Compare commits

..

1 Commits

Author SHA1 Message Date
srikanthccv
7e1947800b chore: temp commit 2026-02-08 11:03:17 +05:30
73 changed files with 7381 additions and 1540 deletions

View File

@@ -2665,7 +2665,6 @@ paths:
parameters:
- in: query
name: metricName
required: true
schema:
type: string
responses:
@@ -2720,7 +2719,6 @@ paths:
parameters:
- in: query
name: metricName
required: true
schema:
type: string
responses:
@@ -2776,7 +2774,6 @@ paths:
parameters:
- in: query
name: metricName
required: true
schema:
type: string
responses:
@@ -2943,7 +2940,6 @@ paths:
parameters:
- in: query
name: metricName
required: true
schema:
type: string
responses:
@@ -3811,9 +3807,6 @@ components:
type: string
alertName:
type: string
required:
- alertName
- alertId
type: object
MetricsexplorertypesMetricAlertsResponse:
properties:
@@ -3822,8 +3815,6 @@ components:
$ref: '#/components/schemas/MetricsexplorertypesMetricAlert'
nullable: true
type: array
required:
- alerts
type: object
MetricsexplorertypesMetricAttribute:
properties:
@@ -3837,10 +3828,6 @@ components:
type: string
nullable: true
type: array
required:
- key
- values
- valueCount
type: object
MetricsexplorertypesMetricAttributesRequest:
properties:
@@ -3852,8 +3839,6 @@ components:
start:
nullable: true
type: integer
required:
- metricName
type: object
MetricsexplorertypesMetricAttributesResponse:
properties:
@@ -3865,9 +3850,6 @@ components:
totalKeys:
format: int64
type: integer
required:
- attributes
- totalKeys
type: object
MetricsexplorertypesMetricDashboard:
properties:
@@ -3879,11 +3861,6 @@ components:
type: string
widgetName:
type: string
required:
- dashboardName
- dashboardId
- widgetId
- widgetName
type: object
MetricsexplorertypesMetricDashboardsResponse:
properties:
@@ -3892,8 +3869,6 @@ components:
$ref: '#/components/schemas/MetricsexplorertypesMetricDashboard'
nullable: true
type: array
required:
- dashboards
type: object
MetricsexplorertypesMetricHighlightsResponse:
properties:
@@ -3909,11 +3884,6 @@ components:
totalTimeSeries:
minimum: 0
type: integer
required:
- dataPoints
- lastReceived
- totalTimeSeries
- activeTimeSeries
type: object
MetricsexplorertypesMetricMetadata:
properties:
@@ -3922,27 +3892,11 @@ components:
isMonotonic:
type: boolean
temporality:
enum:
- delta
- cumulative
- unspecified
type: string
type:
enum:
- gauge
- sum
- histogram
- summary
- exponentialhistogram
type: string
unit:
type: string
required:
- description
- type
- unit
- temporality
- isMonotonic
type: object
MetricsexplorertypesStat:
properties:
@@ -3957,22 +3911,9 @@ components:
minimum: 0
type: integer
type:
enum:
- gauge
- sum
- histogram
- summary
- exponentialhistogram
type: string
unit:
type: string
required:
- metricName
- description
- type
- unit
- timeseries
- samples
type: object
MetricsexplorertypesStatsRequest:
properties:
@@ -3990,10 +3931,6 @@ components:
start:
format: int64
type: integer
required:
- start
- end
- limit
type: object
MetricsexplorertypesStatsResponse:
properties:
@@ -4005,9 +3942,6 @@ components:
total:
minimum: 0
type: integer
required:
- metrics
- total
type: object
MetricsexplorertypesTreemapEntry:
properties:
@@ -4019,10 +3953,6 @@ components:
totalValue:
minimum: 0
type: integer
required:
- metricName
- percentage
- totalValue
type: object
MetricsexplorertypesTreemapRequest:
properties:
@@ -4034,18 +3964,10 @@ components:
limit:
type: integer
mode:
enum:
- timeseries
- samples
type: string
start:
format: int64
type: integer
required:
- start
- end
- limit
- mode
type: object
MetricsexplorertypesTreemapResponse:
properties:
@@ -4059,9 +3981,6 @@ components:
$ref: '#/components/schemas/MetricsexplorertypesTreemapEntry'
nullable: true
type: array
required:
- timeseries
- samples
type: object
MetricsexplorertypesUpdateMetricMetadataRequest:
properties:
@@ -4072,28 +3991,11 @@ components:
metricName:
type: string
temporality:
enum:
- delta
- cumulative
- unspecified
type: string
type:
enum:
- gauge
- sum
- histogram
- summary
- exponentialhistogram
type: string
unit:
type: string
required:
- metricName
- type
- description
- unit
- temporality
- isMonotonic
type: object
PreferencetypesPreference:
properties:

View File

@@ -26,6 +26,7 @@ func PrepareTaskFunc(opts baserules.PrepareTaskOptions) (baserules.Task, error)
if err != nil {
return nil, errors.NewInvalidInputf(errors.CodeInvalidInput, "evaluation is invalid: %v", err)
}
if opts.Rule.RuleType == ruletypes.RuleTypeThreshold {
// create a threshold rule
tr, err := baserules.NewThresholdRule(

View File

@@ -61,8 +61,6 @@ module.exports = {
curly: 'error', // Requires curly braces for all control statements
eqeqeq: ['error', 'always', { null: 'ignore' }], // Enforces === and !== (allows == null for null/undefined check)
'no-console': ['error', { allow: ['warn', 'error'] }], // Warns on console.log, allows console.warn/error
// TODO: Change this to error in May 2026
'max-params': ['warn', 3], // a function can have max 3 params after which it should become an object
// TypeScript rules
'@typescript-eslint/explicit-function-return-type': 'error', // Requires explicit return types on functions
@@ -118,7 +116,7 @@ module.exports = {
},
],
'import/no-extraneous-dependencies': ['error', { devDependencies: true }], // Prevents importing packages not in package.json
'import/no-cycle': 'warn', // Warns about circular dependencies
// 'import/no-cycle': 'warn', // TODO: Enable later to detect circular dependencies
// Import sorting rules
'simple-import-sort/imports': [
@@ -148,19 +146,6 @@ module.exports = {
'sonarjs/no-duplicate-string': 'off', // Disabled - can be noisy (enable periodically to check)
},
overrides: [
{
files: [
'**/*.test.{js,jsx,ts,tsx}',
'**/*.spec.{js,jsx,ts,tsx}',
'**/__tests__/**/*.{js,jsx,ts,tsx}',
],
rules: {
// Tests often have intentional duplication and complexity - disable SonarJS rules
'sonarjs/cognitive-complexity': 'off', // Tests can be complex
'sonarjs/no-identical-functions': 'off', // Similar test patterns are OK
'sonarjs/no-small-switch': 'off', // Small switches are OK in tests
},
},
{
files: ['src/api/generated/**/*.ts'],
rules: {
@@ -168,6 +153,7 @@ module.exports = {
'@typescript-eslint/explicit-module-boundary-types': 'off',
'no-nested-ternary': 'off',
'@typescript-eslint/no-unused-vars': 'warn',
'sonarjs/no-duplicate-string': 'off',
},
},
],

View File

@@ -2,11 +2,6 @@
Embrace the spirit of collaboration and contribute to the success of our open-source project by adhering to these frontend development guidelines with precision and passion.
### Export Style
- **React components** (`src/components/`, `src/container/`, `src/pages/`): Prefer **default exports** for the main component in each file
- **Utilities, hooks, APIs, types, constants** (`src/utils/`, `src/hooks/`, `src/api/`, `src/lib/`, `src/types/`, `src/constants/`): Prefer **named exports** for better tree-shaking and explicit imports
### React and Components
- Strive to create small and modular components, ensuring they are divided into individual pieces for improved maintainability and reusability.

View File

@@ -28,10 +28,8 @@ import type {
GatewaytypesPostableIngestionKeyLimitDTO,
GatewaytypesUpdatableIngestionKeyLimitDTO,
GetIngestionKeys200,
GetIngestionKeysParams,
RenderErrorResponseDTO,
SearchIngestionKeys200,
SearchIngestionKeysParams,
UpdateIngestionKeyLimitPathParameters,
UpdateIngestionKeyPathParameters,
} from '../sigNoz.schemas';
@@ -44,44 +42,35 @@ type Awaited<O> = O extends AwaitedInput<infer T> ? T : never;
* This endpoint returns the ingestion keys for a workspace
* @summary Get ingestion keys for workspace
*/
export const getIngestionKeys = (
params?: GetIngestionKeysParams,
signal?: AbortSignal,
) => {
export const getIngestionKeys = (signal?: AbortSignal) => {
return GeneratedAPIInstance<GetIngestionKeys200>({
url: `/api/v2/gateway/ingestion_keys`,
method: 'GET',
params,
signal,
});
};
export const getGetIngestionKeysQueryKey = (
params?: GetIngestionKeysParams,
) => {
return ['getIngestionKeys', ...(params ? [params] : [])] as const;
export const getGetIngestionKeysQueryKey = () => {
return ['getIngestionKeys'] as const;
};
export const getGetIngestionKeysQueryOptions = <
TData = Awaited<ReturnType<typeof getIngestionKeys>>,
TError = RenderErrorResponseDTO
>(
params?: GetIngestionKeysParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getIngestionKeys>>,
TError,
TData
>;
},
) => {
>(options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getIngestionKeys>>,
TError,
TData
>;
}) => {
const { query: queryOptions } = options ?? {};
const queryKey = queryOptions?.queryKey ?? getGetIngestionKeysQueryKey(params);
const queryKey = queryOptions?.queryKey ?? getGetIngestionKeysQueryKey();
const queryFn: QueryFunction<Awaited<ReturnType<typeof getIngestionKeys>>> = ({
signal,
}) => getIngestionKeys(params, signal);
}) => getIngestionKeys(signal);
return { queryKey, queryFn, ...queryOptions } as UseQueryOptions<
Awaited<ReturnType<typeof getIngestionKeys>>,
@@ -102,17 +91,14 @@ export type GetIngestionKeysQueryError = RenderErrorResponseDTO;
export function useGetIngestionKeys<
TData = Awaited<ReturnType<typeof getIngestionKeys>>,
TError = RenderErrorResponseDTO
>(
params?: GetIngestionKeysParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getIngestionKeys>>,
TError,
TData
>;
},
): UseQueryResult<TData, TError> & { queryKey: QueryKey } {
const queryOptions = getGetIngestionKeysQueryOptions(params, options);
>(options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getIngestionKeys>>,
TError,
TData
>;
}): UseQueryResult<TData, TError> & { queryKey: QueryKey } {
const queryOptions = getGetIngestionKeysQueryOptions(options);
const query = useQuery(queryOptions) as UseQueryResult<TData, TError> & {
queryKey: QueryKey;
@@ -128,11 +114,10 @@ export function useGetIngestionKeys<
*/
export const invalidateGetIngestionKeys = async (
queryClient: QueryClient,
params?: GetIngestionKeysParams,
options?: InvalidateOptions,
): Promise<QueryClient> => {
await queryClient.invalidateQueries(
{ queryKey: getGetIngestionKeysQueryKey(params) },
{ queryKey: getGetIngestionKeysQueryKey() },
options,
);
@@ -677,45 +662,35 @@ export const useUpdateIngestionKeyLimit = <
* This endpoint returns the ingestion keys for a workspace
* @summary Search ingestion keys for workspace
*/
export const searchIngestionKeys = (
params?: SearchIngestionKeysParams,
signal?: AbortSignal,
) => {
export const searchIngestionKeys = (signal?: AbortSignal) => {
return GeneratedAPIInstance<SearchIngestionKeys200>({
url: `/api/v2/gateway/ingestion_keys/search`,
method: 'GET',
params,
signal,
});
};
export const getSearchIngestionKeysQueryKey = (
params?: SearchIngestionKeysParams,
) => {
return ['searchIngestionKeys', ...(params ? [params] : [])] as const;
export const getSearchIngestionKeysQueryKey = () => {
return ['searchIngestionKeys'] as const;
};
export const getSearchIngestionKeysQueryOptions = <
TData = Awaited<ReturnType<typeof searchIngestionKeys>>,
TError = RenderErrorResponseDTO
>(
params?: SearchIngestionKeysParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof searchIngestionKeys>>,
TError,
TData
>;
},
) => {
>(options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof searchIngestionKeys>>,
TError,
TData
>;
}) => {
const { query: queryOptions } = options ?? {};
const queryKey =
queryOptions?.queryKey ?? getSearchIngestionKeysQueryKey(params);
const queryKey = queryOptions?.queryKey ?? getSearchIngestionKeysQueryKey();
const queryFn: QueryFunction<
Awaited<ReturnType<typeof searchIngestionKeys>>
> = ({ signal }) => searchIngestionKeys(params, signal);
> = ({ signal }) => searchIngestionKeys(signal);
return { queryKey, queryFn, ...queryOptions } as UseQueryOptions<
Awaited<ReturnType<typeof searchIngestionKeys>>,
@@ -736,17 +711,14 @@ export type SearchIngestionKeysQueryError = RenderErrorResponseDTO;
export function useSearchIngestionKeys<
TData = Awaited<ReturnType<typeof searchIngestionKeys>>,
TError = RenderErrorResponseDTO
>(
params?: SearchIngestionKeysParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof searchIngestionKeys>>,
TError,
TData
>;
},
): UseQueryResult<TData, TError> & { queryKey: QueryKey } {
const queryOptions = getSearchIngestionKeysQueryOptions(params, options);
>(options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof searchIngestionKeys>>,
TError,
TData
>;
}): UseQueryResult<TData, TError> & { queryKey: QueryKey } {
const queryOptions = getSearchIngestionKeysQueryOptions(options);
const query = useQuery(queryOptions) as UseQueryResult<TData, TError> & {
queryKey: QueryKey;
@@ -762,11 +734,10 @@ export function useSearchIngestionKeys<
*/
export const invalidateSearchIngestionKeys = async (
queryClient: QueryClient,
params?: SearchIngestionKeysParams,
options?: InvalidateOptions,
): Promise<QueryClient> => {
await queryClient.invalidateQueries(
{ queryKey: getSearchIngestionKeysQueryKey(params) },
{ queryKey: getSearchIngestionKeysQueryKey() },
options,
);

View File

@@ -47,7 +47,7 @@ type Awaited<O> = O extends AwaitedInput<infer T> ? T : never;
* @summary Get metric alerts
*/
export const getMetricAlerts = (
params: GetMetricAlertsParams,
params?: GetMetricAlertsParams,
signal?: AbortSignal,
) => {
return GeneratedAPIInstance<GetMetricAlerts200>({
@@ -66,7 +66,7 @@ export const getGetMetricAlertsQueryOptions = <
TData = Awaited<ReturnType<typeof getMetricAlerts>>,
TError = RenderErrorResponseDTO
>(
params: GetMetricAlertsParams,
params?: GetMetricAlertsParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getMetricAlerts>>,
@@ -103,7 +103,7 @@ export function useGetMetricAlerts<
TData = Awaited<ReturnType<typeof getMetricAlerts>>,
TError = RenderErrorResponseDTO
>(
params: GetMetricAlertsParams,
params?: GetMetricAlertsParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getMetricAlerts>>,
@@ -128,7 +128,7 @@ export function useGetMetricAlerts<
*/
export const invalidateGetMetricAlerts = async (
queryClient: QueryClient,
params: GetMetricAlertsParams,
params?: GetMetricAlertsParams,
options?: InvalidateOptions,
): Promise<QueryClient> => {
await queryClient.invalidateQueries(
@@ -144,7 +144,7 @@ export const invalidateGetMetricAlerts = async (
* @summary Get metric dashboards
*/
export const getMetricDashboards = (
params: GetMetricDashboardsParams,
params?: GetMetricDashboardsParams,
signal?: AbortSignal,
) => {
return GeneratedAPIInstance<GetMetricDashboards200>({
@@ -165,7 +165,7 @@ export const getGetMetricDashboardsQueryOptions = <
TData = Awaited<ReturnType<typeof getMetricDashboards>>,
TError = RenderErrorResponseDTO
>(
params: GetMetricDashboardsParams,
params?: GetMetricDashboardsParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getMetricDashboards>>,
@@ -203,7 +203,7 @@ export function useGetMetricDashboards<
TData = Awaited<ReturnType<typeof getMetricDashboards>>,
TError = RenderErrorResponseDTO
>(
params: GetMetricDashboardsParams,
params?: GetMetricDashboardsParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getMetricDashboards>>,
@@ -228,7 +228,7 @@ export function useGetMetricDashboards<
*/
export const invalidateGetMetricDashboards = async (
queryClient: QueryClient,
params: GetMetricDashboardsParams,
params?: GetMetricDashboardsParams,
options?: InvalidateOptions,
): Promise<QueryClient> => {
await queryClient.invalidateQueries(
@@ -244,7 +244,7 @@ export const invalidateGetMetricDashboards = async (
* @summary Get metric highlights
*/
export const getMetricHighlights = (
params: GetMetricHighlightsParams,
params?: GetMetricHighlightsParams,
signal?: AbortSignal,
) => {
return GeneratedAPIInstance<GetMetricHighlights200>({
@@ -265,7 +265,7 @@ export const getGetMetricHighlightsQueryOptions = <
TData = Awaited<ReturnType<typeof getMetricHighlights>>,
TError = RenderErrorResponseDTO
>(
params: GetMetricHighlightsParams,
params?: GetMetricHighlightsParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getMetricHighlights>>,
@@ -303,7 +303,7 @@ export function useGetMetricHighlights<
TData = Awaited<ReturnType<typeof getMetricHighlights>>,
TError = RenderErrorResponseDTO
>(
params: GetMetricHighlightsParams,
params?: GetMetricHighlightsParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getMetricHighlights>>,
@@ -328,7 +328,7 @@ export function useGetMetricHighlights<
*/
export const invalidateGetMetricHighlights = async (
queryClient: QueryClient,
params: GetMetricHighlightsParams,
params?: GetMetricHighlightsParams,
options?: InvalidateOptions,
): Promise<QueryClient> => {
await queryClient.invalidateQueries(
@@ -526,7 +526,7 @@ export const useGetMetricAttributes = <
* @summary Get metric metadata
*/
export const getMetricMetadata = (
params: GetMetricMetadataParams,
params?: GetMetricMetadataParams,
signal?: AbortSignal,
) => {
return GeneratedAPIInstance<GetMetricMetadata200>({
@@ -547,7 +547,7 @@ export const getGetMetricMetadataQueryOptions = <
TData = Awaited<ReturnType<typeof getMetricMetadata>>,
TError = RenderErrorResponseDTO
>(
params: GetMetricMetadataParams,
params?: GetMetricMetadataParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getMetricMetadata>>,
@@ -585,7 +585,7 @@ export function useGetMetricMetadata<
TData = Awaited<ReturnType<typeof getMetricMetadata>>,
TError = RenderErrorResponseDTO
>(
params: GetMetricMetadataParams,
params?: GetMetricMetadataParams,
options?: {
query?: UseQueryOptions<
Awaited<ReturnType<typeof getMetricMetadata>>,
@@ -610,7 +610,7 @@ export function useGetMetricMetadata<
*/
export const invalidateGetMetricMetadata = async (
queryClient: QueryClient,
params: GetMetricMetadataParams,
params?: GetMetricMetadataParams,
options?: InvalidateOptions,
): Promise<QueryClient> => {
await queryClient.invalidateQueries(

View File

@@ -650,11 +650,11 @@ export interface MetricsexplorertypesMetricAlertDTO {
/**
* @type string
*/
alertId: string;
alertId?: string;
/**
* @type string
*/
alertName: string;
alertName?: string;
}
export interface MetricsexplorertypesMetricAlertsResponseDTO {
@@ -662,24 +662,24 @@ export interface MetricsexplorertypesMetricAlertsResponseDTO {
* @type array
* @nullable true
*/
alerts: MetricsexplorertypesMetricAlertDTO[] | null;
alerts?: MetricsexplorertypesMetricAlertDTO[] | null;
}
export interface MetricsexplorertypesMetricAttributeDTO {
/**
* @type string
*/
key: string;
key?: string;
/**
* @type integer
* @minimum 0
*/
valueCount: number;
valueCount?: number;
/**
* @type array
* @nullable true
*/
values: string[] | null;
values?: string[] | null;
}
export interface MetricsexplorertypesMetricAttributesRequestDTO {
@@ -691,7 +691,7 @@ export interface MetricsexplorertypesMetricAttributesRequestDTO {
/**
* @type string
*/
metricName: string;
metricName?: string;
/**
* @type integer
* @nullable true
@@ -704,31 +704,31 @@ export interface MetricsexplorertypesMetricAttributesResponseDTO {
* @type array
* @nullable true
*/
attributes: MetricsexplorertypesMetricAttributeDTO[] | null;
attributes?: MetricsexplorertypesMetricAttributeDTO[] | null;
/**
* @type integer
* @format int64
*/
totalKeys: number;
totalKeys?: number;
}
export interface MetricsexplorertypesMetricDashboardDTO {
/**
* @type string
*/
dashboardId: string;
dashboardId?: string;
/**
* @type string
*/
dashboardName: string;
dashboardName?: string;
/**
* @type string
*/
widgetId: string;
widgetId?: string;
/**
* @type string
*/
widgetName: string;
widgetName?: string;
}
export interface MetricsexplorertypesMetricDashboardsResponseDTO {
@@ -736,7 +736,7 @@ export interface MetricsexplorertypesMetricDashboardsResponseDTO {
* @type array
* @nullable true
*/
dashboards: MetricsexplorertypesMetricDashboardDTO[] | null;
dashboards?: MetricsexplorertypesMetricDashboardDTO[] | null;
}
export interface MetricsexplorertypesMetricHighlightsResponseDTO {
@@ -744,96 +744,74 @@ export interface MetricsexplorertypesMetricHighlightsResponseDTO {
* @type integer
* @minimum 0
*/
activeTimeSeries: number;
activeTimeSeries?: number;
/**
* @type integer
* @minimum 0
*/
dataPoints: number;
dataPoints?: number;
/**
* @type integer
* @minimum 0
*/
lastReceived: number;
lastReceived?: number;
/**
* @type integer
* @minimum 0
*/
totalTimeSeries: number;
totalTimeSeries?: number;
}
export enum MetricsexplorertypesMetricMetadataDTOTemporality {
delta = 'delta',
cumulative = 'cumulative',
unspecified = 'unspecified',
}
export enum MetricsexplorertypesMetricMetadataDTOType {
gauge = 'gauge',
sum = 'sum',
histogram = 'histogram',
summary = 'summary',
exponentialhistogram = 'exponentialhistogram',
}
export interface MetricsexplorertypesMetricMetadataDTO {
/**
* @type string
*/
description: string;
description?: string;
/**
* @type boolean
*/
isMonotonic: boolean;
/**
* @enum delta,cumulative,unspecified
* @type string
*/
temporality: MetricsexplorertypesMetricMetadataDTOTemporality;
/**
* @enum gauge,sum,histogram,summary,exponentialhistogram
* @type string
*/
type: MetricsexplorertypesMetricMetadataDTOType;
isMonotonic?: boolean;
/**
* @type string
*/
unit: string;
temporality?: string;
/**
* @type string
*/
type?: string;
/**
* @type string
*/
unit?: string;
}
export enum MetricsexplorertypesStatDTOType {
gauge = 'gauge',
sum = 'sum',
histogram = 'histogram',
summary = 'summary',
exponentialhistogram = 'exponentialhistogram',
}
export interface MetricsexplorertypesStatDTO {
/**
* @type string
*/
description: string;
description?: string;
/**
* @type string
*/
metricName: string;
metricName?: string;
/**
* @type integer
* @minimum 0
*/
samples: number;
samples?: number;
/**
* @type integer
* @minimum 0
*/
timeseries: number;
/**
* @enum gauge,sum,histogram,summary,exponentialhistogram
* @type string
*/
type: MetricsexplorertypesStatDTOType;
timeseries?: number;
/**
* @type string
*/
unit: string;
type?: string;
/**
* @type string
*/
unit?: string;
}
export interface MetricsexplorertypesStatsRequestDTO {
@@ -841,12 +819,12 @@ export interface MetricsexplorertypesStatsRequestDTO {
* @type integer
* @format int64
*/
end: number;
end?: number;
filter?: Querybuildertypesv5FilterDTO;
/**
* @type integer
*/
limit: number;
limit?: number;
/**
* @type integer
*/
@@ -856,7 +834,7 @@ export interface MetricsexplorertypesStatsRequestDTO {
* @type integer
* @format int64
*/
start: number;
start?: number;
}
export interface MetricsexplorertypesStatsResponseDTO {
@@ -864,56 +842,51 @@ export interface MetricsexplorertypesStatsResponseDTO {
* @type array
* @nullable true
*/
metrics: MetricsexplorertypesStatDTO[] | null;
metrics?: MetricsexplorertypesStatDTO[] | null;
/**
* @type integer
* @minimum 0
*/
total: number;
total?: number;
}
export interface MetricsexplorertypesTreemapEntryDTO {
/**
* @type string
*/
metricName: string;
metricName?: string;
/**
* @type number
* @format double
*/
percentage: number;
percentage?: number;
/**
* @type integer
* @minimum 0
*/
totalValue: number;
totalValue?: number;
}
export enum MetricsexplorertypesTreemapRequestDTOMode {
timeseries = 'timeseries',
samples = 'samples',
}
export interface MetricsexplorertypesTreemapRequestDTO {
/**
* @type integer
* @format int64
*/
end: number;
end?: number;
filter?: Querybuildertypesv5FilterDTO;
/**
* @type integer
*/
limit: number;
limit?: number;
/**
* @enum timeseries,samples
* @type string
*/
mode: MetricsexplorertypesTreemapRequestDTOMode;
mode?: string;
/**
* @type integer
* @format int64
*/
start: number;
start?: number;
}
export interface MetricsexplorertypesTreemapResponseDTO {
@@ -921,53 +894,39 @@ export interface MetricsexplorertypesTreemapResponseDTO {
* @type array
* @nullable true
*/
samples: MetricsexplorertypesTreemapEntryDTO[] | null;
samples?: MetricsexplorertypesTreemapEntryDTO[] | null;
/**
* @type array
* @nullable true
*/
timeseries: MetricsexplorertypesTreemapEntryDTO[] | null;
timeseries?: MetricsexplorertypesTreemapEntryDTO[] | null;
}
export enum MetricsexplorertypesUpdateMetricMetadataRequestDTOTemporality {
delta = 'delta',
cumulative = 'cumulative',
unspecified = 'unspecified',
}
export enum MetricsexplorertypesUpdateMetricMetadataRequestDTOType {
gauge = 'gauge',
sum = 'sum',
histogram = 'histogram',
summary = 'summary',
exponentialhistogram = 'exponentialhistogram',
}
export interface MetricsexplorertypesUpdateMetricMetadataRequestDTO {
/**
* @type string
*/
description: string;
description?: string;
/**
* @type boolean
*/
isMonotonic: boolean;
isMonotonic?: boolean;
/**
* @type string
*/
metricName: string;
/**
* @enum delta,cumulative,unspecified
* @type string
*/
temporality: MetricsexplorertypesUpdateMetricMetadataRequestDTOTemporality;
/**
* @enum gauge,sum,histogram,summary,exponentialhistogram
* @type string
*/
type: MetricsexplorertypesUpdateMetricMetadataRequestDTOType;
metricName?: string;
/**
* @type string
*/
unit: string;
temporality?: string;
/**
* @type string
*/
type?: string;
/**
* @type string
*/
unit?: string;
}
export interface PreferencetypesPreferenceDTO {
@@ -1892,19 +1851,6 @@ export type GetFeatures200 = {
status?: string;
};
export type GetIngestionKeysParams = {
/**
* @type integer
* @description undefined
*/
page?: number;
/**
* @type integer
* @description undefined
*/
per_page?: number;
};
export type GetIngestionKeys200 = {
data?: GatewaytypesGettableIngestionKeysDTO;
/**
@@ -1944,24 +1890,6 @@ export type DeleteIngestionKeyLimitPathParameters = {
export type UpdateIngestionKeyLimitPathParameters = {
limitId: string;
};
export type SearchIngestionKeysParams = {
/**
* @type string
* @description undefined
*/
name?: string;
/**
* @type integer
* @description undefined
*/
page?: number;
/**
* @type integer
* @description undefined
*/
per_page?: number;
};
export type SearchIngestionKeys200 = {
data?: GatewaytypesGettableIngestionKeysDTO;
/**
@@ -1975,7 +1903,7 @@ export type GetMetricAlertsParams = {
* @type string
* @description undefined
*/
metricName: string;
metricName?: string;
};
export type GetMetricAlerts200 = {
@@ -1991,7 +1919,7 @@ export type GetMetricDashboardsParams = {
* @type string
* @description undefined
*/
metricName: string;
metricName?: string;
};
export type GetMetricDashboards200 = {
@@ -2007,7 +1935,7 @@ export type GetMetricHighlightsParams = {
* @type string
* @description undefined
*/
metricName: string;
metricName?: string;
};
export type GetMetricHighlights200 = {
@@ -2034,7 +1962,7 @@ export type GetMetricMetadataParams = {
* @type string
* @description undefined
*/
metricName: string;
metricName?: string;
};
export type GetMetricMetadata200 = {

View File

@@ -93,7 +93,7 @@ function EndPointDetails({
return currentFilters; // No change needed, prevents loop
}
// Rebuild filters: Keep non-http_url filters and add/update http_url filter based on prop
// Rebuild filters: Keep non-http.url filters and add/update http.url filter based on prop
const otherFilters = currentFilters?.items?.filter(
(item) => item.key?.key !== httpUrlKey.key,
);
@@ -125,7 +125,7 @@ function EndPointDetails({
(newFilters: IBuilderQuery['filters']): void => {
// 1. Update local filters state immediately
setFilters(newFilters);
// Filter out http_url filter before saving to params
// Filter out http.url filter before saving to params
const filteredNewFilters = {
op: 'AND',
items:

View File

@@ -56,7 +56,7 @@ function TopErrors({
{
items: endPointName
? [
// Remove any existing http_url filters from initialFilters to avoid duplicates
// Remove any existing http.url filters from initialFilters to avoid duplicates
...(initialFilters?.items?.filter(
(item) => item.key?.key !== SPAN_ATTRIBUTES.URL_PATH,
) || []),

View File

@@ -126,8 +126,12 @@ describe('DomainMetrics - V5 Query Payload Tests', () => {
'count()',
);
// Verify exact domain filter expression structure
expect(queryA.filter.expression).toContain("http_host = '0.0.0.0'");
expect(queryA.filter.expression).toContain('http_url EXISTS');
expect(queryA.filter.expression).toContain(
"(net.peer.name = '0.0.0.0' OR server.address = '0.0.0.0')",
);
expect(queryA.filter.expression).toContain(
'url.full EXISTS OR http.url EXISTS',
);
// Verify Query B - p99 latency
const queryB = queryData.find((q: any) => q.queryName === 'B');
@@ -138,13 +142,17 @@ describe('DomainMetrics - V5 Query Payload Tests', () => {
'p99(duration_nano)',
);
// Verify exact domain filter expression structure
expect(queryB.filter.expression).toContain("http_host = '0.0.0.0'");
expect(queryB.filter.expression).toContain(
"(net.peer.name = '0.0.0.0' OR server.address = '0.0.0.0')",
);
// Verify Query C - error count (disabled)
const queryC = queryData.find((q: any) => q.queryName === 'C');
expect(queryC).toBeDefined();
expect(queryC.disabled).toBe(true);
expect(queryC.filter.expression).toContain("http_host = '0.0.0.0'");
expect(queryC.filter.expression).toContain(
"(net.peer.name = '0.0.0.0' OR server.address = '0.0.0.0')",
);
expect(queryC.aggregations?.[0]).toBeDefined();
expect((queryC.aggregations?.[0] as TraceAggregation)?.expression).toBe(
'count()',
@@ -161,7 +169,9 @@ describe('DomainMetrics - V5 Query Payload Tests', () => {
'max(timestamp)',
);
// Verify exact domain filter expression structure
expect(queryD.filter.expression).toContain("http_host = '0.0.0.0'");
expect(queryD.filter.expression).toContain(
"(net.peer.name = '0.0.0.0' OR server.address = '0.0.0.0')",
);
// Verify Formula F1 - error rate calculation
const formulas = payload.query.builder.queryFormulas;

View File

@@ -153,7 +153,7 @@ describe('EndPointMetrics - V5 Query Payload Tests', () => {
// Verify exact domain filter expression structure
if (queryA.filter) {
expect(queryA.filter.expression).toContain(
`http_host = 'api.example.com'`,
"(net.peer.name = 'api.example.com' OR server.address = 'api.example.com')",
);
expect(queryA.filter.expression).toContain("kind_string = 'Client'");
}
@@ -171,7 +171,7 @@ describe('EndPointMetrics - V5 Query Payload Tests', () => {
// Verify exact domain filter expression structure
if (queryB.filter) {
expect(queryB.filter.expression).toContain(
`http_host = 'api.example.com'`,
"(net.peer.name = 'api.example.com' OR server.address = 'api.example.com')",
);
expect(queryB.filter.expression).toContain("kind_string = 'Client'");
}
@@ -185,7 +185,7 @@ describe('EndPointMetrics - V5 Query Payload Tests', () => {
expect(queryC.aggregateOperator).toBe('count');
if (queryC.filter) {
expect(queryC.filter.expression).toContain(
`http_host = 'api.example.com'`,
"(net.peer.name = 'api.example.com' OR server.address = 'api.example.com')",
);
expect(queryC.filter.expression).toContain("kind_string = 'Client'");
expect(queryC.filter.expression).toContain('has_error = true');
@@ -204,7 +204,7 @@ describe('EndPointMetrics - V5 Query Payload Tests', () => {
// Verify exact domain filter expression structure
if (queryD.filter) {
expect(queryD.filter.expression).toContain(
`http_host = 'api.example.com'`,
"(net.peer.name = 'api.example.com' OR server.address = 'api.example.com')",
);
expect(queryD.filter.expression).toContain("kind_string = 'Client'");
}
@@ -221,7 +221,7 @@ describe('EndPointMetrics - V5 Query Payload Tests', () => {
}
if (queryE.filter) {
expect(queryE.filter.expression).toContain(
`http_host = 'api.example.com'`,
"(net.peer.name = 'api.example.com' OR server.address = 'api.example.com')",
);
expect(queryE.filter.expression).toContain("kind_string = 'Client'");
}
@@ -291,7 +291,7 @@ describe('EndPointMetrics - V5 Query Payload Tests', () => {
expect(query.filter.expression).toContain('staging');
// Also verify domain filter is still present
expect(query.filter.expression).toContain(
"http_host = 'api.internal.com'",
"(net.peer.name = 'api.internal.com' OR server.address = 'api.internal.com')",
);
// Verify client kind filter is present
expect(query.filter.expression).toContain("kind_string = 'Client'");

View File

@@ -166,7 +166,6 @@ function StatusCodeBarCharts({
xValue,
TWO_AND_HALF_MINUTES_IN_MILLISECONDS,
);
handleGraphClick({
xValue,
yValue,

View File

@@ -12,8 +12,8 @@ export const VIEW_TYPES = {
// Span attribute keys - these are the source of truth for all attribute keys
export const SPAN_ATTRIBUTES = {
URL_PATH: 'http_url',
URL_PATH: 'http.url',
RESPONSE_STATUS_CODE: 'response_status_code',
SERVER_NAME: 'http_host',
SERVER_NAME: 'net.peer.name',
SERVER_PORT: 'net.peer.port',
} as const;

View File

@@ -350,7 +350,7 @@ describe('API Monitoring Utils', () => {
data: {
// eslint-disable-next-line sonarjs/no-duplicate-string
[URL_PATH_KEY]: '/api/users',
URL_PATH_KEY: 'http://example.com/api/users',
'url.full': 'http://example.com/api/users',
A: 150, // count or other metric
},
},
@@ -358,7 +358,7 @@ describe('API Monitoring Utils', () => {
data: {
// eslint-disable-next-line sonarjs/no-duplicate-string
[URL_PATH_KEY]: '/api/orders',
URL_PATH_KEY: 'http://example.com/api/orders',
'url.full': 'http://example.com/api/orders',
A: 75,
},
},
@@ -741,6 +741,15 @@ describe('API Monitoring Utils', () => {
if (domainFilter) {
expect(domainFilter.value).toBe(domainName);
}
// Should have endpoint filter if provided
const endpointFilter = queryData.filters?.items?.find(
(item) => item.key && item.key.key === SPAN_ATTRIBUTES.URL_PATH,
);
expect(endpointFilter).toBeDefined();
if (endpointFilter) {
expect(endpointFilter.value).toBe(endPointName);
}
});
it('should include custom filters in the widget configuration', () => {

View File

@@ -6,10 +6,10 @@
* These tests validate the migration from V4 to V5 format for getAllEndpointsWidgetData:
* - Filter format change: filters.items[] → filter.expression
* - Aggregation format: aggregateAttribute → aggregations[] array
* - Domain filter: http_host = '${domainName}'
* - Domain filter: (net.peer.name OR server.address)
* - Kind filter: kind_string = 'Client'
* - Four queries: A (count), B (p99 latency), C (max timestamp), D (error count - disabled)
* - GroupBy: http_url with type 'attribute'
* - GroupBy: Both http.url AND url.full with type 'attribute'
*/
import { getAllEndpointsWidgetData } from 'container/ApiMonitoring/utils';
import {
@@ -92,28 +92,28 @@ describe('AllEndpointsWidget - V5 Migration Validation', () => {
const [queryA, queryB, queryC, queryD] = widget.query.builder.queryData;
const baseExpression = `http_host = '${mockDomainName}' AND kind_string = 'Client'`;
const baseExpression = `(net.peer.name = '${mockDomainName}' OR server.address = '${mockDomainName}') AND kind_string = 'Client'`;
// Queries A, B, C have identical base filter
expect(queryA.filter?.expression).toBe(
`${baseExpression} AND http_url EXISTS`,
`${baseExpression} AND (http.url EXISTS OR url.full EXISTS)`,
);
expect(queryB.filter?.expression).toBe(
`${baseExpression} AND http_url EXISTS`,
`${baseExpression} AND (http.url EXISTS OR url.full EXISTS)`,
);
expect(queryC.filter?.expression).toBe(
`${baseExpression} AND http_url EXISTS`,
`${baseExpression} AND (http.url EXISTS OR url.full EXISTS)`,
);
// Query D has additional has_error filter
expect(queryD.filter?.expression).toBe(
`${baseExpression} AND has_error = true AND http_url EXISTS`,
`${baseExpression} AND has_error = true AND (http.url EXISTS OR url.full EXISTS)`,
);
});
});
describe('2. GroupBy Structure', () => {
it('default groupBy includes both http_url and http_host with type attribute', () => {
it('default groupBy includes both http.url and url.full with type attribute', () => {
const widget = getAllEndpointsWidgetData(
emptyGroupBy,
mockDomainName,
@@ -124,13 +124,23 @@ describe('AllEndpointsWidget - V5 Migration Validation', () => {
// All queries should have the same default groupBy
queryData.forEach((query) => {
expect(query.groupBy).toHaveLength(1);
expect(query.groupBy).toHaveLength(2);
// http.url
expect(query.groupBy).toContainEqual({
dataType: DataTypes.String,
isColumn: false,
isJSON: false,
key: 'http_url',
key: 'http.url',
type: 'attribute',
});
// url.full
expect(query.groupBy).toContainEqual({
dataType: DataTypes.String,
isColumn: false,
isJSON: false,
key: 'url.full',
type: 'attribute',
});
});
@@ -160,18 +170,19 @@ describe('AllEndpointsWidget - V5 Migration Validation', () => {
// All queries should have defaults + custom groupBy
queryData.forEach((query) => {
expect(query.groupBy).toHaveLength(3); // 1 default + 2 custom
expect(query.groupBy).toHaveLength(4); // 2 defaults + 2 custom
// First two should be defaults (http_url)
expect(query.groupBy[0].key).toBe('http_url');
// First two should be defaults (http.url, url.full)
expect(query.groupBy[0].key).toBe('http.url');
expect(query.groupBy[1].key).toBe('url.full');
// Last two should be custom (matching subset of properties)
expect(query.groupBy[1]).toMatchObject({
expect(query.groupBy[2]).toMatchObject({
dataType: DataTypes.String,
key: 'service.name',
type: 'resource',
});
expect(query.groupBy[2]).toMatchObject({
expect(query.groupBy[3]).toMatchObject({
dataType: DataTypes.String,
key: 'deployment.environment',
type: 'resource',

View File

@@ -6,11 +6,11 @@
* These tests validate the migration from V4 to V5 format for the third payload
* in getEndPointDetailsQueryPayload (endpoint dropdown data):
* - Filter format change: filters.items[] → filter.expression
* - Domain handling: http_host = '${domainName}'
* - Domain handling: (net.peer.name OR server.address)
* - Kind filter: kind_string = 'Client'
* - Existence check: http_url EXISTS
* - Existence check: (http.url EXISTS OR url.full EXISTS)
* - Aggregation: count() expression
* - GroupBy: http_url with type 'attribute'
* - GroupBy: Both http.url AND url.full with type 'attribute'
*/
import { getEndPointDetailsQueryPayload } from 'container/ApiMonitoring/utils';
import { IBuilderQuery } from 'types/api/queryBuilder/queryBuilderData';
@@ -43,16 +43,18 @@ describe('EndpointDropdown - V5 Migration Validation', () => {
expect(typeof queryA.filter?.expression).toBe('string');
expect(queryA).not.toHaveProperty('filters');
// Base filter 1: Domain http_host = '${domainName}'
// Base filter 1: Domain (net.peer.name OR server.address)
expect(queryA.filter?.expression).toContain(
`http_host = '${mockDomainName}'`,
`(net.peer.name = '${mockDomainName}' OR server.address = '${mockDomainName}')`,
);
// Base filter 2: Kind
expect(queryA.filter?.expression).toContain("kind_string = 'Client'");
// Base filter 3: Existence check
expect(queryA.filter?.expression).toContain('http_url EXISTS');
expect(queryA.filter?.expression).toContain(
'(http.url EXISTS OR url.full EXISTS)',
);
// V5 Aggregation format: aggregations array (not aggregateAttribute)
expect(queryA.aggregations).toBeDefined();
@@ -62,10 +64,15 @@ describe('EndpointDropdown - V5 Migration Validation', () => {
});
expect(queryA).not.toHaveProperty('aggregateAttribute');
// GroupBy: http_url
expect(queryA.groupBy).toHaveLength(1);
// GroupBy: Both http.url and url.full
expect(queryA.groupBy).toHaveLength(2);
expect(queryA.groupBy).toContainEqual({
key: 'http_url',
key: 'http.url',
dataType: 'string',
type: 'attribute',
});
expect(queryA.groupBy).toContainEqual({
key: 'url.full',
dataType: 'string',
type: 'attribute',
});
@@ -113,19 +120,19 @@ describe('EndpointDropdown - V5 Migration Validation', () => {
// Exact filter expression with custom filters merged
expect(expression).toBe(
"http_host = 'api.example.com' AND kind_string = 'Client' AND http_url EXISTS service.name = 'user-service' AND deployment.environment = 'production'",
"(net.peer.name = 'api.example.com' OR server.address = 'api.example.com') AND kind_string = 'Client' AND (http.url EXISTS OR url.full EXISTS) service.name = 'user-service' AND deployment.environment = 'production'",
);
});
});
describe('3. HTTP URL Filter Special Handling', () => {
it('converts http_url filter to http_url expression', () => {
it('converts http.url filter to (http.url OR url.full) expression', () => {
const filtersWithHttpUrl: IBuilderQuery['filters'] = {
items: [
{
id: 'http-url-filter',
key: {
key: 'http_url',
key: 'http.url',
dataType: 'string' as any,
type: 'tag',
},
@@ -157,8 +164,9 @@ describe('EndpointDropdown - V5 Migration Validation', () => {
const expression =
dropdownQuery.query.builder.queryData[0].filter?.expression;
// CRITICAL: Exact filter expression with http.url converted to OR logic
expect(expression).toBe(
"http_host = 'api.example.com' AND kind_string = 'Client' AND http_url EXISTS http_url = '/api/users' AND service.name = 'user-service'",
"(net.peer.name = 'api.example.com' OR server.address = 'api.example.com') AND kind_string = 'Client' AND (http.url EXISTS OR url.full EXISTS) service.name = 'user-service' AND (http.url = '/api/users' OR url.full = '/api/users')",
);
});
});

View File

@@ -33,7 +33,7 @@ describe('MetricOverTime - V5 Migration Validation', () => {
expect(queryData).not.toHaveProperty('filters.items');
});
it('uses new domain filter format: (http_url)', () => {
it('uses new domain filter format: (net.peer.name OR server.address)', () => {
const widget = getRateOverTimeWidgetData(
mockDomainName,
mockEndpointName,
@@ -44,7 +44,7 @@ describe('MetricOverTime - V5 Migration Validation', () => {
// Verify EXACT new filter format with OR operator
expect(queryData?.filter?.expression).toContain(
`http_host = '${mockDomainName}'`,
`(net.peer.name = '${mockDomainName}' OR server.address = '${mockDomainName}')`,
);
// Endpoint name is used in legend, not filter
@@ -90,7 +90,7 @@ describe('MetricOverTime - V5 Migration Validation', () => {
// Verify domain filter is present
expect(queryData?.filter?.expression).toContain(
`http_host = '${mockDomainName}'`,
`(net.peer.name = '${mockDomainName}' OR server.address = '${mockDomainName}')`,
);
// Verify custom filters are merged into the expression
@@ -120,7 +120,7 @@ describe('MetricOverTime - V5 Migration Validation', () => {
expect(queryData).not.toHaveProperty('filters.items');
});
it('uses new domain filter format: (http_host)', () => {
it('uses new domain filter format: (net.peer.name OR server.address)', () => {
const widget = getLatencyOverTimeWidgetData(
mockDomainName,
mockEndpointName,
@@ -132,7 +132,7 @@ describe('MetricOverTime - V5 Migration Validation', () => {
// Verify EXACT new filter format with OR operator
expect(queryData.filter).toBeDefined();
expect(queryData?.filter?.expression).toContain(
`http_host = '${mockDomainName}'`,
`(net.peer.name = '${mockDomainName}' OR server.address = '${mockDomainName}')`,
);
// Endpoint name is used in legend, not filter
@@ -166,7 +166,7 @@ describe('MetricOverTime - V5 Migration Validation', () => {
// Verify domain filter is present
expect(queryData?.filter?.expression).toContain(
`http_host = '${mockDomainName}' service.name = 'user-service'`,
`(net.peer.name = '${mockDomainName}' OR server.address = '${mockDomainName}') service.name = 'user-service'`,
);
});
});

View File

@@ -10,7 +10,7 @@
*
* V5 Changes:
* - Filter format change: filters.items[] → filter.expression
* - Domain filter: (http_host)
* - Domain filter: (net.peer.name OR server.address)
* - Kind filter: kind_string = 'Client'
* - stepInterval: 60 → null
* - Grouped by response_status_code
@@ -47,9 +47,9 @@ describe('StatusCodeBarCharts - V5 Migration Validation', () => {
expect(typeof queryA.filter?.expression).toBe('string');
expect(queryA).not.toHaveProperty('filters.items');
// Base filter 1: Domain (http_host)
// Base filter 1: Domain (net.peer.name OR server.address)
expect(queryA.filter?.expression).toContain(
`http_host = '${mockDomainName}'`,
`(net.peer.name = '${mockDomainName}' OR server.address = '${mockDomainName}')`,
);
// Base filter 2: Kind
@@ -96,9 +96,9 @@ describe('StatusCodeBarCharts - V5 Migration Validation', () => {
expect(typeof queryA.filter?.expression).toBe('string');
expect(queryA).not.toHaveProperty('filters.items');
// Base filter 1: Domain (http_host)
// Base filter 1: Domain (net.peer.name OR server.address)
expect(queryA.filter?.expression).toContain(
`http_host = '${mockDomainName}'`,
`(net.peer.name = '${mockDomainName}' OR server.address = '${mockDomainName}')`,
);
// Base filter 2: Kind
@@ -177,7 +177,7 @@ describe('StatusCodeBarCharts - V5 Migration Validation', () => {
expect(callsExpression).toBe(latencyExpression);
// Verify base filters
expect(callsExpression).toContain('http_host');
expect(callsExpression).toContain('net.peer.name');
expect(callsExpression).toContain("kind_string = 'Client'");
// Verify custom filters are merged
@@ -189,13 +189,13 @@ describe('StatusCodeBarCharts - V5 Migration Validation', () => {
});
describe('4. HTTP URL Filter Handling', () => {
it('converts http_url filter to http_url expression in both charts', () => {
it('converts http.url filter to (http.url OR url.full) expression in both charts', () => {
const filtersWithHttpUrl: IBuilderQuery['filters'] = {
items: [
{
id: 'http-url-filter',
key: {
key: 'http_url',
key: 'http.url',
dataType: 'string' as any,
type: 'tag',
},
@@ -221,10 +221,16 @@ describe('StatusCodeBarCharts - V5 Migration Validation', () => {
const latencyExpression =
latencyChartQuery.query.builder.queryData[0].filter?.expression;
expect(callsExpression).toContain("http_url = '/api/metrics'");
expect(latencyExpression).toContain("http_url = '/api/metrics'");
// CRITICAL: http.url converted to OR logic
expect(callsExpression).toContain(
"(http.url = '/api/metrics' OR url.full = '/api/metrics')",
);
expect(latencyExpression).toContain(
"(http.url = '/api/metrics' OR url.full = '/api/metrics')",
);
expect(callsExpression).toContain('http_host');
// Base filters still present
expect(callsExpression).toContain('net.peer.name');
expect(callsExpression).toContain("kind_string = 'Client'");
});
});

View File

@@ -6,8 +6,8 @@
* These tests validate the migration from V4 to V5 format for the second payload
* in getEndPointDetailsQueryPayload (status code table data):
* - Filter format change: filters.items[] → filter.expression
* - URL handling: Special logic for http_url
* - Domain filter: http_host = '${domainName}'
* - URL handling: Special logic for (http.url OR url.full)
* - Domain filter: (net.peer.name OR server.address)
* - Kind filter: kind_string = 'Client'
* - Kind filter: response_status_code EXISTS
* - Three queries: A (count), B (p99 latency), C (rate)
@@ -45,9 +45,9 @@ describe('StatusCodeTable - V5 Migration Validation', () => {
expect(typeof queryA.filter?.expression).toBe('string');
expect(queryA).not.toHaveProperty('filters.items');
// Base filter 1: Domain (http_host)
// Base filter 1: Domain (net.peer.name OR server.address)
expect(queryA.filter?.expression).toContain(
`http_host = '${mockDomainName}'`,
`(net.peer.name = '${mockDomainName}' OR server.address = '${mockDomainName}')`,
);
// Base filter 2: Kind
@@ -149,7 +149,7 @@ describe('StatusCodeTable - V5 Migration Validation', () => {
statusCodeQuery.query.builder.queryData[0].filter?.expression;
// Base filters present
expect(expression).toContain('http_host');
expect(expression).toContain('net.peer.name');
expect(expression).toContain("kind_string = 'Client'");
expect(expression).toContain('response_status_code EXISTS');
@@ -167,13 +167,13 @@ describe('StatusCodeTable - V5 Migration Validation', () => {
});
describe('4. HTTP URL Filter Handling', () => {
it('converts http_url filter to http_url expression', () => {
it('converts http.url filter to (http.url OR url.full) expression', () => {
const filtersWithHttpUrl: IBuilderQuery['filters'] = {
items: [
{
id: 'http-url-filter',
key: {
key: 'http_url',
key: 'http.url',
dataType: 'string' as any,
type: 'tag',
},
@@ -205,14 +205,17 @@ describe('StatusCodeTable - V5 Migration Validation', () => {
const expression =
statusCodeQuery.query.builder.queryData[0].filter?.expression;
expect(expression).toContain("http_url = '/api/users'");
// CRITICAL: http.url converted to OR logic
expect(expression).toContain(
"(http.url = '/api/users' OR url.full = '/api/users')",
);
// Other filters still present
expect(expression).toContain('service.name');
expect(expression).toContain('user-service');
// Base filters present
expect(expression).toContain('http_host');
expect(expression).toContain('net.peer.name');
expect(expression).toContain("kind_string = 'Client'");
expect(expression).toContain('response_status_code EXISTS');

View File

@@ -83,7 +83,7 @@ describe('TopErrors', () => {
{
columns: [
{
name: 'http_url',
name: 'http.url',
fieldDataType: 'string',
fieldContext: 'attribute',
},
@@ -123,7 +123,7 @@ describe('TopErrors', () => {
table: {
rows: [
{
http_url: '/api/test',
'http.url': '/api/test',
A: 100,
},
],
@@ -205,7 +205,7 @@ describe('TopErrors', () => {
expect(navigateMock).toHaveBeenCalledWith({
filters: expect.arrayContaining([
expect.objectContaining({
key: expect.objectContaining({ key: 'http_url' }),
key: expect.objectContaining({ key: 'http.url' }),
op: '=',
value: '/api/test',
}),
@@ -215,7 +215,7 @@ describe('TopErrors', () => {
value: 'true',
}),
expect.objectContaining({
key: expect.objectContaining({ key: 'http_host' }),
key: expect.objectContaining({ key: 'net.peer.name' }),
op: '=',
value: 'test-domain',
}),
@@ -334,7 +334,7 @@ describe('TopErrors', () => {
// Verify all required filters are present
expect(filterExpression).toContain(
`kind_string = 'Client' AND http_url EXISTS AND http_host = 'test-domain' AND has_error = true`,
`kind_string = 'Client' AND (http.url EXISTS OR url.full EXISTS) AND (net.peer.name = 'test-domain' OR server.address = 'test-domain') AND has_error = true`,
);
});
});

View File

@@ -57,12 +57,12 @@ export const getDisplayValue = (value: unknown): string =>
isEmptyFilterValue(value) ? '-' : String(value);
export const getDomainNameFilterExpression = (domainName: string): string =>
`http_host = '${domainName}'`;
`(net.peer.name = '${domainName}' OR server.address = '${domainName}')`;
export const clientKindExpression = `kind_string = 'Client'`;
/**
* Converts filters to expression
* Converts filters to expression, handling http.url specially by creating (http.url OR url.full) condition
* @param filters Filters to convert
* @param baseExpression Base expression to combine with filters
* @returns Filter expression string
@@ -75,6 +75,34 @@ export const convertFiltersWithUrlHandling = (
return baseExpression;
}
// Check if filters contain http.url (SPAN_ATTRIBUTES.URL_PATH)
const httpUrlFilter = filters.items?.find(
(item) => item.key?.key === SPAN_ATTRIBUTES.URL_PATH,
);
// If http.url filter exists, create modified filters with (http.url OR url.full)
if (httpUrlFilter && httpUrlFilter.value) {
// Remove ALL http.url filters from items (guards against duplicates)
const otherFilters = filters.items?.filter(
(item) => item.key?.key !== SPAN_ATTRIBUTES.URL_PATH,
);
// Convert to expression first with other filters
const {
filter: intermediateFilter,
} = convertFiltersToExpressionWithExistingQuery(
{ ...filters, items: otherFilters || [] },
baseExpression,
);
// Add the OR condition for http.url and url.full
const urlValue = httpUrlFilter.value;
const urlCondition = `(http.url = '${urlValue}' OR url.full = '${urlValue}')`;
return intermediateFilter.expression.trim()
? `${intermediateFilter.expression} AND ${urlCondition}`
: urlCondition;
}
const { filter } = convertFiltersToExpressionWithExistingQuery(
filters,
baseExpression,
@@ -343,7 +371,7 @@ export const formatDataForTable = (
});
};
const urlExpression = `http_url EXISTS`;
const urlExpression = `(url.full EXISTS OR http.url EXISTS)`;
export const getDomainMetricsQueryPayload = (
domainName: string,
@@ -563,6 +591,13 @@ const defaultGroupBy = [
key: SPAN_ATTRIBUTES.URL_PATH,
type: 'attribute',
},
{
dataType: DataTypes.String,
isColumn: false,
isJSON: false,
key: 'url.full',
type: 'attribute',
},
// {
// key: SPAN_ATTRIBUTES.SERVER_PORT,
// dataType: DataTypes.Float64,
@@ -832,8 +867,8 @@ function buildFilterExpression(
): string {
const baseFilterParts = [
`kind_string = 'Client'`,
`http_url EXISTS`,
`http_host = '${domainName}'`,
`(http.url EXISTS OR url.full EXISTS)`,
`(net.peer.name = '${domainName}' OR server.address = '${domainName}')`,
`has_error = true`,
];
if (showStatusCodeErrors) {
@@ -875,7 +910,12 @@ export const getTopErrorsQueryPayload = (
filter: { expression: filterExpression },
groupBy: [
{
name: 'http_url',
name: 'http.url',
fieldDataType: 'string',
fieldContext: 'attribute',
},
{
name: 'url.full',
fieldDataType: 'string',
fieldContext: 'attribute',
},
@@ -1222,7 +1262,9 @@ export const formatTopErrorsDataForTable = (
return {
key: v4(),
endpointName: getDisplayValue(rowObj[SPAN_ATTRIBUTES.URL_PATH]),
endpointName: getDisplayValue(
rowObj[SPAN_ATTRIBUTES.URL_PATH] || rowObj['url.full'],
),
statusCode: getDisplayValue(rowObj[SPAN_ATTRIBUTES.RESPONSE_STATUS_CODE]),
statusMessage: getDisplayValue(rowObj.status_message),
count: getDisplayValue(rowObj.__result_0),
@@ -1239,10 +1281,10 @@ export const getTopErrorsCoRelationQueryFilters = (
{
id: 'ea16470b',
key: {
key: 'http_url',
key: 'http.url',
dataType: DataTypes.String,
type: 'tag',
id: 'http_url--string--tag--false',
id: 'http.url--string--tag--false',
},
op: '=',
value: endPointName,
@@ -1260,7 +1302,7 @@ export const getTopErrorsCoRelationQueryFilters = (
{
id: 'e8a043b7',
key: {
key: 'http_host',
key: 'net.peer.name',
dataType: DataTypes.String,
type: '',
},
@@ -1739,7 +1781,7 @@ export const getEndPointDetailsQueryPayload = (
filters || { items: [], op: 'AND' },
`${getDomainNameFilterExpression(
domainName,
)} AND ${clientKindExpression} AND http_url EXISTS`,
)} AND ${clientKindExpression} AND (http.url EXISTS OR url.full EXISTS)`,
),
},
expression: 'A',
@@ -1755,6 +1797,11 @@ export const getEndPointDetailsQueryPayload = (
dataType: DataTypes.String,
type: 'attribute',
},
{
key: 'url.full',
dataType: DataTypes.String,
type: 'attribute',
},
],
reduceTo: ReduceOperators.AVG,
},
@@ -2373,6 +2420,7 @@ export const statusCodeWidgetInfo = [
interface EndPointDropDownResponseRow {
data: {
[SPAN_ATTRIBUTES.URL_PATH]: string;
'url.full': string;
A: number;
};
}
@@ -2391,8 +2439,8 @@ export const getFormattedEndPointDropDownData = (
}
return data.map((row) => ({
key: v4(),
label: row.data[SPAN_ATTRIBUTES.URL_PATH] || '-',
value: row.data[SPAN_ATTRIBUTES.URL_PATH] || '-',
label: row.data[SPAN_ATTRIBUTES.URL_PATH] || row.data['url.full'] || '-',
value: row.data[SPAN_ATTRIBUTES.URL_PATH] || row.data['url.full'] || '-',
}));
};
@@ -2750,6 +2798,20 @@ export const getStatusCodeBarChartWidgetData = (
op: '=',
value: domainName,
},
...(endPointName
? [
{
id: '8b1be6f0',
key: {
dataType: DataTypes.String,
key: SPAN_ATTRIBUTES.URL_PATH,
type: 'tag',
},
op: '=',
value: endPointName,
},
]
: []),
...(filters?.items || []),
],
op: 'AND',
@@ -2871,7 +2933,7 @@ export const getAllEndpointsWidgetData = (
filters,
`${getDomainNameFilterExpression(
domainName,
)} AND ${clientKindExpression} AND http_url EXISTS`,
)} AND ${clientKindExpression} AND (http.url EXISTS OR url.full EXISTS)`,
),
},
functions: [],
@@ -2903,7 +2965,7 @@ export const getAllEndpointsWidgetData = (
filters,
`${getDomainNameFilterExpression(
domainName,
)} AND ${clientKindExpression} AND http_url EXISTS`,
)} AND ${clientKindExpression} AND (http.url EXISTS OR url.full EXISTS)`,
),
},
functions: [],
@@ -2935,7 +2997,7 @@ export const getAllEndpointsWidgetData = (
filters,
`${getDomainNameFilterExpression(
domainName,
)} AND ${clientKindExpression} AND http_url EXISTS`,
)} AND ${clientKindExpression} AND (http.url EXISTS OR url.full EXISTS)`,
),
},
functions: [],
@@ -2967,7 +3029,7 @@ export const getAllEndpointsWidgetData = (
filters,
`${getDomainNameFilterExpression(
domainName,
)} AND ${clientKindExpression} AND has_error = true AND http_url EXISTS`,
)} AND ${clientKindExpression} AND has_error = true AND (http.url EXISTS OR url.full EXISTS)`,
),
},
functions: [],
@@ -3005,10 +3067,10 @@ export const getAllEndpointsWidgetData = (
// First try to use the url from the column value
let urlValue = url;
// If url is empty/null and we have the record, fallback to http_url
// If url is empty/null and we have the record, fallback to url.full
if (isEmptyFilterValue(url) && record) {
const { http_url } = record;
urlValue = http_url;
const { 'url.full': urlFull } = record;
urlValue = urlFull;
}
if (!urlValue || urlValue === 'n/a') {
@@ -3071,7 +3133,7 @@ export const getAllEndpointsWidgetData = (
widget.customColTitles = {
[SPAN_ATTRIBUTES.URL_PATH]: 'Endpoint',
http_port: 'Port',
'net.peer.port': 'Port',
};
widget.title = (
@@ -3096,10 +3158,12 @@ export const getAllEndpointsWidgetData = (
</div>
);
widget.hiddenColumns = ['url.full'];
return widget;
};
const keysToRemove = ['http_url', 'A', 'B', 'C', 'F1'];
const keysToRemove = ['http.url', 'url.full', 'A', 'B', 'C', 'F1'];
export const getGroupByFiltersFromGroupByValues = (
rowData: any,
@@ -3157,7 +3221,7 @@ export const getRateOverTimeWidgetData = (
filter: {
expression: convertFiltersWithUrlHandling(
filters || { items: [], op: 'AND' },
`http_host = '${domainName}'`,
`(net.peer.name = '${domainName}' OR server.address = '${domainName}')`,
),
},
functions: [],
@@ -3208,7 +3272,7 @@ export const getLatencyOverTimeWidgetData = (
filter: {
expression: convertFiltersWithUrlHandling(
filters || { items: [], op: 'AND' },
`http_host = '${domainName}'`,
`(net.peer.name = '${domainName}' OR server.address = '${domainName}')`,
),
},
functions: [],

View File

@@ -25,6 +25,12 @@
}
}
&.focused {
.variable-value {
outline: 1px solid var(--bg-robin-400);
}
}
.variable-value {
display: flex;
min-width: 120px;
@@ -42,11 +48,6 @@
font-style: normal;
font-weight: 400;
line-height: 16px; /* 133.333% */
&:hover,
&:focus-within {
outline: 1px solid var(--bg-robin-400);
}
}
.variable-select {
@@ -98,6 +99,12 @@
.lightMode {
.variable-item {
&.focused {
.variable-value {
border: 1px solid var(--bg-robin-400);
}
}
.variable-name {
border: 1px solid var(--bg-vanilla-300);
background: var(--bg-vanilla-100);
@@ -108,11 +115,6 @@
border: 1px solid var(--bg-vanilla-300);
background: var(--bg-vanilla-100);
color: var(--bg-ink-400);
&:hover,
&:focus-within {
outline: 1px solid var(--bg-robin-400);
}
}
}
}
@@ -122,9 +124,3 @@
padding: 4px 12px;
font-size: 12px;
}
.dashboard-variables-selection-container {
display: flex;
flex-wrap: wrap;
gap: 12px;
}

View File

@@ -1,4 +1,4 @@
import { memo, useCallback, useEffect, useMemo } from 'react';
import { memo, useEffect } from 'react';
import { useSelector } from 'react-redux';
import { Row } from 'antd';
import { ALL_SELECTED_VALUE } from 'components/NewSelect/utils';
@@ -7,6 +7,7 @@ import {
useDashboardVariablesSelector,
} from 'hooks/dashboard/useDashboardVariables';
import useVariablesFromUrl from 'hooks/dashboard/useVariablesFromUrl';
import { isEmpty } from 'lodash-es';
import { useDashboard } from 'providers/Dashboard/Dashboard';
import { initializeDefaultVariables } from 'providers/Dashboard/initializeDefaultVariables';
import { AppState } from 'store/reducers';
@@ -21,6 +22,7 @@ import './DashboardVariableSelection.styles.scss';
function DashboardVariableSelection(): JSX.Element | null {
const {
selectedDashboard,
setSelectedDashboard,
updateLocalStorageDashboardVariables,
variablesToGetUpdated,
@@ -50,36 +52,34 @@ function DashboardVariableSelection(): JSX.Element | null {
);
}, [getUrlVariables, updateUrlVariable, dashboardVariables]);
// Memoize the order key to avoid unnecessary triggers
const dependencyOrderKey = useMemo(
() => dependencyData?.order?.join(',') ?? '',
[dependencyData?.order],
);
// Trigger refetch when dependency order changes or global time changes
useEffect(() => {
if (dependencyData?.order && dependencyData.order.length > 0) {
setVariablesToGetUpdated(dependencyData?.order || []);
}
// this handles the case where the dependency order changes i.e. variable list updated via creation or deletion etc. and we need to refetch the variables
// also trigger when the global time changes
useEffect(
() => {
if (!isEmpty(dependencyData?.order)) {
setVariablesToGetUpdated(dependencyData?.order || []);
}
},
// eslint-disable-next-line react-hooks/exhaustive-deps
}, [dependencyOrderKey, minTime, maxTime]);
[JSON.stringify(dependencyData?.order), minTime, maxTime],
);
// Performance optimization: For dynamic variables with allSelected=true, we don't store
// individual values in localStorage since we can always derive them from available options.
// This makes localStorage much lighter and more efficient.
const onValueUpdate = useCallback(
(
name: string,
id: string,
value: IDashboardVariable['selectedValue'],
allSelected: boolean,
haveCustomValuesSelected?: boolean,
// eslint-disable-next-line sonarjs/cognitive-complexity
): void => {
const onValueUpdate = (
name: string,
id: string,
value: IDashboardVariable['selectedValue'],
allSelected: boolean,
haveCustomValuesSelected?: boolean,
// eslint-disable-next-line sonarjs/cognitive-complexity
): void => {
if (id) {
// For dynamic variables, only store in localStorage when NOT allSelected
// This makes localStorage much lighter by avoiding storing all individual values
const variable = dashboardVariables[id] || dashboardVariables[name];
const isDynamic = variable.type === 'DYNAMIC';
const variable = dashboardVariables?.[id] || dashboardVariables?.[name];
const isDynamic = variable?.type === 'DYNAMIC';
updateLocalStorageDashboardVariables(name, value, allSelected, isDynamic);
if (allSelected) {
@@ -88,39 +88,41 @@ function DashboardVariableSelection(): JSX.Element | null {
updateUrlVariable(name || id, value);
}
setSelectedDashboard((prev) => {
if (prev) {
const oldVariables = { ...prev?.data.variables };
// this is added to handle case where we have two different
// schemas for variable response
if (oldVariables?.[id]) {
oldVariables[id] = {
...oldVariables[id],
selectedValue: value,
allSelected,
haveCustomValuesSelected,
};
}
if (oldVariables?.[name]) {
oldVariables[name] = {
...oldVariables[name],
selectedValue: value,
allSelected,
haveCustomValuesSelected,
};
}
return {
...prev,
data: {
...prev?.data,
variables: {
...oldVariables,
if (selectedDashboard) {
setSelectedDashboard((prev) => {
if (prev) {
const oldVariables = prev?.data.variables;
// this is added to handle case where we have two different
// schemas for variable response
if (oldVariables?.[id]) {
oldVariables[id] = {
...oldVariables[id],
selectedValue: value,
allSelected,
haveCustomValuesSelected,
};
}
if (oldVariables?.[name]) {
oldVariables[name] = {
...oldVariables[name],
selectedValue: value,
allSelected,
haveCustomValuesSelected,
};
}
return {
...prev,
data: {
...prev?.data,
variables: {
...oldVariables,
},
},
},
};
}
return prev;
});
};
}
return prev;
});
}
if (dependencyData) {
const updatedVariables: string[] = [];
@@ -136,19 +138,11 @@ function DashboardVariableSelection(): JSX.Element | null {
} else {
setVariablesToGetUpdated((prev) => prev.filter((v) => v !== name));
}
},
[
dashboardVariables,
updateLocalStorageDashboardVariables,
dependencyData,
updateUrlVariable,
setSelectedDashboard,
setVariablesToGetUpdated,
],
);
}
};
return (
<Row className="dashboard-variables-selection-container">
<Row style={{ display: 'flex', gap: '12px' }}>
{sortedVariablesArray.map((variable) => {
const key = `${variable.name}${variable.id}${variable.order}`;

View File

@@ -1,66 +0,0 @@
import { memo, useCallback, useRef, useState } from 'react';
import { Input, InputRef } from 'antd';
import { IDashboardVariable } from 'types/api/dashboard/getAll';
interface TextboxVariableInputProps {
variableData: IDashboardVariable;
handleChange: (inputValue: string) => void;
}
function TextboxVariableInput({
variableData,
handleChange,
}: TextboxVariableInputProps): JSX.Element {
const textboxInputRef = useRef<InputRef>(null);
const [textboxInputValue, setTextboxInputValue] = useState<string>(
(variableData.selectedValue?.toString() ||
variableData.defaultValue?.toString()) ??
'',
);
const handleInputOnChange = useCallback(
(event: React.ChangeEvent<HTMLInputElement>) => {
setTextboxInputValue(event.target.value);
},
[setTextboxInputValue],
);
const handleInputOnBlur = useCallback(
(event: React.FocusEvent<HTMLInputElement>): void => {
const value = event.target.value.trim();
// If empty, reset to default value
if (!value && variableData.defaultValue) {
setTextboxInputValue(variableData.defaultValue.toString());
handleChange(variableData.defaultValue.toString());
} else {
handleChange(value);
}
},
[handleChange, variableData.defaultValue],
);
const handleInputOnKeyDown = useCallback(
(event: React.KeyboardEvent<HTMLInputElement>): void => {
if (event.key === 'Enter') {
textboxInputRef.current?.blur();
}
},
[],
);
return (
<Input
ref={textboxInputRef}
placeholder="Enter value"
data-testid={`variable-textbox-${variableData.id}`}
bordered={false}
value={textboxInputValue}
title={textboxInputValue}
onChange={handleInputOnChange}
onBlur={handleInputOnBlur}
onKeyDown={handleInputOnKeyDown}
/>
);
}
export default memo(TextboxVariableInput);

View File

@@ -4,18 +4,18 @@
/* eslint-disable @typescript-eslint/no-explicit-any */
/* eslint-disable react/jsx-props-no-spreading */
/* eslint-disable no-nested-ternary */
import { memo, useCallback, useEffect, useMemo, useState } from 'react';
import { memo, useCallback, useEffect, useMemo, useRef, useState } from 'react';
import { useQuery } from 'react-query';
import { useSelector } from 'react-redux';
import { orange } from '@ant-design/colors';
import { InfoCircleOutlined, WarningOutlined } from '@ant-design/icons';
import { Popover, Tooltip, Typography } from 'antd';
import { Input, InputRef, Popover, Tooltip, Typography } from 'antd';
import dashboardVariablesQuery from 'api/dashboard/variables/dashboardVariablesQuery';
import { CustomMultiSelect, CustomSelect } from 'components/NewSelect';
import { REACT_QUERY_KEY } from 'constants/reactQueryKeys';
import { commaValuesParser } from 'lib/dashbaordVariables/customCommaValuesParser';
import sortValues from 'lib/dashbaordVariables/sortVariableValues';
import { isArray, isEmpty, isString } from 'lodash-es';
import { debounce, isArray, isEmpty, isString } from 'lodash-es';
import { IDependencyData } from 'providers/Dashboard/store/dashboardVariables/dashboardVariablesStoreTypes';
import { AppState } from 'store/reducers';
import { IDashboardVariable } from 'types/api/dashboard/getAll';
@@ -25,7 +25,6 @@ import { popupContainer } from 'utils/selectPopupContainer';
import { ALL_SELECT_VALUE, variablePropsToPayloadVariables } from '../utils';
import { SelectItemStyle } from './styles';
import TextboxVariableInput from './TextboxVariableInput';
import { areArraysEqual, checkAPIInvocation } from './util';
import './DashboardVariableSelection.styles.scss';
@@ -73,6 +72,15 @@ function VariableItem({
string | string[] | undefined
>(undefined);
// Local state for textbox input to ensure smooth editing experience
const [textboxInputValue, setTextboxInputValue] = useState<string>(
(variableData.selectedValue?.toString() ||
variableData.defaultValue?.toString()) ??
'',
);
const [isTextboxFocused, setIsTextboxFocused] = useState<boolean>(false);
const textboxInputRef = useRef<InputRef>(null);
const { maxTime, minTime } = useSelector<AppState, GlobalReducer>(
(state) => state.globalTime,
);
@@ -296,6 +304,7 @@ function VariableItem({
};
// do not debounce the above function as we do not need debounce in select variables
const debouncedHandleChange = debounce(handleChange, 500);
const { selectedValue } = variableData;
const selectedValueStringified = useMemo(
@@ -372,7 +381,7 @@ function VariableItem({
}, [variableData.type, variableData.customValue]);
return (
<div className="variable-item">
<div className={`variable-item${isTextboxFocused ? ' focused' : ''}`}>
<Typography.Text className="variable-name" ellipsis>
${variableData.name}
{variableData.description && (
@@ -384,9 +393,42 @@ function VariableItem({
<div className="variable-value">
{variableData.type === 'TEXTBOX' ? (
<TextboxVariableInput
variableData={variableData}
handleChange={handleChange}
<Input
ref={textboxInputRef}
placeholder="Enter value"
data-testid={`variable-textbox-${variableData.id}`}
bordered={false}
value={textboxInputValue}
title={textboxInputValue}
onChange={(e): void => {
setTextboxInputValue(e.target.value);
}}
onFocus={(): void => {
setIsTextboxFocused(true);
}}
onBlur={(e): void => {
setIsTextboxFocused(false);
const value = e.target.value.trim();
// If empty, reset to default value
if (!value && variableData.defaultValue) {
setTextboxInputValue(variableData.defaultValue.toString());
debouncedHandleChange(variableData.defaultValue.toString());
} else {
debouncedHandleChange(value);
}
}}
onKeyDown={(e): void => {
if (e.key === 'Enter') {
const value = textboxInputValue.trim();
if (!value && variableData.defaultValue) {
setTextboxInputValue(variableData.defaultValue.toString());
debouncedHandleChange(variableData.defaultValue.toString());
} else {
debouncedHandleChange(value);
}
textboxInputRef.current?.blur();
}
}}
/>
) : (
optionsData &&

View File

@@ -1,21 +0,0 @@
.chart-manager-container {
width: 100%;
max-height: calc(40% - 40px);
display: flex;
flex-direction: column;
gap: 16px;
.chart-manager-header {
display: flex;
justify-content: space-between;
align-items: center;
gap: 16px;
.chart-manager-actions-container {
display: flex;
justify-content: flex-end;
align-items: center;
gap: 8px;
}
}
}

View File

@@ -1,147 +0,0 @@
import { useCallback, useEffect, useMemo, useState } from 'react';
import { Button, Input } from 'antd';
import { ResizeTable } from 'components/ResizeTable';
import { getGraphManagerTableColumns } from 'container/GridCardLayout/GridCard/FullView/TableRender/GraphManagerColumns';
import { ExtendedChartDataset } from 'container/GridCardLayout/GridCard/FullView/types';
import { getDefaultTableDataSet } from 'container/GridCardLayout/GridCard/FullView/utils';
import { useNotifications } from 'hooks/useNotifications';
import { UPlotConfigBuilder } from 'lib/uPlotV2/config/UPlotConfigBuilder';
import { usePlotContext } from 'lib/uPlotV2/context/PlotContext';
import useLegendsSync from 'lib/uPlotV2/hooks/useLegendsSync';
import { useDashboard } from 'providers/Dashboard/Dashboard';
import './ChartManager.styles.scss';
interface ChartManagerProps {
config: UPlotConfigBuilder;
alignedData: uPlot.AlignedData;
yAxisUnit?: string;
onCancel?: () => void;
}
/**
* ChartManager provides a tabular view to manage the visibility of
* individual series on a uPlot chart.
*
* It syncs with the legend state coming from the plot context and
* allows users to:
* - filter series by label
* - toggle individual series on/off
* - persist the visibility configuration to local storage.
*
* @param config - `UPlotConfigBuilder` instance used to derive chart options.
* @param alignedData - uPlot aligned data used to build the initial table dataset.
* @param yAxisUnit - Optional unit label for Y-axis values shown in the table.
* @param onCancel - Optional callback invoked when the user cancels the dialog.
*/
export default function ChartManager({
config,
alignedData,
yAxisUnit,
onCancel,
}: ChartManagerProps): JSX.Element {
const { notifications } = useNotifications();
const { legendItemsMap } = useLegendsSync({
config,
subscribeToFocusChange: false,
});
const {
onToggleSeriesOnOff,
onToggleSeriesVisibility,
syncSeriesVisibilityToLocalStorage,
} = usePlotContext();
const { isDashboardLocked } = useDashboard();
const [tableDataSet, setTableDataSet] = useState<ExtendedChartDataset[]>(() =>
getDefaultTableDataSet(config.getConfig() as uPlot.Options, alignedData),
);
const graphVisibilityState = useMemo(
() =>
Object.entries(legendItemsMap).reduce<boolean[]>((acc, [key, item]) => {
acc[Number(key)] = item.show;
return acc;
}, []),
[legendItemsMap],
);
useEffect(() => {
setTableDataSet(
getDefaultTableDataSet(config.getConfig() as uPlot.Options, alignedData),
);
}, [alignedData, config]);
const filterHandler = useCallback(
(event: React.ChangeEvent<HTMLInputElement>): void => {
const value = event.target.value.toString().toLowerCase();
const updatedDataSet = tableDataSet.map((item) => {
if (item.label?.toLocaleLowerCase().includes(value)) {
return { ...item, show: true };
}
return { ...item, show: false };
});
setTableDataSet(updatedDataSet);
},
[tableDataSet],
);
const dataSource = useMemo(
() =>
tableDataSet.filter(
(item, index) => index !== 0 && item.show, // skipping the first item as it is the x-axis
),
[tableDataSet],
);
const columns = useMemo(
() =>
getGraphManagerTableColumns({
tableDataSet,
checkBoxOnChangeHandler: (_e, index) => {
onToggleSeriesOnOff(index);
},
graphVisibilityState,
labelClickedHandler: onToggleSeriesVisibility,
yAxisUnit,
isGraphDisabled: isDashboardLocked,
}),
// eslint-disable-next-line react-hooks/exhaustive-deps
[tableDataSet, graphVisibilityState, yAxisUnit, isDashboardLocked],
);
const handleSave = useCallback((): void => {
syncSeriesVisibilityToLocalStorage();
notifications.success({
message: 'The updated graphs & legends are saved',
});
if (onCancel) {
onCancel();
}
}, [syncSeriesVisibilityToLocalStorage, notifications, onCancel]);
return (
<div className="chart-manager-container">
<div className="chart-manager-header">
<Input onChange={filterHandler} placeholder="Filter Series" />
<div className="chart-manager-actions-container">
<Button type="default" onClick={onCancel}>
Cancel
</Button>
<Button type="primary" onClick={handleSave}>
Save
</Button>
</div>
</div>
<div className="chart-manager-table-container">
<ResizeTable
columns={columns}
dataSource={dataSource}
virtual
rowKey="index"
scroll={{ y: 200 }}
pagination={false}
/>
</div>
</div>
);
}

View File

@@ -1,120 +0,0 @@
import { useCallback } from 'react';
import { UseQueryResult } from 'react-query';
import {
getTimeRangeFromStepInterval,
isApmMetric,
} from 'container/PanelWrapper/utils';
import { getUplotClickData } from 'container/QueryTable/Drilldown/drilldownUtils';
import useGraphContextMenu from 'container/QueryTable/Drilldown/useGraphContextMenu';
import {
PopoverPosition,
useCoordinates,
} from 'periscope/components/ContextMenu';
import { SuccessResponse } from 'types/api';
import { Widgets } from 'types/api/dashboard/getAll';
import { MetricRangePayloadProps } from 'types/api/metrics/getQueryRange';
import { DataSource } from 'types/common/queryBuilder';
interface UseTimeSeriesContextMenuParams {
widget: Widgets;
queryResponse: UseQueryResult<
SuccessResponse<MetricRangePayloadProps, unknown>,
Error
>;
}
export const usePanelContextMenu = ({
widget,
queryResponse,
}: UseTimeSeriesContextMenuParams): {
coordinates: { x: number; y: number } | null;
popoverPosition: PopoverPosition | null;
onClose: () => void;
menuItemsConfig: {
header?: string | React.ReactNode;
items?: React.ReactNode;
};
clickHandlerWithContextMenu: (...args: any[]) => void;
} => {
const {
coordinates,
popoverPosition,
clickedData,
onClose,
subMenu,
onClick,
setSubMenu,
} = useCoordinates();
const { menuItemsConfig } = useGraphContextMenu({
widgetId: widget.id || '',
query: widget.query,
graphData: clickedData,
onClose,
coordinates,
subMenu,
setSubMenu,
contextLinks: widget.contextLinks,
panelType: widget.panelTypes,
queryRange: queryResponse,
});
const clickHandlerWithContextMenu = useCallback(
(...args: any[]) => {
const [
xValue,
_yvalue,
_mouseX,
_mouseY,
metric,
queryData,
absoluteMouseX,
absoluteMouseY,
axesData,
focusedSeries,
] = args;
const data = getUplotClickData({
metric,
queryData,
absoluteMouseX,
absoluteMouseY,
focusedSeries,
});
let timeRange;
if (axesData && queryData?.queryName) {
const compositeQuery = (queryResponse?.data?.params as any)?.compositeQuery;
if (compositeQuery?.queries) {
const specificQuery = compositeQuery.queries.find(
(query: any) => query.spec?.name === queryData.queryName,
);
const stepInterval = specificQuery?.spec?.stepInterval || 60;
timeRange = getTimeRangeFromStepInterval(
stepInterval,
metric?.clickedTimestamp || xValue,
specificQuery?.spec?.signal === DataSource.METRICS &&
isApmMetric(specificQuery?.spec?.aggregations[0]?.metricName),
);
}
}
if (data && data?.record?.queryName) {
onClick(data.coord, { ...data.record, label: data.label, timeRange });
}
},
[onClick, queryResponse],
);
return {
coordinates,
popoverPosition,
onClose,
menuItemsConfig,
clickHandlerWithContextMenu,
};
};

View File

@@ -1,4 +0,0 @@
.panel-container {
height: 100%;
width: 100%;
}

View File

@@ -1,176 +0,0 @@
import { useEffect, useMemo, useRef, useState } from 'react';
import TimeSeries from 'container/DashboardContainer/visualization/charts/TimeSeries/TimeSeries';
import ChartManager from 'container/DashboardContainer/visualization/components/ChartManager/ChartManager';
import { usePanelContextMenu } from 'container/DashboardContainer/visualization/hooks/usePanelContextMenu';
import { PanelWrapperProps } from 'container/PanelWrapper/panelWrapper.types';
import { useIsDarkMode } from 'hooks/useDarkMode';
import { useResizeObserver } from 'hooks/useDimensions';
import { LegendPosition } from 'lib/uPlotV2/components/types';
import { LineInterpolation } from 'lib/uPlotV2/config/types';
import { ContextMenu } from 'periscope/components/ContextMenu';
import { useDashboard } from 'providers/Dashboard/Dashboard';
import { useTimezone } from 'providers/Timezone';
import { MetricRangePayloadProps } from 'types/api/metrics/getQueryRange';
import uPlot from 'uplot';
import { getTimeRange } from 'utils/getTimeRange';
import { prepareChartData, prepareUPlotConfig } from '../TimeSeriesPanel/utils';
import '../Panel.styles.scss';
function TimeSeriesPanel(props: PanelWrapperProps): JSX.Element {
const {
panelMode,
queryResponse,
widget,
onDragSelect,
isFullViewMode,
onToggleModelHandler,
} = props;
const { toScrollWidgetId, setToScrollWidgetId } = useDashboard();
const graphRef = useRef<HTMLDivElement>(null);
const [minTimeScale, setMinTimeScale] = useState<number>();
const [maxTimeScale, setMaxTimeScale] = useState<number>();
const containerDimensions = useResizeObserver(graphRef);
const isDarkMode = useIsDarkMode();
const { timezone } = useTimezone();
useEffect(() => {
if (toScrollWidgetId === widget.id) {
graphRef.current?.scrollIntoView({
behavior: 'smooth',
block: 'center',
});
graphRef.current?.focus();
setToScrollWidgetId('');
}
}, [toScrollWidgetId, setToScrollWidgetId, widget.id]);
useEffect((): void => {
const { startTime, endTime } = getTimeRange(queryResponse);
setMinTimeScale(startTime);
setMaxTimeScale(endTime);
}, [queryResponse]);
const {
coordinates,
popoverPosition,
onClose,
menuItemsConfig,
clickHandlerWithContextMenu,
} = usePanelContextMenu({
widget,
queryResponse,
});
const chartData = useMemo(() => {
if (!queryResponse?.data?.payload) {
return [];
}
return prepareChartData(queryResponse?.data?.payload);
}, [queryResponse?.data?.payload]);
const config = useMemo(() => {
const tzDate = (timestamp: number): Date =>
uPlot.tzDate(new Date(timestamp * 1e3), timezone.value);
return prepareUPlotConfig({
widgetId: widget.id || '',
apiResponse: queryResponse?.data?.payload as MetricRangePayloadProps,
tzDate,
minTimeScale: minTimeScale,
maxTimeScale: maxTimeScale,
isLogScale: widget?.isLogScale ?? false,
thresholds: {
scaleKey: 'y',
thresholds: (widget.thresholds || []).map((threshold) => ({
thresholdValue: threshold.thresholdValue ?? 0,
thresholdColor: threshold.thresholdColor,
thresholdUnit: threshold.thresholdUnit,
thresholdLabel: threshold.thresholdLabel,
})),
yAxisUnit: widget.yAxisUnit,
},
yAxisUnit: widget.yAxisUnit || '',
softMin: widget.softMin === undefined ? null : widget.softMin,
softMax: widget.softMax === undefined ? null : widget.softMax,
spanGaps: false,
colorMapping: widget.customLegendColors ?? {},
lineInterpolation: LineInterpolation.Spline,
isDarkMode,
onClick: clickHandlerWithContextMenu,
onDragSelect,
currentQuery: widget.query,
panelMode,
});
}, [
widget.id,
maxTimeScale,
minTimeScale,
timezone.value,
widget.customLegendColors,
widget.isLogScale,
widget.softMax,
widget.softMin,
isDarkMode,
queryResponse?.data?.payload,
widget.query,
widget.thresholds,
widget.yAxisUnit,
panelMode,
clickHandlerWithContextMenu,
onDragSelect,
]);
const layoutChildren = useMemo(() => {
if (!isFullViewMode) {
return null;
}
return (
<ChartManager
config={config}
alignedData={chartData}
yAxisUnit={widget.yAxisUnit}
onCancel={onToggleModelHandler}
/>
);
}, [
isFullViewMode,
config,
chartData,
widget.yAxisUnit,
onToggleModelHandler,
]);
return (
<div className="panel-container" ref={graphRef}>
{containerDimensions.width > 0 && containerDimensions.height > 0 && (
<TimeSeries
config={config}
legendConfig={{
position: widget?.legendPosition ?? LegendPosition.BOTTOM,
}}
yAxisUnit={widget.yAxisUnit}
decimalPrecision={widget.decimalPrecision}
timezone={timezone.value}
data={chartData as uPlot.AlignedData}
width={containerDimensions.width}
height={containerDimensions.height}
layoutChildren={layoutChildren}
>
<ContextMenu
coordinates={coordinates}
popoverPosition={popoverPosition}
title={menuItemsConfig.header as string}
items={menuItemsConfig.items}
onClose={onClose}
/>
</TimeSeries>
)}
</div>
);
}
export default TimeSeriesPanel;

View File

@@ -1,170 +0,0 @@
import { PANEL_TYPES } from 'constants/queryBuilder';
import {
fillMissingXAxisTimestamps,
getXAxisTimestamps,
} from 'container/DashboardContainer/visualization/panels/utils';
import { getLegend } from 'lib/dashboard/getQueryResults';
import getLabelName from 'lib/getLabelName';
import onClickPlugin, {
OnClickPluginOpts,
} from 'lib/uPlotLib/plugins/onClickPlugin';
import {
DistributionType,
DrawStyle,
LineInterpolation,
LineStyle,
SelectionPreferencesSource,
VisibilityMode,
} from 'lib/uPlotV2/config/types';
import { UPlotConfigBuilder } from 'lib/uPlotV2/config/UPlotConfigBuilder';
import { ThresholdsDrawHookOptions } from 'lib/uPlotV2/hooks/types';
import { MetricRangePayloadProps } from 'types/api/metrics/getQueryRange';
import { Query } from 'types/api/queryBuilder/queryBuilderData';
import { PanelMode } from '../types';
export const prepareChartData = (
apiResponse: MetricRangePayloadProps,
): uPlot.AlignedData => {
const seriesList = apiResponse?.data?.result || [];
const timestampArr = getXAxisTimestamps(seriesList);
const yAxisValuesArr = fillMissingXAxisTimestamps(timestampArr, seriesList);
return [timestampArr, ...yAxisValuesArr];
};
export const prepareUPlotConfig = ({
widgetId,
apiResponse,
tzDate,
minTimeScale,
maxTimeScale,
isLogScale,
thresholds,
softMin,
softMax,
spanGaps,
colorMapping,
lineInterpolation,
isDarkMode,
currentQuery,
onDragSelect,
onClick,
yAxisUnit,
panelMode,
}: {
widgetId: string;
apiResponse: MetricRangePayloadProps;
tzDate: uPlot.LocalDateFromUnix;
minTimeScale: number | undefined;
maxTimeScale: number | undefined;
isLogScale: boolean;
softMin: number | null;
softMax: number | null;
spanGaps: boolean;
colorMapping: Record<string, string>;
lineInterpolation: LineInterpolation;
isDarkMode: boolean;
thresholds: ThresholdsDrawHookOptions;
currentQuery: Query;
yAxisUnit: string;
onDragSelect: (startTime: number, endTime: number) => void;
onClick?: OnClickPluginOpts['onClick'];
panelMode: PanelMode;
}): UPlotConfigBuilder => {
const builder = new UPlotConfigBuilder({
onDragSelect,
widgetId,
tzDate,
shouldSaveSelectionPreference: panelMode === PanelMode.DASHBOARD_VIEW,
selectionPreferencesSource: [
PanelMode.DASHBOARD_VIEW,
PanelMode.STANDALONE_VIEW,
].includes(panelMode)
? SelectionPreferencesSource.LOCAL_STORAGE
: SelectionPreferencesSource.IN_MEMORY,
});
// X scale time axis
builder.addScale({
scaleKey: 'x',
time: true,
min: minTimeScale,
max: maxTimeScale,
logBase: isLogScale ? 10 : undefined,
distribution: isLogScale
? DistributionType.Logarithmic
: DistributionType.Linear,
});
// Y scale value axis, driven primarily by softMin/softMax and data
builder.addScale({
scaleKey: 'y',
time: false,
min: undefined,
max: undefined,
softMin: softMin ?? undefined,
softMax: softMax ?? undefined,
thresholds,
logBase: isLogScale ? 10 : undefined,
distribution: isLogScale
? DistributionType.Logarithmic
: DistributionType.Linear,
});
builder.addThresholds(thresholds);
if (typeof onClick === 'function') {
builder.addPlugin(
onClickPlugin({
onClick,
apiResponse,
}),
);
}
builder.addAxis({
scaleKey: 'x',
show: true,
side: 2,
isDarkMode,
isLogScale: false,
panelType: PANEL_TYPES.TIME_SERIES,
});
builder.addAxis({
scaleKey: 'y',
show: true,
side: 3,
isDarkMode,
isLogScale: false,
yAxisUnit,
panelType: PANEL_TYPES.TIME_SERIES,
});
apiResponse.data?.result?.forEach((series) => {
const baseLabelName = getLabelName(
series.metric,
series.queryName || '', // query
series.legend || '',
);
const label = currentQuery
? getLegend(series, currentQuery, baseLabelName)
: baseLabelName;
builder.addSeries({
scaleKey: 'y',
drawStyle: DrawStyle.Line,
label: label,
colorMapping,
spanGaps,
lineStyle: LineStyle.Solid,
lineInterpolation,
showPoints: VisibilityMode.Never,
pointSize: 5,
isDarkMode,
});
});
return builder;
};

View File

@@ -1,54 +0,0 @@
import { normalizePlotValue } from 'lib/uPlotV2/utils/dataUtils';
import { QueryData } from 'types/api/widgets/getQuery';
export function getXAxisTimestamps(seriesList: QueryData[]): number[] {
const timestamps = new Set<number>();
seriesList.forEach((series: { values?: [number, string][] }) => {
if (series?.values) {
series.values.forEach((value) => {
timestamps.add(value[0]);
});
}
});
const timestampsArr = Array.from(timestamps);
timestampsArr.sort((a, b) => a - b);
return timestampsArr;
}
export function fillMissingXAxisTimestamps(
timestampArr: number[],
data: Array<{ values?: [number, string][] }>,
): (number | null)[][] {
// Ensure we work with a sorted, deduplicated list of x-axis timestamps
const canonicalTimestamps = Array.from(new Set(timestampArr)).sort(
(a, b) => a - b,
);
return data.map(({ values }) =>
buildSeriesYValues(canonicalTimestamps, values),
);
}
function buildSeriesYValues(
timestamps: number[],
values?: [number, string][],
): (number | null)[] {
if (!values?.length) {
return [];
}
const valueByTimestamp = new Map<number, number | null>();
for (let i = 0; i < values.length; i++) {
const [timestamp, rawValue] = values[i];
valueByTimestamp.set(timestamp, normalizePlotValue(rawValue));
}
return timestamps.map((timestamp) => {
const value = valueByTimestamp.get(timestamp);
return value !== undefined ? value : null;
});
}

View File

@@ -108,7 +108,7 @@ const createMockSpan = (): Span => ({
statusMessage: '',
tagMap: {
'http.method': 'GET',
http_url: '/api/users?page=1',
'http.url': '/api/users?page=1',
'http.status_code': '200',
'service.name': 'frontend-service',
'span.kind': 'server',

View File

@@ -878,7 +878,7 @@ describe('SpanDetailsDrawer', () => {
// Verify only matching attributes are shown (use getAllByText for all since they appear in multiple places)
expect(screen.getAllByText('http.method').length).toBeGreaterThan(0);
expect(screen.getAllByText('http_url').length).toBeGreaterThan(0);
expect(screen.getAllByText('http.url').length).toBeGreaterThan(0);
expect(screen.getAllByText('http.status_code').length).toBeGreaterThan(0);
});
@@ -1126,7 +1126,7 @@ describe('SpanDetailsDrawer - Search Visibility User Flows', () => {
// User sees all attributes initially
expect(screen.getByText('http.method')).toBeInTheDocument();
expect(screen.getByText('http_url')).toBeInTheDocument();
expect(screen.getByText('http.url')).toBeInTheDocument();
expect(screen.getByText('http.status_code')).toBeInTheDocument();
// User types "method" in search
@@ -1136,7 +1136,7 @@ describe('SpanDetailsDrawer - Search Visibility User Flows', () => {
// User sees only matching attributes
await waitFor(() => {
expect(screen.getByText('http.method')).toBeInTheDocument();
expect(screen.queryByText('http_url')).not.toBeInTheDocument();
expect(screen.queryByText('http.url')).not.toBeInTheDocument();
expect(screen.queryByText('http.status_code')).not.toBeInTheDocument();
});
});

View File

@@ -22,7 +22,7 @@ export const mockSpan: Span = {
event: [],
tagMap: {
'http.method': 'GET',
http_url: '/api/test',
'http.url': '/api/test',
'http.status_code': '200',
},
hasError: false,

View File

@@ -24,14 +24,6 @@
opacity: 1;
}
.legend-empty-state {
font-size: 12px;
color: var(--bg-vanilla-400);
text-align: center;
padding: 12px;
padding: 2rem 0;
}
.legend-virtuoso-container {
height: 100%;
width: 100%;

View File

@@ -81,13 +81,6 @@ export default function Legend({
[focusedSeriesIndex, position],
);
const isEmptyState = useMemo(() => {
if (position !== LegendPosition.RIGHT || !legendSearchQuery.trim()) {
return false;
}
return visibleLegendItems.length === 0;
}, [position, legendSearchQuery, visibleLegendItems]);
return (
<div
ref={legendContainerRef}
@@ -110,21 +103,15 @@ export default function Legend({
/>
</div>
)}
{isEmptyState ? (
<div className="legend-empty-state">
No series found matching &quot;{legendSearchQuery}&quot;
</div>
) : (
<VirtuosoGrid
className={cx(
'legend-virtuoso-container',
`legend-virtuoso-container-${position.toLowerCase()}`,
{ 'legend-virtuoso-container-single-row': isSingleRow },
)}
data={visibleLegendItems}
itemContent={(_, item): JSX.Element => renderLegendItem(item)}
/>
)}
<VirtuosoGrid
className={cx(
'legend-virtuoso-container',
`legend-virtuoso-container-${position.toLowerCase()}`,
{ 'legend-virtuoso-container-single-row': isSingleRow },
)}
data={visibleLegendItems}
itemContent={(_, item): JSX.Element => renderLegendItem(item)}
/>
</div>
);
}

View File

@@ -35,7 +35,7 @@ export const traceDetailResponse = [
'component',
'host.name',
'http.method',
'http_url',
'http.url',
'ip',
'http.status_code',
'opencensus.exporterversion',
@@ -84,7 +84,7 @@ export const traceDetailResponse = [
'signoz.collector.id',
'component',
'http.method',
'http_url',
'http.url',
'ip',
],
[
@@ -741,7 +741,7 @@ export const traceDetailResponse = [
'component',
'http.method',
'http.status_code',
'http_url',
'http.url',
'net/http.reused',
'net/http.was_idle',
'service.name',
@@ -833,7 +833,7 @@ export const traceDetailResponse = [
'opencensus.exporterversion',
'signoz.collector.id',
'host.name',
'http_url',
'http.url',
'net/http.reused',
'net/http.was_idle',
],
@@ -916,7 +916,7 @@ export const traceDetailResponse = [
'net/http.was_idle',
'component',
'host.name',
'http_url',
'http.url',
'ip',
'service.name',
'signoz.collector.id',

View File

@@ -31,7 +31,7 @@ export const AllTraceFilterKeyValue: Record<string, string> = {
httpRoute: 'HTTP Route',
'http.route': 'HTTP Route',
httpUrl: 'HTTP URL',
http_url: 'HTTP URL',
'http.url': 'HTTP URL',
traceID: 'Trace ID',
trace_id: 'Trace ID',
} as const;

View File

@@ -4,7 +4,6 @@ import {
} from 'container/DashboardContainer/DashboardVariablesSelection/util';
import { IDashboardVariable } from 'types/api/dashboard/getAll';
import { initializeVariableFetchStore } from '../variableFetchStore';
import {
IDashboardVariables,
IDashboardVariablesStoreState,
@@ -63,30 +62,9 @@ export function buildDependencyData(
};
}
/**
* Initialize the variable fetch store with the computed dependency data
*/
function initializeFetchStore(
sortedVariablesArray: IDashboardVariable[],
dependencyData: IDependencyData | null,
): void {
if (dependencyData) {
const allVariableNames = sortedVariablesArray
.map((v) => v.name)
.filter((name): name is string => !!name);
initializeVariableFetchStore(
allVariableNames,
dependencyData.graph,
dependencyData.parentDependencyGraph,
);
}
}
/**
* Compute derived values from variables
* This is a composition of buildSortedVariablesArray and buildDependencyData
* Also initializes the variable fetch store with the new dependency data
*/
export function computeDerivedValues(
variables: IDashboardVariablesStoreState['variables'],
@@ -97,22 +75,15 @@ export function computeDerivedValues(
const sortedVariablesArray = buildSortedVariablesArray(variables);
const dependencyData = buildDependencyData(sortedVariablesArray);
// Initialize the variable fetch store when dependency data is computed
initializeFetchStore(sortedVariablesArray, dependencyData);
return { sortedVariablesArray, dependencyData };
}
/**
* Update derived values in the store state (for use with immer)
* Also initializes the variable fetch store with the new dependency data
*/
export function updateDerivedValues(
draft: IDashboardVariablesStoreState,
): void {
draft.sortedVariablesArray = buildSortedVariablesArray(draft.variables);
draft.dependencyData = buildDependencyData(draft.sortedVariablesArray);
// Initialize the variable fetch store when dependency data is updated
initializeFetchStore(draft.sortedVariablesArray, draft.dependencyData);
}

View File

@@ -1,57 +0,0 @@
import { VariableGraph } from 'container/DashboardContainer/DashboardVariablesSelection/util';
import createStore from './store';
// Fetch state for each variable
export type VariableFetchState =
| 'idle' // stable state - initial or complete
| 'loading' // actively fetching data (first time)
| 'revalidating' // refetching existing data
| 'waiting' // blocked on parent dependencies
| 'error';
export interface IVariableFetchStoreState {
// Per-variable fetch state
states: Record<string, VariableFetchState>;
// Dependency graphs (set once when variables change)
dependencyGraph: VariableGraph; // variable -> children that depend on it
parentGraph: VariableGraph; // variable -> parents it depends on
// Track last update timestamp per variable to trigger re-fetches
lastUpdated: Record<string, number>;
}
const initialState: IVariableFetchStoreState = {
states: {},
dependencyGraph: {},
parentGraph: {},
lastUpdated: {},
};
export const variableFetchStore = createStore<IVariableFetchStoreState>(
initialState,
);
// ============== Actions ==============
/**
* Initialize the store with dependency graphs and set initial states
*/
export function initializeVariableFetchStore(
variableNames: string[],
dependencyGraph: VariableGraph,
parentGraph: VariableGraph,
): void {
variableFetchStore.update((draft) => {
draft.dependencyGraph = dependencyGraph;
draft.parentGraph = parentGraph;
// Initialize all variables to idle, preserving existing ready states
variableNames.forEach((name) => {
if (!draft.states[name]) {
draft.states[name] = 'idle';
}
});
});
}

View File

@@ -74,6 +74,21 @@ type Alertmanager interface {
CreateInhibitRules(ctx context.Context, orgID valuer.UUID, rules []amConfig.InhibitRule) error
DeleteAllInhibitRulesByRuleId(ctx context.Context, orgID valuer.UUID, ruleId string) error
// Planned Maintenance CRUD
GetAllPlannedMaintenance(ctx context.Context, orgID string) ([]*alertmanagertypes.GettablePlannedMaintenance, error)
GetPlannedMaintenanceByID(ctx context.Context, id valuer.UUID) (*alertmanagertypes.GettablePlannedMaintenance, error)
CreatePlannedMaintenance(ctx context.Context, maintenance alertmanagertypes.GettablePlannedMaintenance) (valuer.UUID, error)
EditPlannedMaintenance(ctx context.Context, maintenance alertmanagertypes.GettablePlannedMaintenance, id valuer.UUID) error
DeletePlannedMaintenance(ctx context.Context, id valuer.UUID) error
// Rule State History
RecordRuleStateHistory(ctx context.Context, orgID string, entries []alertmanagertypes.RuleStateHistory) error
GetLastSavedRuleStateHistory(ctx context.Context, ruleID string) ([]alertmanagertypes.RuleStateHistory, error)
GetRuleStateHistoryTimeline(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStateTimeline, error)
GetRuleStateHistoryTopContributors(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateHistoryContributor, error)
GetOverallStateTransitions(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateTransition, error)
GetRuleStats(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStats, error)
// Collects stats for the organization.
statsreporter.StatsCollector
}

View File

@@ -68,20 +68,30 @@ type Server struct {
wg sync.WaitGroup
stopc chan struct{}
notificationManager nfmanager.NotificationManager
// maintenanceExprMuter is an optional muter for expression-based maintenance scoping
maintenanceExprMuter types.Muter
// muteStageMetrics are created once and reused across SetConfig calls
muteStageMetrics *notify.Metrics
// signozRegisterer is used for metrics in the pipeline
signozRegisterer prometheus.Registerer
}
func New(ctx context.Context, logger *slog.Logger, registry prometheus.Registerer, srvConfig Config, orgID string, stateStore alertmanagertypes.StateStore, nfManager nfmanager.NotificationManager) (*Server, error) {
func New(ctx context.Context, logger *slog.Logger, registry prometheus.Registerer, srvConfig Config, orgID string, stateStore alertmanagertypes.StateStore, nfManager nfmanager.NotificationManager, maintenanceExprMuter types.Muter) (*Server, error) {
server := &Server{
logger: logger.With("pkg", "go.signoz.io/pkg/alertmanager/alertmanagerserver"),
registry: registry,
srvConfig: srvConfig,
orgID: orgID,
stateStore: stateStore,
stopc: make(chan struct{}),
notificationManager: nfManager,
logger: logger.With("pkg", "go.signoz.io/pkg/alertmanager/alertmanagerserver"),
registry: registry,
srvConfig: srvConfig,
orgID: orgID,
stateStore: stateStore,
stopc: make(chan struct{}),
notificationManager: nfManager,
maintenanceExprMuter: maintenanceExprMuter,
}
signozRegisterer := prometheus.WrapRegistererWithPrefix("signoz_", registry)
signozRegisterer = prometheus.WrapRegistererWith(prometheus.Labels{"org_id": server.orgID}, signozRegisterer)
server.signozRegisterer = prometheus.WrapRegistererWithPrefix("signoz_", registry)
server.signozRegisterer = prometheus.WrapRegistererWith(prometheus.Labels{"org_id": server.orgID}, server.signozRegisterer)
signozRegisterer := server.signozRegisterer
// initialize marker
server.marker = alertmanagertypes.NewMarker(signozRegisterer)
@@ -198,6 +208,11 @@ func New(ctx context.Context, logger *slog.Logger, registry prometheus.Registere
server.pipelineBuilder = notify.NewPipelineBuilder(signozRegisterer, featurecontrol.NoopFlags{})
server.dispatcherMetrics = NewDispatcherMetrics(false, signozRegisterer)
if server.maintenanceExprMuter != nil {
muteRegisterer := prometheus.WrapRegistererWithPrefix("maintenance_mute_", signozRegisterer)
server.muteStageMetrics = notify.NewMetrics(muteRegisterer, featurecontrol.NoopFlags{})
}
return server, nil
}
@@ -205,6 +220,9 @@ func (server *Server) GetAlerts(ctx context.Context, params alertmanagertypes.Ge
return alertmanagertypes.NewGettableAlertsFromAlertProvider(server.alerts, server.alertmanagerConfig, server.marker.Status, func(labels model.LabelSet) {
server.inhibitor.Mutes(labels)
server.silencer.Mutes(labels)
if server.maintenanceExprMuter != nil {
server.maintenanceExprMuter.Mutes(labels)
}
}, params)
}
@@ -293,6 +311,14 @@ func (server *Server) SetConfig(ctx context.Context, alertmanagerConfig *alertma
pipelinePeer,
)
// Inject expression-based maintenance muter into the pipeline
if server.maintenanceExprMuter != nil {
ms := notify.NewMuteStage(server.maintenanceExprMuter, server.muteStageMetrics)
for name, stage := range pipeline {
pipeline[name] = notify.MultiStage{ms, stage}
}
}
timeoutFunc := func(d time.Duration) time.Duration {
if d < notify.MinTimeout {
d = notify.MinTimeout

View File

@@ -4,6 +4,7 @@ import (
"context"
"log/slog"
"net/http"
"sync"
"testing"
"time"
@@ -22,9 +23,35 @@ import (
"github.com/prometheus/client_golang/prometheus"
"github.com/prometheus/common/model"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
// testMuter implements types.Muter for testing maintenance expression muting.
type testMuter struct {
mu sync.RWMutex
muteFunc func(model.LabelSet) bool
calls []model.LabelSet
}
func (m *testMuter) Mutes(labels model.LabelSet) bool {
m.mu.Lock()
defer m.mu.Unlock()
m.calls = append(m.calls, labels)
if m.muteFunc != nil {
return m.muteFunc(labels)
}
return false
}
func (m *testMuter) getCalls() []model.LabelSet {
m.mu.RLock()
defer m.mu.RUnlock()
result := make([]model.LabelSet, len(m.calls))
copy(result, m.calls)
return result
}
func TestEndToEndAlertManagerFlow(t *testing.T) {
ctx := context.Background()
providerSettings := instrumentationtest.New().ToProviderSettings()
@@ -90,7 +117,7 @@ func TestEndToEndAlertManagerFlow(t *testing.T) {
stateStore := alertmanagertypestest.NewStateStore()
registry := prometheus.NewRegistry()
logger := slog.New(slog.DiscardHandler)
server, err := New(context.Background(), logger, registry, srvCfg, orgID, stateStore, notificationManager)
server, err := New(context.Background(), logger, registry, srvCfg, orgID, stateStore, notificationManager, nil)
require.NoError(t, err)
amConfig, err := alertmanagertypes.NewDefaultConfig(srvCfg.Global, srvCfg.Route, orgID)
require.NoError(t, err)
@@ -221,3 +248,257 @@ func TestEndToEndAlertManagerFlow(t *testing.T) {
require.Equal(t, "{__receiver__=\"webhook\"}:{cluster=\"prod-cluster\", instance=\"server-03\", ruleId=\"high-cpu-usage\"}", alertGroups[2].GroupKey)
})
}
// TestEndToEndMaintenanceMuting verifies that the maintenance expression muter
// integrates correctly with the alertmanager server pipeline:
// 1. MuteStage is injected into the notification pipeline when a muter is provided
// 2. Alerts remain visible in GetAlerts during maintenance (muting suppresses
// notifications, not alert visibility)
// 3. The muter is called during GetAlerts for status resolution
func TestEndToEndMaintenanceMuting(t *testing.T) {
ctx := context.Background()
providerSettings := instrumentationtest.New().ToProviderSettings()
store := nfroutingstoretest.NewMockSQLRouteStore()
store.MatchExpectationsInOrder(false)
notificationManager, err := rulebasednotification.New(ctx, providerSettings, nfmanager.Config{}, store)
require.NoError(t, err)
orgID := "test-org-maintenance"
// Create a muter that mutes alerts with severity == "critical"
muter := &testMuter{
muteFunc: func(labels model.LabelSet) bool {
return string(labels["severity"]) == "critical"
},
}
srvCfg := NewConfig()
stateStore := alertmanagertypestest.NewStateStore()
registry := prometheus.NewRegistry()
logger := slog.New(slog.DiscardHandler)
// Create server WITH the maintenance muter
server, err := New(ctx, logger, registry, srvCfg, orgID, stateStore, notificationManager, muter)
require.NoError(t, err)
amConfig, err := alertmanagertypes.NewDefaultConfig(srvCfg.Global, srvCfg.Route, orgID)
require.NoError(t, err)
err = server.SetConfig(ctx, amConfig)
require.NoError(t, err)
// Put a mix of alerts: 2 critical (should be muted) and 1 warning (should not)
now := time.Now()
testAlerts := []*alertmanagertypes.PostableAlert{
{
Alert: alertmanagertypes.AlertModel{
Labels: map[string]string{
"ruleId": "disk-usage",
"severity": "critical",
"env": "prod",
"alertname": "DiskUsageHigh",
},
},
Annotations: map[string]string{"summary": "Disk usage critical"},
StartsAt: strfmt.DateTime(now.Add(-5 * time.Minute)),
EndsAt: strfmt.DateTime(time.Time{}),
},
{
Alert: alertmanagertypes.AlertModel{
Labels: map[string]string{
"ruleId": "disk-usage",
"severity": "warning",
"env": "prod",
"alertname": "DiskUsageHigh",
},
},
Annotations: map[string]string{"summary": "Disk usage warning"},
StartsAt: strfmt.DateTime(now.Add(-3 * time.Minute)),
EndsAt: strfmt.DateTime(time.Time{}),
},
{
Alert: alertmanagertypes.AlertModel{
Labels: map[string]string{
"ruleId": "memory-usage",
"severity": "critical",
"env": "staging",
"alertname": "MemoryUsageHigh",
},
},
Annotations: map[string]string{"summary": "Memory usage critical"},
StartsAt: strfmt.DateTime(now.Add(-2 * time.Minute)),
EndsAt: strfmt.DateTime(time.Time{}),
},
}
err = server.PutAlerts(ctx, testAlerts)
require.NoError(t, err)
time.Sleep(2 * time.Second)
t.Run("alerts_visible_during_maintenance", func(t *testing.T) {
// Maintenance muting suppresses notifications, NOT alert visibility.
// All 3 alerts should still be returned by GetAlerts.
req, err := http.NewRequest(http.MethodGet, "/alerts", nil)
require.NoError(t, err)
params, err := alertmanagertypes.NewGettableAlertsParams(req)
require.NoError(t, err)
alerts, err := server.GetAlerts(ctx, params)
require.NoError(t, err)
require.Len(t, alerts, 3, "All alerts should be visible during maintenance")
// Verify labels are intact
severities := map[string]int{}
for _, alert := range alerts {
severities[alert.Alert.Labels["severity"]]++
}
assert.Equal(t, 2, severities["critical"])
assert.Equal(t, 1, severities["warning"])
})
t.Run("muter_called_during_get_alerts", func(t *testing.T) {
// The muter should have been called for each alert during GetAlerts.
calls := muter.getCalls()
assert.GreaterOrEqual(t, len(calls), 3, "Muter should be called for each alert")
})
t.Run("muter_correctly_identifies_targets", func(t *testing.T) {
// Verify the muter returns correct results for different label sets
assert.True(t, muter.Mutes(model.LabelSet{"severity": "critical", "env": "prod"}),
"Should mute critical alerts")
assert.False(t, muter.Mutes(model.LabelSet{"severity": "warning", "env": "prod"}),
"Should not mute warning alerts")
assert.True(t, muter.Mutes(model.LabelSet{"severity": "critical", "env": "staging"}),
"Should mute critical regardless of env")
})
}
// TestEndToEndMaintenanceCatchAll verifies that a catch-all muter (always returns true)
// mutes all alerts while keeping them visible.
func TestEndToEndMaintenanceCatchAll(t *testing.T) {
ctx := context.Background()
providerSettings := instrumentationtest.New().ToProviderSettings()
store := nfroutingstoretest.NewMockSQLRouteStore()
store.MatchExpectationsInOrder(false)
notificationManager, err := rulebasednotification.New(ctx, providerSettings, nfmanager.Config{}, store)
require.NoError(t, err)
orgID := "test-org-catchall"
// Catch-all muter: mutes everything
muter := &testMuter{
muteFunc: func(labels model.LabelSet) bool {
return true
},
}
srvCfg := NewConfig()
stateStore := alertmanagertypestest.NewStateStore()
registry := prometheus.NewRegistry()
logger := slog.New(slog.DiscardHandler)
server, err := New(ctx, logger, registry, srvCfg, orgID, stateStore, notificationManager, muter)
require.NoError(t, err)
amConfig, err := alertmanagertypes.NewDefaultConfig(srvCfg.Global, srvCfg.Route, orgID)
require.NoError(t, err)
err = server.SetConfig(ctx, amConfig)
require.NoError(t, err)
now := time.Now()
testAlerts := []*alertmanagertypes.PostableAlert{
{
Alert: alertmanagertypes.AlertModel{
Labels: map[string]string{
"ruleId": "rule-1", "alertname": "Alert1", "env": "prod",
},
},
StartsAt: strfmt.DateTime(now.Add(-1 * time.Minute)),
EndsAt: strfmt.DateTime(time.Time{}),
},
{
Alert: alertmanagertypes.AlertModel{
Labels: map[string]string{
"ruleId": "rule-2", "alertname": "Alert2", "env": "staging",
},
},
StartsAt: strfmt.DateTime(now.Add(-1 * time.Minute)),
EndsAt: strfmt.DateTime(time.Time{}),
},
}
err = server.PutAlerts(ctx, testAlerts)
require.NoError(t, err)
time.Sleep(2 * time.Second)
req, err := http.NewRequest(http.MethodGet, "/alerts", nil)
require.NoError(t, err)
params, err := alertmanagertypes.NewGettableAlertsParams(req)
require.NoError(t, err)
alerts, err := server.GetAlerts(ctx, params)
require.NoError(t, err)
assert.Len(t, alerts, 2, "All alerts should remain visible even when catch-all muter is active")
// Verify the muter was called for each alert
calls := muter.getCalls()
assert.GreaterOrEqual(t, len(calls), 2, "Muter should be called for each alert")
}
// TestEndToEndNoMuter verifies the server works correctly without a muter (nil),
// matching the existing behavior where no maintenance muting is configured.
func TestEndToEndNoMuter(t *testing.T) {
ctx := context.Background()
providerSettings := instrumentationtest.New().ToProviderSettings()
store := nfroutingstoretest.NewMockSQLRouteStore()
store.MatchExpectationsInOrder(false)
notificationManager, err := rulebasednotification.New(ctx, providerSettings, nfmanager.Config{}, store)
require.NoError(t, err)
orgID := "test-org-nomuter"
srvCfg := NewConfig()
stateStore := alertmanagertypestest.NewStateStore()
registry := prometheus.NewRegistry()
logger := slog.New(slog.DiscardHandler)
// Create server WITHOUT a muter (nil)
server, err := New(ctx, logger, registry, srvCfg, orgID, stateStore, notificationManager, nil)
require.NoError(t, err)
amConfig, err := alertmanagertypes.NewDefaultConfig(srvCfg.Global, srvCfg.Route, orgID)
require.NoError(t, err)
err = server.SetConfig(ctx, amConfig)
require.NoError(t, err)
now := time.Now()
testAlerts := []*alertmanagertypes.PostableAlert{
{
Alert: alertmanagertypes.AlertModel{
Labels: map[string]string{
"ruleId": "rule-1", "alertname": "Alert1", "severity": "critical",
},
},
StartsAt: strfmt.DateTime(now.Add(-1 * time.Minute)),
EndsAt: strfmt.DateTime(time.Time{}),
},
}
err = server.PutAlerts(ctx, testAlerts)
require.NoError(t, err)
time.Sleep(2 * time.Second)
req, err := http.NewRequest(http.MethodGet, "/alerts", nil)
require.NoError(t, err)
params, err := alertmanagertypes.NewGettableAlertsParams(req)
require.NoError(t, err)
alerts, err := server.GetAlerts(ctx, params)
require.NoError(t, err)
assert.Len(t, alerts, 1, "Alert should be returned when no muter is configured")
assert.Equal(t, "critical", alerts[0].Alert.Labels["severity"])
}

View File

@@ -25,7 +25,7 @@ import (
func TestServerSetConfigAndStop(t *testing.T) {
notificationManager := nfmanagertest.NewMock()
server, err := New(context.Background(), slog.New(slog.DiscardHandler), prometheus.NewRegistry(), NewConfig(), "1", alertmanagertypestest.NewStateStore(), notificationManager)
server, err := New(context.Background(), slog.New(slog.DiscardHandler), prometheus.NewRegistry(), NewConfig(), "1", alertmanagertypestest.NewStateStore(), notificationManager, nil)
require.NoError(t, err)
amConfig, err := alertmanagertypes.NewDefaultConfig(alertmanagertypes.GlobalConfig{}, alertmanagertypes.RouteConfig{GroupInterval: 1 * time.Minute, RepeatInterval: 1 * time.Minute, GroupWait: 1 * time.Minute}, "1")
@@ -37,7 +37,7 @@ func TestServerSetConfigAndStop(t *testing.T) {
func TestServerTestReceiverTypeWebhook(t *testing.T) {
notificationManager := nfmanagertest.NewMock()
server, err := New(context.Background(), slog.New(slog.DiscardHandler), prometheus.NewRegistry(), NewConfig(), "1", alertmanagertypestest.NewStateStore(), notificationManager)
server, err := New(context.Background(), slog.New(slog.DiscardHandler), prometheus.NewRegistry(), NewConfig(), "1", alertmanagertypestest.NewStateStore(), notificationManager, nil)
require.NoError(t, err)
amConfig, err := alertmanagertypes.NewDefaultConfig(alertmanagertypes.GlobalConfig{}, alertmanagertypes.RouteConfig{GroupInterval: 1 * time.Minute, RepeatInterval: 1 * time.Minute, GroupWait: 1 * time.Minute}, "1")
@@ -85,7 +85,7 @@ func TestServerPutAlerts(t *testing.T) {
srvCfg := NewConfig()
srvCfg.Route.GroupInterval = 1 * time.Second
notificationManager := nfmanagertest.NewMock()
server, err := New(context.Background(), slog.New(slog.DiscardHandler), prometheus.NewRegistry(), srvCfg, "1", stateStore, notificationManager)
server, err := New(context.Background(), slog.New(slog.DiscardHandler), prometheus.NewRegistry(), srvCfg, "1", stateStore, notificationManager, nil)
require.NoError(t, err)
amConfig, err := alertmanagertypes.NewDefaultConfig(srvCfg.Global, srvCfg.Route, "1")
@@ -133,7 +133,7 @@ func TestServerTestAlert(t *testing.T) {
srvCfg := NewConfig()
srvCfg.Route.GroupInterval = 1 * time.Second
notificationManager := nfmanagertest.NewMock()
server, err := New(context.Background(), slog.New(slog.DiscardHandler), prometheus.NewRegistry(), srvCfg, "1", stateStore, notificationManager)
server, err := New(context.Background(), slog.New(slog.DiscardHandler), prometheus.NewRegistry(), srvCfg, "1", stateStore, notificationManager, nil)
require.NoError(t, err)
amConfig, err := alertmanagertypes.NewDefaultConfig(srvCfg.Global, srvCfg.Route, "1")
@@ -238,7 +238,7 @@ func TestServerTestAlertContinuesOnFailure(t *testing.T) {
srvCfg := NewConfig()
srvCfg.Route.GroupInterval = 1 * time.Second
notificationManager := nfmanagertest.NewMock()
server, err := New(context.Background(), slog.New(slog.DiscardHandler), prometheus.NewRegistry(), srvCfg, "1", stateStore, notificationManager)
server, err := New(context.Background(), slog.New(slog.DiscardHandler), prometheus.NewRegistry(), srvCfg, "1", stateStore, notificationManager, nil)
require.NoError(t, err)
amConfig, err := alertmanagertypes.NewDefaultConfig(srvCfg.Global, srvCfg.Route, "1")

View File

@@ -0,0 +1,531 @@
package clickhousealertmanagerstore
import (
"context"
"encoding/json"
"fmt"
"strings"
"github.com/ClickHouse/clickhouse-go/v2"
"github.com/SigNoz/signoz/pkg/types/alertmanagertypes"
"github.com/SigNoz/signoz/pkg/valuer"
)
const (
signozHistoryDBName = "signoz_analytics"
ruleStateHistoryTableName = "distributed_rule_state_history_v2"
maxPointsInTimeSeries = 300
)
type stateHistoryStore struct {
conn clickhouse.Conn
}
func NewStateHistoryStore(conn clickhouse.Conn) alertmanagertypes.StateHistoryStore {
return &stateHistoryStore{conn: conn}
}
func (s *stateHistoryStore) WriteRuleStateHistory(ctx context.Context, entries []alertmanagertypes.RuleStateHistory) error {
if len(entries) == 0 {
return nil
}
statement, err := s.conn.PrepareBatch(ctx, fmt.Sprintf(
"INSERT INTO %s.%s (org_id, rule_id, rule_name, overall_state, overall_state_changed, state, state_changed, unix_milli, labels, fingerprint, value) VALUES ($1, $2, $3, $4, $5, $6, $7, $8, $9, $10, $11)",
signozHistoryDBName, ruleStateHistoryTableName))
if err != nil {
return err
}
defer statement.Abort()
for _, h := range entries {
if err := statement.Append(
h.OrgID,
h.RuleID, h.RuleName,
h.OverallState, h.OverallStateChanged,
h.State, h.StateChanged,
h.UnixMilli, h.Labels,
h.Fingerprint, h.Value,
); err != nil {
return err
}
}
return statement.Send()
}
func (s *stateHistoryStore) GetLastSavedRuleStateHistory(ctx context.Context, ruleID string) ([]alertmanagertypes.RuleStateHistory, error) {
query := fmt.Sprintf(
"SELECT org_id, rule_id, rule_name, overall_state, overall_state_changed, state, state_changed, unix_milli, labels, fingerprint, value FROM %s.%s WHERE rule_id = '%s' AND state_changed = true ORDER BY unix_milli DESC LIMIT 1 BY fingerprint",
signozHistoryDBName, ruleStateHistoryTableName, ruleID)
rows, err := s.conn.Query(ctx, query)
if err != nil {
return nil, err
}
defer rows.Close()
var results []alertmanagertypes.RuleStateHistory
for rows.Next() {
var h alertmanagertypes.RuleStateHistory
if err := rows.Scan(
&h.OrgID,
&h.RuleID, &h.RuleName,
&h.OverallState, &h.OverallStateChanged,
&h.State, &h.StateChanged,
&h.UnixMilli, &h.Labels,
&h.Fingerprint, &h.Value,
); err != nil {
return nil, err
}
results = append(results, h)
}
return results, rows.Err()
}
func (s *stateHistoryStore) GetRuleStateHistoryTimeline(
ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory,
) (*alertmanagertypes.RuleStateTimeline, error) {
var conditions []string
conditions = append(conditions, fmt.Sprintf("org_id = '%s'", orgID))
conditions = append(conditions, fmt.Sprintf("rule_id = '%s'", ruleID))
conditions = append(conditions, fmt.Sprintf("unix_milli >= %d AND unix_milli < %d", params.Start, params.End))
if params.State.StringValue() != "" {
conditions = append(conditions, fmt.Sprintf("state = '%s'", params.State.StringValue()))
}
whereClause := strings.Join(conditions, " AND ")
// Main query — paginated results.
query := fmt.Sprintf(
"SELECT org_id, rule_id, rule_name, overall_state, overall_state_changed, state, state_changed, unix_milli, labels, fingerprint, value FROM %s.%s WHERE %s ORDER BY unix_milli %s LIMIT %d OFFSET %d",
signozHistoryDBName, ruleStateHistoryTableName, whereClause, params.Order.StringValue(), params.Limit, params.Offset)
rows, err := s.conn.Query(ctx, query)
if err != nil {
return nil, err
}
defer rows.Close()
var items []alertmanagertypes.RuleStateHistory
for rows.Next() {
var h alertmanagertypes.RuleStateHistory
if err := rows.Scan(
&h.OrgID,
&h.RuleID, &h.RuleName,
&h.OverallState, &h.OverallStateChanged,
&h.State, &h.StateChanged,
&h.UnixMilli, &h.Labels,
&h.Fingerprint, &h.Value,
); err != nil {
return nil, err
}
items = append(items, h)
}
if err := rows.Err(); err != nil {
return nil, err
}
// Count query.
var total uint64
countQuery := fmt.Sprintf("SELECT count(*) FROM %s.%s WHERE %s",
signozHistoryDBName, ruleStateHistoryTableName, whereClause)
if err := s.conn.QueryRow(ctx, countQuery).Scan(&total); err != nil {
return nil, err
}
// Labels query — distinct labels for the rule.
labelsQuery := fmt.Sprintf("SELECT DISTINCT labels FROM %s.%s WHERE org_id = '%s' AND rule_id = '%s'",
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID)
labelRows, err := s.conn.Query(ctx, labelsQuery)
if err != nil {
return nil, err
}
defer labelRows.Close()
labelsMap := make(map[string][]string)
for labelRows.Next() {
var rawLabel string
if err := labelRows.Scan(&rawLabel); err != nil {
return nil, err
}
label := map[string]string{}
if err := json.Unmarshal([]byte(rawLabel), &label); err != nil {
continue
}
for k, v := range label {
labelsMap[k] = append(labelsMap[k], v)
}
}
if items == nil {
items = []alertmanagertypes.RuleStateHistory{}
}
return &alertmanagertypes.RuleStateTimeline{
Items: items,
Total: total,
Labels: labelsMap,
}, nil
}
func (s *stateHistoryStore) GetRuleStateHistoryTopContributors(
ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory,
) ([]alertmanagertypes.RuleStateHistoryContributor, error) {
query := fmt.Sprintf(`SELECT
fingerprint,
any(labels) as labels,
count(*) as count
FROM %s.%s
WHERE org_id = '%s' AND rule_id = '%s' AND (state_changed = true) AND (state = 'firing') AND unix_milli >= %d AND unix_milli <= %d
GROUP BY fingerprint
HAVING labels != '{}'
ORDER BY count DESC`,
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.Start, params.End)
rows, err := s.conn.Query(ctx, query)
if err != nil {
return nil, err
}
defer rows.Close()
var contributors []alertmanagertypes.RuleStateHistoryContributor
for rows.Next() {
var c alertmanagertypes.RuleStateHistoryContributor
if err := rows.Scan(&c.Fingerprint, &c.Labels, &c.Count); err != nil {
return nil, err
}
contributors = append(contributors, c)
}
if contributors == nil {
contributors = []alertmanagertypes.RuleStateHistoryContributor{}
}
return contributors, rows.Err()
}
func (s *stateHistoryStore) GetOverallStateTransitions(
ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory,
) ([]alertmanagertypes.RuleStateTransition, error) {
tmpl := `WITH firing_events AS (
SELECT
rule_id,
state,
unix_milli AS firing_time
FROM %s.%s
WHERE overall_state = 'firing'
AND overall_state_changed = true
AND org_id = '%s'
AND rule_id = '%s'
AND unix_milli >= %d AND unix_milli <= %d
),
resolution_events AS (
SELECT
rule_id,
state,
unix_milli AS resolution_time
FROM %s.%s
WHERE overall_state = 'inactive'
AND overall_state_changed = true
AND org_id = '%s'
AND rule_id = '%s'
AND unix_milli >= %d AND unix_milli <= %d
),
matched_events AS (
SELECT
f.rule_id,
f.state,
f.firing_time,
MIN(r.resolution_time) AS resolution_time
FROM firing_events f
LEFT JOIN resolution_events r
ON f.rule_id = r.rule_id
WHERE r.resolution_time > f.firing_time
GROUP BY f.rule_id, f.state, f.firing_time
)
SELECT *
FROM matched_events
ORDER BY firing_time ASC;`
query := fmt.Sprintf(tmpl,
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.Start, params.End,
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.Start, params.End)
type transition struct {
RuleID string `ch:"rule_id"`
State string `ch:"state"`
FiringTime int64 `ch:"firing_time"`
ResolutionTime int64 `ch:"resolution_time"`
}
rows, err := s.conn.Query(ctx, query)
if err != nil {
return nil, err
}
defer rows.Close()
var transitions []transition
for rows.Next() {
var t transition
if err := rows.Scan(&t.RuleID, &t.State, &t.FiringTime, &t.ResolutionTime); err != nil {
return nil, err
}
transitions = append(transitions, t)
}
if err := rows.Err(); err != nil {
return nil, err
}
var stateItems []alertmanagertypes.RuleStateTransition
for idx, item := range transitions {
stateItems = append(stateItems, alertmanagertypes.RuleStateTransition{
State: alertmanagertypes.AlertState{String: valuer.NewString(item.State)},
Start: item.FiringTime,
End: item.ResolutionTime,
})
if idx < len(transitions)-1 {
nextStart := transitions[idx+1].FiringTime
if nextStart > item.ResolutionTime {
stateItems = append(stateItems, alertmanagertypes.RuleStateTransition{
State: alertmanagertypes.AlertStateInactive,
Start: item.ResolutionTime,
End: nextStart,
})
}
}
}
// Fetch the most recent state to fill in edges.
var lastStateStr string
stateQuery := fmt.Sprintf(
"SELECT state FROM %s.%s WHERE org_id = '%s' AND rule_id = '%s' AND unix_milli <= %d ORDER BY unix_milli DESC LIMIT 1",
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.End)
if err := s.conn.QueryRow(ctx, stateQuery).Scan(&lastStateStr); err != nil {
lastStateStr = "inactive"
}
if len(transitions) == 0 {
stateItems = append(stateItems, alertmanagertypes.RuleStateTransition{
State: alertmanagertypes.AlertState{String: valuer.NewString(lastStateStr)},
Start: params.Start,
End: params.End,
})
} else {
if lastStateStr == "inactive" {
stateItems = append(stateItems, alertmanagertypes.RuleStateTransition{
State: alertmanagertypes.AlertStateInactive,
Start: transitions[len(transitions)-1].ResolutionTime,
End: params.End,
})
} else {
// Find the most recent firing event.
var firingTime int64
firingQuery := fmt.Sprintf(
"SELECT unix_milli FROM %s.%s WHERE org_id = '%s' AND rule_id = '%s' AND overall_state_changed = true AND overall_state = 'firing' AND unix_milli <= %d ORDER BY unix_milli DESC LIMIT 1",
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.End)
if err := s.conn.QueryRow(ctx, firingQuery).Scan(&firingTime); err != nil {
firingTime = transitions[len(transitions)-1].ResolutionTime
}
stateItems = append(stateItems, alertmanagertypes.RuleStateTransition{
State: alertmanagertypes.AlertStateInactive,
Start: transitions[len(transitions)-1].ResolutionTime,
End: firingTime,
})
stateItems = append(stateItems, alertmanagertypes.RuleStateTransition{
State: alertmanagertypes.AlertStateFiring,
Start: firingTime,
End: params.End,
})
}
}
return stateItems, nil
}
func (s *stateHistoryStore) GetTotalTriggers(
ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory,
) (uint64, error) {
query := fmt.Sprintf(
"SELECT count(*) FROM %s.%s WHERE org_id = '%s' AND rule_id = '%s' AND (state_changed = true) AND (state = 'firing') AND unix_milli >= %d AND unix_milli <= %d",
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.Start, params.End)
var total uint64
if err := s.conn.QueryRow(ctx, query).Scan(&total); err != nil {
return 0, err
}
return total, nil
}
func (s *stateHistoryStore) GetTriggersByInterval(
ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory,
) (*alertmanagertypes.Series, error) {
step := minAllowedStepInterval(params.Start, params.End)
query := fmt.Sprintf(
"SELECT count(*), toStartOfInterval(toDateTime(intDiv(unix_milli, 1000)), INTERVAL %d SECOND) as ts FROM %s.%s WHERE org_id = '%s' AND rule_id = '%s' AND (state_changed = true) AND (state = 'firing') AND unix_milli >= %d AND unix_milli <= %d GROUP BY ts ORDER BY ts ASC",
step, signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.Start, params.End)
return s.queryTimeSeries(ctx, query)
}
func (s *stateHistoryStore) GetAvgResolutionTime(
ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory,
) (float64, error) {
tmpl := `
WITH firing_events AS (
SELECT
rule_id,
state,
unix_milli AS firing_time
FROM %s.%s
WHERE overall_state = 'firing'
AND overall_state_changed = true
AND org_id = '%s'
AND rule_id = '%s'
AND unix_milli >= %d AND unix_milli <= %d
),
resolution_events AS (
SELECT
rule_id,
state,
unix_milli AS resolution_time
FROM %s.%s
WHERE overall_state = 'inactive'
AND overall_state_changed = true
AND org_id = '%s'
AND rule_id = '%s'
AND unix_milli >= %d AND unix_milli <= %d
),
matched_events AS (
SELECT
f.rule_id,
f.state,
f.firing_time,
MIN(r.resolution_time) AS resolution_time
FROM firing_events f
LEFT JOIN resolution_events r
ON f.rule_id = r.rule_id
WHERE r.resolution_time > f.firing_time
GROUP BY f.rule_id, f.state, f.firing_time
)
SELECT AVG(resolution_time - firing_time) / 1000 AS avg_resolution_time
FROM matched_events;`
query := fmt.Sprintf(tmpl,
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.Start, params.End,
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.Start, params.End)
var avgResolutionTime float64
if err := s.conn.QueryRow(ctx, query).Scan(&avgResolutionTime); err != nil {
return 0, err
}
return avgResolutionTime, nil
}
func (s *stateHistoryStore) GetAvgResolutionTimeByInterval(
ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory,
) (*alertmanagertypes.Series, error) {
step := minAllowedStepInterval(params.Start, params.End)
tmpl := `
WITH firing_events AS (
SELECT
rule_id,
state,
unix_milli AS firing_time
FROM %s.%s
WHERE overall_state = 'firing'
AND overall_state_changed = true
AND org_id = '%s'
AND rule_id = '%s'
AND unix_milli >= %d AND unix_milli <= %d
),
resolution_events AS (
SELECT
rule_id,
state,
unix_milli AS resolution_time
FROM %s.%s
WHERE overall_state = 'inactive'
AND overall_state_changed = true
AND org_id = '%s'
AND rule_id = '%s'
AND unix_milli >= %d AND unix_milli <= %d
),
matched_events AS (
SELECT
f.rule_id,
f.state,
f.firing_time,
MIN(r.resolution_time) AS resolution_time
FROM firing_events f
LEFT JOIN resolution_events r
ON f.rule_id = r.rule_id
WHERE r.resolution_time > f.firing_time
GROUP BY f.rule_id, f.state, f.firing_time
)
SELECT toStartOfInterval(toDateTime(firing_time / 1000), INTERVAL %d SECOND) AS ts, AVG(resolution_time - firing_time) / 1000 AS avg_resolution_time
FROM matched_events
GROUP BY ts
ORDER BY ts ASC;`
query := fmt.Sprintf(tmpl,
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.Start, params.End,
signozHistoryDBName, ruleStateHistoryTableName, orgID, ruleID, params.Start, params.End, step)
return s.queryTimeSeries(ctx, query)
}
func (s *stateHistoryStore) queryTimeSeries(ctx context.Context, query string) (*alertmanagertypes.Series, error) {
rows, err := s.conn.Query(ctx, query)
if err != nil {
return nil, err
}
defer rows.Close()
series := &alertmanagertypes.Series{
Labels: map[string]string{},
}
for rows.Next() {
var value float64
var ts interface{}
if err := rows.Scan(&value, &ts); err != nil {
return nil, err
}
// The timestamp may come back as time.Time from ClickHouse.
var timestamp int64
switch v := ts.(type) {
case int64:
timestamp = v
default:
// Try time.Time
if t, ok := ts.(interface{ UnixMilli() int64 }); ok {
timestamp = t.UnixMilli()
}
}
series.Points = append(series.Points, alertmanagertypes.Point{
Timestamp: timestamp,
Value: value,
})
}
if len(series.Points) == 0 {
return nil, nil
}
return series, rows.Err()
}
func minAllowedStepInterval(start, end int64) int64 {
step := (end - start) / maxPointsInTimeSeries / 1000
if step < 60 {
return 60
}
return step - step%60
}

View File

@@ -0,0 +1,165 @@
package sqlalertmanagerstore
import (
"context"
"encoding/json"
"time"
"github.com/SigNoz/signoz/pkg/sqlstore"
"github.com/SigNoz/signoz/pkg/types"
"github.com/SigNoz/signoz/pkg/types/alertmanagertypes"
"github.com/SigNoz/signoz/pkg/types/authtypes"
"github.com/SigNoz/signoz/pkg/valuer"
)
type maintenance struct {
sqlstore sqlstore.SQLStore
}
func NewMaintenanceStore(store sqlstore.SQLStore) alertmanagertypes.MaintenanceStore {
return &maintenance{sqlstore: store}
}
func (r *maintenance) GetAllPlannedMaintenance(ctx context.Context, orgID string) ([]*alertmanagertypes.GettablePlannedMaintenance, error) {
storables := make([]*alertmanagertypes.StorablePlannedMaintenance, 0)
err := r.sqlstore.
BunDB().
NewSelect().
Model(&storables).
Where("org_id = ?", orgID).
Scan(ctx)
if err != nil {
return nil, err
}
result := make([]*alertmanagertypes.GettablePlannedMaintenance, 0, len(storables))
for _, s := range storables {
result = append(result, alertmanagertypes.ConvertStorableToGettable(s))
}
return result, nil
}
func (r *maintenance) GetPlannedMaintenanceByID(ctx context.Context, id valuer.UUID) (*alertmanagertypes.GettablePlannedMaintenance, error) {
storable := new(alertmanagertypes.StorablePlannedMaintenance)
err := r.sqlstore.
BunDB().
NewSelect().
Model(storable).
Where("id = ?", id.StringValue()).
Scan(ctx)
if err != nil {
return nil, err
}
return alertmanagertypes.ConvertStorableToGettable(storable), nil
}
func (r *maintenance) CreatePlannedMaintenance(ctx context.Context, maintenance alertmanagertypes.GettablePlannedMaintenance) (valuer.UUID, error) {
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
return valuer.UUID{}, err
}
var ruleIDsStr string
if len(maintenance.RuleIDs) > 0 {
data, err := json.Marshal(maintenance.RuleIDs)
if err != nil {
return valuer.UUID{}, err
}
ruleIDsStr = string(data)
}
storable := alertmanagertypes.StorablePlannedMaintenance{
Identifiable: types.Identifiable{
ID: valuer.GenerateUUID(),
},
TimeAuditable: types.TimeAuditable{
CreatedAt: time.Now(),
UpdatedAt: time.Now(),
},
UserAuditable: types.UserAuditable{
CreatedBy: claims.Email,
UpdatedBy: claims.Email,
},
Name: maintenance.Name,
Description: maintenance.Description,
Schedule: maintenance.Schedule,
RuleIDs: ruleIDsStr,
Expression: maintenance.Expression,
OrgID: claims.OrgID,
}
_, err = r.sqlstore.
BunDB().
NewInsert().
Model(&storable).
Exec(ctx)
if err != nil {
return valuer.UUID{}, err
}
return storable.ID, nil
}
func (r *maintenance) DeletePlannedMaintenance(ctx context.Context, id valuer.UUID) error {
_, err := r.sqlstore.
BunDB().
NewDelete().
Model(new(alertmanagertypes.StorablePlannedMaintenance)).
Where("id = ?", id.StringValue()).
Exec(ctx)
if err != nil {
return err
}
return nil
}
func (r *maintenance) EditPlannedMaintenance(ctx context.Context, maintenance alertmanagertypes.GettablePlannedMaintenance, id valuer.UUID) error {
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
return err
}
var ruleIDsStr string
if len(maintenance.RuleIDs) > 0 {
data, err := json.Marshal(maintenance.RuleIDs)
if err != nil {
return err
}
ruleIDsStr = string(data)
}
storable := alertmanagertypes.StorablePlannedMaintenance{
Identifiable: types.Identifiable{
ID: id,
},
TimeAuditable: types.TimeAuditable{
CreatedAt: maintenance.CreatedAt,
UpdatedAt: time.Now(),
},
UserAuditable: types.UserAuditable{
CreatedBy: maintenance.CreatedBy,
UpdatedBy: claims.Email,
},
Name: maintenance.Name,
Description: maintenance.Description,
Schedule: maintenance.Schedule,
RuleIDs: ruleIDsStr,
Expression: maintenance.Expression,
OrgID: claims.OrgID,
}
_, err = r.sqlstore.
BunDB().
NewUpdate().
Model(&storable).
Where("id = ?", storable.ID.StringValue()).
Exec(ctx)
if err != nil {
return err
}
return nil
}

View File

@@ -1845,3 +1845,216 @@ func (_c *MockAlertmanager_UpdateRoutePolicyByID_Call) RunAndReturn(run func(ctx
_c.Call.Return(run)
return _c
}
// GetAllPlannedMaintenance provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) GetAllPlannedMaintenance(ctx context.Context, orgID string) ([]*alertmanagertypes.GettablePlannedMaintenance, error) {
ret := _mock.Called(ctx, orgID)
if len(ret) == 0 {
panic("no return value specified for GetAllPlannedMaintenance")
}
var r0 []*alertmanagertypes.GettablePlannedMaintenance
var r1 error
if returnFunc, ok := ret.Get(0).(func(context.Context, string) ([]*alertmanagertypes.GettablePlannedMaintenance, error)); ok {
return returnFunc(ctx, orgID)
}
if ret.Get(0) != nil {
r0 = ret.Get(0).([]*alertmanagertypes.GettablePlannedMaintenance)
}
r1 = ret.Error(1)
return r0, r1
}
// GetPlannedMaintenanceByID provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) GetPlannedMaintenanceByID(ctx context.Context, id valuer.UUID) (*alertmanagertypes.GettablePlannedMaintenance, error) {
ret := _mock.Called(ctx, id)
if len(ret) == 0 {
panic("no return value specified for GetPlannedMaintenanceByID")
}
var r0 *alertmanagertypes.GettablePlannedMaintenance
var r1 error
if returnFunc, ok := ret.Get(0).(func(context.Context, valuer.UUID) (*alertmanagertypes.GettablePlannedMaintenance, error)); ok {
return returnFunc(ctx, id)
}
if ret.Get(0) != nil {
r0 = ret.Get(0).(*alertmanagertypes.GettablePlannedMaintenance)
}
r1 = ret.Error(1)
return r0, r1
}
// CreatePlannedMaintenance provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) CreatePlannedMaintenance(ctx context.Context, maintenance alertmanagertypes.GettablePlannedMaintenance) (valuer.UUID, error) {
ret := _mock.Called(ctx, maintenance)
if len(ret) == 0 {
panic("no return value specified for CreatePlannedMaintenance")
}
var r0 valuer.UUID
var r1 error
if returnFunc, ok := ret.Get(0).(func(context.Context, alertmanagertypes.GettablePlannedMaintenance) (valuer.UUID, error)); ok {
return returnFunc(ctx, maintenance)
}
if returnFunc, ok := ret.Get(0).(func(context.Context, alertmanagertypes.GettablePlannedMaintenance) valuer.UUID); ok {
r0 = returnFunc(ctx, maintenance)
} else {
r0 = ret.Get(0).(valuer.UUID)
}
r1 = ret.Error(1)
return r0, r1
}
// EditPlannedMaintenance provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) EditPlannedMaintenance(ctx context.Context, maintenance alertmanagertypes.GettablePlannedMaintenance, id valuer.UUID) error {
ret := _mock.Called(ctx, maintenance, id)
if len(ret) == 0 {
panic("no return value specified for EditPlannedMaintenance")
}
var r0 error
if returnFunc, ok := ret.Get(0).(func(context.Context, alertmanagertypes.GettablePlannedMaintenance, valuer.UUID) error); ok {
r0 = returnFunc(ctx, maintenance, id)
} else {
r0 = ret.Error(0)
}
return r0
}
// DeletePlannedMaintenance provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) DeletePlannedMaintenance(ctx context.Context, id valuer.UUID) error {
ret := _mock.Called(ctx, id)
if len(ret) == 0 {
panic("no return value specified for DeletePlannedMaintenance")
}
var r0 error
if returnFunc, ok := ret.Get(0).(func(context.Context, valuer.UUID) error); ok {
r0 = returnFunc(ctx, id)
} else {
r0 = ret.Error(0)
}
return r0
}
// RecordRuleStateHistory provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) RecordRuleStateHistory(ctx context.Context, orgID string, entries []alertmanagertypes.RuleStateHistory) error {
ret := _mock.Called(ctx, orgID, entries)
if len(ret) == 0 {
panic("no return value specified for RecordRuleStateHistory")
}
var r0 error
if returnFunc, ok := ret.Get(0).(func(context.Context, string, []alertmanagertypes.RuleStateHistory) error); ok {
r0 = returnFunc(ctx, orgID, entries)
} else {
r0 = ret.Error(0)
}
return r0
}
// GetLastSavedRuleStateHistory provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) GetLastSavedRuleStateHistory(ctx context.Context, ruleID string) ([]alertmanagertypes.RuleStateHistory, error) {
ret := _mock.Called(ctx, ruleID)
if len(ret) == 0 {
panic("no return value specified for GetLastSavedRuleStateHistory")
}
var r0 []alertmanagertypes.RuleStateHistory
var r1 error
if returnFunc, ok := ret.Get(0).(func(context.Context, string) ([]alertmanagertypes.RuleStateHistory, error)); ok {
return returnFunc(ctx, ruleID)
}
if ret.Get(0) != nil {
r0 = ret.Get(0).([]alertmanagertypes.RuleStateHistory)
}
r1 = ret.Error(1)
return r0, r1
}
// GetRuleStateHistoryTimeline provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) GetRuleStateHistoryTimeline(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStateTimeline, error) {
ret := _mock.Called(ctx, orgID, ruleID, params)
if len(ret) == 0 {
panic("no return value specified for GetRuleStateHistoryTimeline")
}
var r0 *alertmanagertypes.RuleStateTimeline
var r1 error
if returnFunc, ok := ret.Get(0).(func(context.Context, string, string, *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStateTimeline, error)); ok {
return returnFunc(ctx, orgID, ruleID, params)
}
if ret.Get(0) != nil {
r0 = ret.Get(0).(*alertmanagertypes.RuleStateTimeline)
}
r1 = ret.Error(1)
return r0, r1
}
// GetRuleStateHistoryTopContributors provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) GetRuleStateHistoryTopContributors(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateHistoryContributor, error) {
ret := _mock.Called(ctx, orgID, ruleID, params)
if len(ret) == 0 {
panic("no return value specified for GetRuleStateHistoryTopContributors")
}
var r0 []alertmanagertypes.RuleStateHistoryContributor
var r1 error
if returnFunc, ok := ret.Get(0).(func(context.Context, string, string, *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateHistoryContributor, error)); ok {
return returnFunc(ctx, orgID, ruleID, params)
}
if ret.Get(0) != nil {
r0 = ret.Get(0).([]alertmanagertypes.RuleStateHistoryContributor)
}
r1 = ret.Error(1)
return r0, r1
}
// GetOverallStateTransitions provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) GetOverallStateTransitions(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateTransition, error) {
ret := _mock.Called(ctx, orgID, ruleID, params)
if len(ret) == 0 {
panic("no return value specified for GetOverallStateTransitions")
}
var r0 []alertmanagertypes.RuleStateTransition
var r1 error
if returnFunc, ok := ret.Get(0).(func(context.Context, string, string, *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateTransition, error)); ok {
return returnFunc(ctx, orgID, ruleID, params)
}
if ret.Get(0) != nil {
r0 = ret.Get(0).([]alertmanagertypes.RuleStateTransition)
}
r1 = ret.Error(1)
return r0, r1
}
// GetRuleStats provides a mock function for the type MockAlertmanager
func (_mock *MockAlertmanager) GetRuleStats(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStats, error) {
ret := _mock.Called(ctx, orgID, ruleID, params)
if len(ret) == 0 {
panic("no return value specified for GetRuleStats")
}
var r0 *alertmanagertypes.RuleStats
var r1 error
if returnFunc, ok := ret.Get(0).(func(context.Context, string, string, *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStats, error)); ok {
return returnFunc(ctx, orgID, ruleID, params)
}
if ret.Get(0) != nil {
r0 = ret.Get(0).(*alertmanagertypes.RuleStats)
}
r1 = ret.Error(1)
return r0, r1
}

View File

@@ -5,6 +5,7 @@ import (
"encoding/json"
"io"
"net/http"
"strconv"
"time"
"github.com/SigNoz/signoz/pkg/errors"
@@ -399,3 +400,312 @@ func (api *API) UpdateRoutePolicy(rw http.ResponseWriter, req *http.Request) {
}
render.Success(rw, http.StatusOK, result)
}
func (api *API) ListDowntimeSchedules(rw http.ResponseWriter, req *http.Request) {
ctx, cancel := context.WithTimeout(req.Context(), 30*time.Second)
defer cancel()
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
render.Error(rw, err)
return
}
schedules, err := api.alertmanager.GetAllPlannedMaintenance(ctx, claims.OrgID)
if err != nil {
render.Error(rw, err)
return
}
if q := req.URL.Query().Get("active"); q != "" {
active, _ := strconv.ParseBool(q)
filtered := make([]*alertmanagertypes.GettablePlannedMaintenance, 0)
for _, schedule := range schedules {
now := time.Now().In(time.FixedZone(schedule.Schedule.Timezone, 0))
if schedule.IsActive(now) == active {
filtered = append(filtered, schedule)
}
}
schedules = filtered
}
if q := req.URL.Query().Get("recurring"); q != "" {
recurring, _ := strconv.ParseBool(q)
filtered := make([]*alertmanagertypes.GettablePlannedMaintenance, 0)
for _, schedule := range schedules {
if schedule.IsRecurring() == recurring {
filtered = append(filtered, schedule)
}
}
schedules = filtered
}
render.Success(rw, http.StatusOK, schedules)
}
func (api *API) GetDowntimeSchedule(rw http.ResponseWriter, req *http.Request) {
ctx, cancel := context.WithTimeout(req.Context(), 30*time.Second)
defer cancel()
vars := mux.Vars(req)
idString, ok := vars["id"]
if !ok {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "id is required in path"))
return
}
id, err := valuer.NewUUID(idString)
if err != nil {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "id is not a valid uuid-v7"))
return
}
schedule, err := api.alertmanager.GetPlannedMaintenanceByID(ctx, id)
if err != nil {
render.Error(rw, err)
return
}
render.Success(rw, http.StatusOK, schedule)
}
func (api *API) CreateDowntimeSchedule(rw http.ResponseWriter, req *http.Request) {
ctx, cancel := context.WithTimeout(req.Context(), 30*time.Second)
defer cancel()
body, err := io.ReadAll(req.Body)
if err != nil {
render.Error(rw, err)
return
}
defer req.Body.Close() //nolint:errcheck
var schedule alertmanagertypes.GettablePlannedMaintenance
if err := json.Unmarshal(body, &schedule); err != nil {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "invalid request body: %v", err))
return
}
if err := schedule.Validate(); err != nil {
render.Error(rw, err)
return
}
_, err = api.alertmanager.CreatePlannedMaintenance(ctx, schedule)
if err != nil {
render.Error(rw, err)
return
}
render.Success(rw, http.StatusOK, nil)
}
func (api *API) EditDowntimeSchedule(rw http.ResponseWriter, req *http.Request) {
ctx, cancel := context.WithTimeout(req.Context(), 30*time.Second)
defer cancel()
vars := mux.Vars(req)
idString, ok := vars["id"]
if !ok {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "id is required in path"))
return
}
id, err := valuer.NewUUID(idString)
if err != nil {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "id is not a valid uuid-v7"))
return
}
body, err := io.ReadAll(req.Body)
if err != nil {
render.Error(rw, err)
return
}
defer req.Body.Close() //nolint:errcheck
var schedule alertmanagertypes.GettablePlannedMaintenance
if err := json.Unmarshal(body, &schedule); err != nil {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "invalid request body: %v", err))
return
}
if err := schedule.Validate(); err != nil {
render.Error(rw, err)
return
}
err = api.alertmanager.EditPlannedMaintenance(ctx, schedule, id)
if err != nil {
render.Error(rw, err)
return
}
render.Success(rw, http.StatusOK, nil)
}
func (api *API) DeleteDowntimeSchedule(rw http.ResponseWriter, req *http.Request) {
ctx, cancel := context.WithTimeout(req.Context(), 30*time.Second)
defer cancel()
vars := mux.Vars(req)
idString, ok := vars["id"]
if !ok {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "id is required in path"))
return
}
id, err := valuer.NewUUID(idString)
if err != nil {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "id is not a valid uuid-v7"))
return
}
err = api.alertmanager.DeletePlannedMaintenance(ctx, id)
if err != nil {
render.Error(rw, err)
return
}
render.Success(rw, http.StatusNoContent, nil)
}
func (api *API) GetRuleStateHistoryTimeline(rw http.ResponseWriter, req *http.Request) {
ctx, cancel := context.WithTimeout(req.Context(), 30*time.Second)
defer cancel()
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
render.Error(rw, err)
return
}
ruleID := mux.Vars(req)["id"]
if ruleID == "" {
render.Error(rw, errors.NewInvalidInputf(errors.CodeInvalidInput, "rule ID is required"))
return
}
var params alertmanagertypes.QueryRuleStateHistory
if err := json.NewDecoder(req.Body).Decode(&params); err != nil {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "invalid request body: %v", err))
return
}
if err := params.Validate(); err != nil {
render.Error(rw, err)
return
}
result, err := api.alertmanager.GetRuleStateHistoryTimeline(ctx, claims.OrgID, ruleID, &params)
if err != nil {
render.Error(rw, err)
return
}
render.Success(rw, http.StatusOK, result)
}
func (api *API) GetRuleStats(rw http.ResponseWriter, req *http.Request) {
ctx, cancel := context.WithTimeout(req.Context(), 30*time.Second)
defer cancel()
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
render.Error(rw, err)
return
}
ruleID := mux.Vars(req)["id"]
if ruleID == "" {
render.Error(rw, errors.NewInvalidInputf(errors.CodeInvalidInput, "rule ID is required"))
return
}
var params alertmanagertypes.QueryRuleStateHistory
if err := json.NewDecoder(req.Body).Decode(&params); err != nil {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "invalid request body: %v", err))
return
}
if err := params.Validate(); err != nil {
render.Error(rw, err)
return
}
result, err := api.alertmanager.GetRuleStats(ctx, claims.OrgID, ruleID, &params)
if err != nil {
render.Error(rw, err)
return
}
render.Success(rw, http.StatusOK, result)
}
func (api *API) GetRuleStateHistoryTopContributors(rw http.ResponseWriter, req *http.Request) {
ctx, cancel := context.WithTimeout(req.Context(), 30*time.Second)
defer cancel()
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
render.Error(rw, err)
return
}
ruleID := mux.Vars(req)["id"]
if ruleID == "" {
render.Error(rw, errors.NewInvalidInputf(errors.CodeInvalidInput, "rule ID is required"))
return
}
var params alertmanagertypes.QueryRuleStateHistory
if err := json.NewDecoder(req.Body).Decode(&params); err != nil {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "invalid request body: %v", err))
return
}
if err := params.Validate(); err != nil {
render.Error(rw, err)
return
}
result, err := api.alertmanager.GetRuleStateHistoryTopContributors(ctx, claims.OrgID, ruleID, &params)
if err != nil {
render.Error(rw, err)
return
}
render.Success(rw, http.StatusOK, result)
}
func (api *API) GetOverallStateTransitions(rw http.ResponseWriter, req *http.Request) {
ctx, cancel := context.WithTimeout(req.Context(), 30*time.Second)
defer cancel()
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
render.Error(rw, err)
return
}
ruleID := mux.Vars(req)["id"]
if ruleID == "" {
render.Error(rw, errors.NewInvalidInputf(errors.CodeInvalidInput, "rule ID is required"))
return
}
var params alertmanagertypes.QueryRuleStateHistory
if err := json.NewDecoder(req.Body).Decode(&params); err != nil {
render.Error(rw, errors.Newf(errors.TypeInvalidInput, errors.CodeInvalidInput, "invalid request body: %v", err))
return
}
if err := params.Validate(); err != nil {
render.Error(rw, err)
return
}
result, err := api.alertmanager.GetOverallStateTransitions(ctx, claims.OrgID, ruleID, &params)
if err != nil {
render.Error(rw, err)
return
}
render.Success(rw, http.StatusOK, result)
}

1123
pkg/alertmanager/api_test.go Normal file

File diff suppressed because it is too large Load Diff

View File

@@ -2,10 +2,15 @@ package alertmanager
import (
"context"
"encoding/json"
"math"
"sync"
"time"
"github.com/prometheus/alertmanager/featurecontrol"
"github.com/prometheus/alertmanager/matcher/compat"
amtypes "github.com/prometheus/alertmanager/types"
"github.com/prometheus/common/model"
"github.com/SigNoz/signoz/pkg/alertmanager/alertmanagerserver"
"github.com/SigNoz/signoz/pkg/alertmanager/nfmanager"
@@ -38,6 +43,15 @@ type Service struct {
serversMtx sync.RWMutex
notificationManager nfmanager.NotificationManager
// maintenanceExprMuter is an optional muter for expression-based maintenance scoping
maintenanceExprMuter amtypes.Muter
// stateHistoryStore writes rule state history to persistent storage (e.g. ClickHouse)
stateHistoryStore alertmanagertypes.StateHistoryStore
// stateTracker tracks alert state transitions for v2 state history recording
stateTracker *stateTracker
}
func New(
@@ -48,16 +62,21 @@ func New(
configStore alertmanagertypes.ConfigStore,
orgGetter organization.Getter,
nfManager nfmanager.NotificationManager,
maintenanceExprMuter amtypes.Muter,
stateHistoryStore alertmanagertypes.StateHistoryStore,
) *Service {
service := &Service{
config: config,
stateStore: stateStore,
configStore: configStore,
orgGetter: orgGetter,
settings: settings,
servers: make(map[string]*alertmanagerserver.Server),
serversMtx: sync.RWMutex{},
notificationManager: nfManager,
config: config,
stateStore: stateStore,
configStore: configStore,
orgGetter: orgGetter,
settings: settings,
servers: make(map[string]*alertmanagerserver.Server),
serversMtx: sync.RWMutex{},
notificationManager: nfManager,
maintenanceExprMuter: maintenanceExprMuter,
stateHistoryStore: stateHistoryStore,
stateTracker: newStateTracker(),
}
return service
@@ -131,7 +150,21 @@ func (service *Service) PutAlerts(ctx context.Context, orgID string, alerts aler
return err
}
return server.PutAlerts(ctx, alerts)
// Convert to typed alerts for state tracking (same conversion the server does).
now := time.Now()
typedAlerts, _ := alertmanagertypes.NewAlertsFromPostableAlerts(
alerts, time.Duration(service.config.Global.ResolveTimeout), now,
)
// Delegate to server for notification pipeline.
if err := server.PutAlerts(ctx, alerts); err != nil {
return err
}
// Record state history from the incoming alerts.
service.recordStateHistoryFromAlerts(ctx, orgID, typedAlerts, now)
return nil
}
func (service *Service) TestReceiver(ctx context.Context, orgID string, receiver alertmanagertypes.Receiver) error {
@@ -176,7 +209,7 @@ func (service *Service) newServer(ctx context.Context, orgID string) (*alertmana
return nil, err
}
server, err := alertmanagerserver.New(ctx, service.settings.Logger(), service.settings.PrometheusRegisterer(), service.config, orgID, service.stateStore, service.notificationManager)
server, err := alertmanagerserver.New(ctx, service.settings.Logger(), service.settings.PrometheusRegisterer(), service.config, orgID, service.stateStore, service.notificationManager, service.maintenanceExprMuter)
if err != nil {
return nil, err
}
@@ -255,6 +288,205 @@ func (service *Service) compareAndSelectConfig(ctx context.Context, incomingConf
}
// RecordRuleStateHistory applies maintenance muting logic and writes state history entries.
// For each entry with State=="firing", if the maintenance muter matches the entry's labels,
// the state is changed to "muted" before writing.
func (service *Service) RecordRuleStateHistory(ctx context.Context, orgID string, entries []alertmanagertypes.RuleStateHistory) error {
if service.stateHistoryStore == nil {
return nil
}
for i := range entries {
entries[i].OrgID = orgID
}
if service.maintenanceExprMuter != nil {
for i := range entries {
if entries[i].State != "firing" {
continue
}
lbls := labelsFromJSON(entries[i].Labels)
if lbls == nil {
continue
}
// Add ruleId to the label set for muter matching.
lbls["ruleId"] = model.LabelValue(entries[i].RuleID)
if service.maintenanceExprMuter.Mutes(lbls) {
entries[i].State = "muted"
}
}
}
return service.stateHistoryStore.WriteRuleStateHistory(ctx, entries)
}
func (service *Service) GetLastSavedRuleStateHistory(ctx context.Context, ruleID string) ([]alertmanagertypes.RuleStateHistory, error) {
if service.stateHistoryStore == nil {
return nil, nil
}
return service.stateHistoryStore.GetLastSavedRuleStateHistory(ctx, ruleID)
}
func (service *Service) GetRuleStateHistoryTimeline(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStateTimeline, error) {
if service.stateHistoryStore == nil {
return &alertmanagertypes.RuleStateTimeline{Items: []alertmanagertypes.RuleStateHistory{}}, nil
}
return service.stateHistoryStore.GetRuleStateHistoryTimeline(ctx, orgID, ruleID, params)
}
func (service *Service) GetRuleStateHistoryTopContributors(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateHistoryContributor, error) {
if service.stateHistoryStore == nil {
return []alertmanagertypes.RuleStateHistoryContributor{}, nil
}
return service.stateHistoryStore.GetRuleStateHistoryTopContributors(ctx, orgID, ruleID, params)
}
func (service *Service) GetOverallStateTransitions(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateTransition, error) {
if service.stateHistoryStore == nil {
return []alertmanagertypes.RuleStateTransition{}, nil
}
return service.stateHistoryStore.GetOverallStateTransitions(ctx, orgID, ruleID, params)
}
func (service *Service) GetRuleStats(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStats, error) {
if service.stateHistoryStore == nil {
return &alertmanagertypes.RuleStats{}, nil
}
store := service.stateHistoryStore
// Current period stats.
totalCurrentTriggers, err := store.GetTotalTriggers(ctx, orgID, ruleID, params)
if err != nil {
return nil, err
}
currentTriggersSeries, err := store.GetTriggersByInterval(ctx, orgID, ruleID, params)
if err != nil {
return nil, err
}
currentAvgResolutionTime, err := store.GetAvgResolutionTime(ctx, orgID, ruleID, params)
if err != nil {
return nil, err
}
currentAvgResolutionTimeSeries, err := store.GetAvgResolutionTimeByInterval(ctx, orgID, ruleID, params)
if err != nil {
return nil, err
}
// Past period stats — shift time window backward.
pastParams := *params
duration := params.End - params.Start
if duration >= 86400000 {
days := int64(math.Ceil(float64(duration) / 86400000))
pastParams.Start -= days * 86400000
pastParams.End -= days * 86400000
} else {
pastParams.Start -= 86400000
pastParams.End -= 86400000
}
totalPastTriggers, err := store.GetTotalTriggers(ctx, orgID, ruleID, &pastParams)
if err != nil {
return nil, err
}
pastTriggersSeries, err := store.GetTriggersByInterval(ctx, orgID, ruleID, &pastParams)
if err != nil {
return nil, err
}
pastAvgResolutionTime, err := store.GetAvgResolutionTime(ctx, orgID, ruleID, &pastParams)
if err != nil {
return nil, err
}
pastAvgResolutionTimeSeries, err := store.GetAvgResolutionTimeByInterval(ctx, orgID, ruleID, &pastParams)
if err != nil {
return nil, err
}
if math.IsNaN(currentAvgResolutionTime) || math.IsInf(currentAvgResolutionTime, 0) {
currentAvgResolutionTime = 0
}
if math.IsNaN(pastAvgResolutionTime) || math.IsInf(pastAvgResolutionTime, 0) {
pastAvgResolutionTime = 0
}
return &alertmanagertypes.RuleStats{
TotalCurrentTriggers: totalCurrentTriggers,
TotalPastTriggers: totalPastTriggers,
CurrentTriggersSeries: currentTriggersSeries,
PastTriggersSeries: pastTriggersSeries,
CurrentAvgResolutionTime: currentAvgResolutionTime,
PastAvgResolutionTime: pastAvgResolutionTime,
CurrentAvgResolutionTimeSeries: currentAvgResolutionTimeSeries,
PastAvgResolutionTimeSeries: pastAvgResolutionTimeSeries,
}, nil
}
// recordStateHistoryFromAlerts detects state transitions from incoming alerts
// and records them via RecordRuleStateHistory (which applies maintenance muting).
func (service *Service) recordStateHistoryFromAlerts(ctx context.Context, orgID string, alerts []*amtypes.Alert, now time.Time) {
if service.stateHistoryStore == nil {
return
}
entries := service.stateTracker.processAlerts(orgID, alerts, now)
if len(entries) == 0 {
return
}
if err := service.RecordRuleStateHistory(ctx, orgID, entries); err != nil {
service.settings.Logger().ErrorContext(ctx, "failed to record state history", "error", err)
}
}
// StartStateHistorySweep starts a background goroutine that periodically checks
// for stale firing alerts and records them as resolved. Call this once after creating the service.
func (service *Service) StartStateHistorySweep(ctx context.Context) {
if service.stateHistoryStore == nil {
return
}
staleTimeout := 2 * time.Duration(service.config.Global.ResolveTimeout)
if staleTimeout == 0 {
staleTimeout = 10 * time.Minute
}
go func() {
ticker := time.NewTicker(1 * time.Minute)
defer ticker.Stop()
for {
select {
case <-ctx.Done():
return
case <-ticker.C:
now := time.Now()
entriesByOrg := service.stateTracker.sweepStale(staleTimeout, now)
for orgID, orgEntries := range entriesByOrg {
if err := service.RecordRuleStateHistory(ctx, orgID, orgEntries); err != nil {
service.settings.Logger().ErrorContext(ctx, "failed to record stale state history", "org_id", orgID, "error", err)
}
}
}
}
}()
}
// labelsFromJSON parses a JSON string of labels into a model.LabelSet.
func labelsFromJSON(labelsJSON string) model.LabelSet {
if labelsJSON == "" {
return nil
}
var m map[string]string
if err := json.Unmarshal([]byte(labelsJSON), &m); err != nil {
return nil
}
ls := make(model.LabelSet, len(m))
for k, v := range m {
ls[model.LabelName(k)] = model.LabelValue(v)
}
return ls
}
// getServer returns the server for the given orgID. It should be called with the lock held.
func (service *Service) getServer(orgID string) (*alertmanagerserver.Server, error) {
server, ok := service.servers[orgID]

View File

@@ -0,0 +1,426 @@
package alertmanager
import (
"context"
"math"
"sync"
"testing"
"github.com/SigNoz/signoz/pkg/types/alertmanagertypes"
"github.com/prometheus/common/model"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
// testMuter implements amtypes.Muter for testing.
type testMuter struct {
mu sync.Mutex
muteFunc func(model.LabelSet) bool
calls []model.LabelSet
}
func (m *testMuter) Mutes(labels model.LabelSet) bool {
m.mu.Lock()
defer m.mu.Unlock()
m.calls = append(m.calls, labels)
if m.muteFunc != nil {
return m.muteFunc(labels)
}
return false
}
func (m *testMuter) getCalls() []model.LabelSet {
m.mu.Lock()
defer m.mu.Unlock()
result := make([]model.LabelSet, len(m.calls))
copy(result, m.calls)
return result
}
// fakeStateHistoryStore captures calls for assertion.
type fakeStateHistoryStore struct {
written []alertmanagertypes.RuleStateHistory
lastErr error
getResult []alertmanagertypes.RuleStateHistory
getErr error
// Stats method returns
totalTriggers uint64
totalTriggersErr error
triggersSeries *alertmanagertypes.Series
triggersSeriesErr error
avgResolutionTime float64
avgResolutionTimeErr error
avgResTimeSeries *alertmanagertypes.Series
avgResTimeSeriesErr error
// Captures params passed to stats methods
statsCalls []*alertmanagertypes.QueryRuleStateHistory
}
func (w *fakeStateHistoryStore) WriteRuleStateHistory(_ context.Context, entries []alertmanagertypes.RuleStateHistory) error {
w.written = append(w.written, entries...)
return w.lastErr
}
func (w *fakeStateHistoryStore) GetLastSavedRuleStateHistory(_ context.Context, _ string) ([]alertmanagertypes.RuleStateHistory, error) {
return w.getResult, w.getErr
}
func (w *fakeStateHistoryStore) GetRuleStateHistoryTimeline(_ context.Context, _ string, _ string, _ *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStateTimeline, error) {
return nil, nil
}
func (w *fakeStateHistoryStore) GetRuleStateHistoryTopContributors(_ context.Context, _ string, _ string, _ *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateHistoryContributor, error) {
return nil, nil
}
func (w *fakeStateHistoryStore) GetOverallStateTransitions(_ context.Context, _ string, _ string, _ *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateTransition, error) {
return nil, nil
}
func (w *fakeStateHistoryStore) GetTotalTriggers(_ context.Context, _ string, _ string, params *alertmanagertypes.QueryRuleStateHistory) (uint64, error) {
w.statsCalls = append(w.statsCalls, params)
return w.totalTriggers, w.totalTriggersErr
}
func (w *fakeStateHistoryStore) GetTriggersByInterval(_ context.Context, _ string, _ string, params *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.Series, error) {
w.statsCalls = append(w.statsCalls, params)
return w.triggersSeries, w.triggersSeriesErr
}
func (w *fakeStateHistoryStore) GetAvgResolutionTime(_ context.Context, _ string, _ string, params *alertmanagertypes.QueryRuleStateHistory) (float64, error) {
w.statsCalls = append(w.statsCalls, params)
return w.avgResolutionTime, w.avgResolutionTimeErr
}
func (w *fakeStateHistoryStore) GetAvgResolutionTimeByInterval(_ context.Context, _ string, _ string, params *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.Series, error) {
w.statsCalls = append(w.statsCalls, params)
return w.avgResTimeSeries, w.avgResTimeSeriesErr
}
func TestLabelsFromJSON(t *testing.T) {
tests := []struct {
name string
input string
want model.LabelSet
}{
{
name: "empty string",
input: "",
want: nil,
},
{
name: "invalid json",
input: "not json",
want: nil,
},
{
name: "valid labels",
input: `{"env":"prod","severity":"critical"}`,
want: model.LabelSet{
"env": "prod",
"severity": "critical",
},
},
{
name: "empty object",
input: `{}`,
want: model.LabelSet{},
},
{
name: "single label",
input: `{"alertname":"HighCPU"}`,
want: model.LabelSet{"alertname": "HighCPU"},
},
}
for _, tc := range tests {
t.Run(tc.name, func(t *testing.T) {
got := labelsFromJSON(tc.input)
assert.Equal(t, tc.want, got)
})
}
}
func TestRecordRuleStateHistory(t *testing.T) {
ctx := context.Background()
t.Run("nil writer returns nil", func(t *testing.T) {
svc := &Service{stateHistoryStore: nil}
err := svc.RecordRuleStateHistory(ctx, "org-1", []alertmanagertypes.RuleStateHistory{
{RuleID: "r1", State: "firing"},
})
require.NoError(t, err)
})
t.Run("no muter writes entries unchanged", func(t *testing.T) {
writer := &fakeStateHistoryStore{}
svc := &Service{
stateHistoryStore: writer,
maintenanceExprMuter: nil,
}
entries := []alertmanagertypes.RuleStateHistory{
{RuleID: "r1", State: "firing", Labels: `{"env":"prod"}`},
{RuleID: "r2", State: "normal", Labels: `{"env":"staging"}`},
}
err := svc.RecordRuleStateHistory(ctx, "org-1", entries)
require.NoError(t, err)
require.Len(t, writer.written, 2)
assert.Equal(t, "firing", writer.written[0].State)
assert.Equal(t, "normal", writer.written[1].State)
})
t.Run("muter changes firing to muted when matched", func(t *testing.T) {
writer := &fakeStateHistoryStore{}
muter := &testMuter{
muteFunc: func(ls model.LabelSet) bool {
return ls["env"] == "prod"
},
}
svc := &Service{
stateHistoryStore: writer,
maintenanceExprMuter: muter,
}
entries := []alertmanagertypes.RuleStateHistory{
{RuleID: "r1", State: "firing", Labels: `{"env":"prod"}`},
}
err := svc.RecordRuleStateHistory(ctx, "org-1", entries)
require.NoError(t, err)
require.Len(t, writer.written, 1)
assert.Equal(t, "muted", writer.written[0].State)
})
t.Run("muter does not change firing when not matched", func(t *testing.T) {
writer := &fakeStateHistoryStore{}
muter := &testMuter{
muteFunc: func(ls model.LabelSet) bool {
return ls["env"] == "prod"
},
}
svc := &Service{
stateHistoryStore: writer,
maintenanceExprMuter: muter,
}
entries := []alertmanagertypes.RuleStateHistory{
{RuleID: "r1", State: "firing", Labels: `{"env":"staging"}`},
}
err := svc.RecordRuleStateHistory(ctx, "org-1", entries)
require.NoError(t, err)
require.Len(t, writer.written, 1)
assert.Equal(t, "firing", writer.written[0].State)
})
t.Run("muter only affects firing entries", func(t *testing.T) {
writer := &fakeStateHistoryStore{}
muter := &testMuter{
muteFunc: func(model.LabelSet) bool { return true }, // mute everything
}
svc := &Service{
stateHistoryStore: writer,
maintenanceExprMuter: muter,
}
entries := []alertmanagertypes.RuleStateHistory{
{RuleID: "r1", State: "normal", Labels: `{"env":"prod"}`},
{RuleID: "r2", State: "no_data", Labels: `{"env":"prod"}`},
{RuleID: "r3", State: "firing", Labels: `{"env":"prod"}`},
}
err := svc.RecordRuleStateHistory(ctx, "org-1", entries)
require.NoError(t, err)
require.Len(t, writer.written, 3)
assert.Equal(t, "normal", writer.written[0].State, "normal should not be muted")
assert.Equal(t, "no_data", writer.written[1].State, "no_data should not be muted")
assert.Equal(t, "muted", writer.written[2].State, "firing should become muted")
})
t.Run("ruleId is injected into labels for muter evaluation", func(t *testing.T) {
writer := &fakeStateHistoryStore{}
muter := &testMuter{
muteFunc: func(ls model.LabelSet) bool {
return ls["ruleId"] == "target-rule"
},
}
svc := &Service{
stateHistoryStore: writer,
maintenanceExprMuter: muter,
}
entries := []alertmanagertypes.RuleStateHistory{
{RuleID: "target-rule", State: "firing", Labels: `{"env":"prod"}`},
{RuleID: "other-rule", State: "firing", Labels: `{"env":"prod"}`},
}
err := svc.RecordRuleStateHistory(ctx, "org-1", entries)
require.NoError(t, err)
require.Len(t, writer.written, 2)
assert.Equal(t, "muted", writer.written[0].State, "target-rule should be muted")
assert.Equal(t, "firing", writer.written[1].State, "other-rule should not be muted")
// Verify the muter received labels with ruleId injected
calls := muter.getCalls()
require.Len(t, calls, 2)
assert.Equal(t, model.LabelValue("target-rule"), calls[0]["ruleId"])
assert.Equal(t, model.LabelValue("other-rule"), calls[1]["ruleId"])
})
t.Run("invalid labels JSON skips muting check", func(t *testing.T) {
writer := &fakeStateHistoryStore{}
muter := &testMuter{
muteFunc: func(model.LabelSet) bool { return true },
}
svc := &Service{
stateHistoryStore: writer,
maintenanceExprMuter: muter,
}
entries := []alertmanagertypes.RuleStateHistory{
{RuleID: "r1", State: "firing", Labels: "not-json"},
{RuleID: "r2", State: "firing", Labels: ""},
}
err := svc.RecordRuleStateHistory(ctx, "org-1", entries)
require.NoError(t, err)
require.Len(t, writer.written, 2)
// Both should stay firing because labels couldn't be parsed
assert.Equal(t, "firing", writer.written[0].State)
assert.Equal(t, "firing", writer.written[1].State)
// Muter should not have been called
assert.Empty(t, muter.getCalls())
})
t.Run("mixed entries with selective muting", func(t *testing.T) {
writer := &fakeStateHistoryStore{}
muter := &testMuter{
muteFunc: func(ls model.LabelSet) bool {
return ls["severity"] == "warning"
},
}
svc := &Service{
stateHistoryStore: writer,
maintenanceExprMuter: muter,
}
entries := []alertmanagertypes.RuleStateHistory{
{RuleID: "r1", State: "firing", Labels: `{"severity":"critical"}`, Fingerprint: 1},
{RuleID: "r2", State: "firing", Labels: `{"severity":"warning"}`, Fingerprint: 2},
{RuleID: "r3", State: "normal", Labels: `{"severity":"warning"}`, Fingerprint: 3},
{RuleID: "r4", State: "firing", Labels: `{"severity":"warning"}`, Fingerprint: 4},
}
err := svc.RecordRuleStateHistory(ctx, "org-1", entries)
require.NoError(t, err)
require.Len(t, writer.written, 4)
assert.Equal(t, "firing", writer.written[0].State, "critical firing stays firing")
assert.Equal(t, "muted", writer.written[1].State, "warning firing becomes muted")
assert.Equal(t, "normal", writer.written[2].State, "normal is never muted")
assert.Equal(t, "muted", writer.written[3].State, "warning firing becomes muted")
})
}
func TestGetLastSavedRuleStateHistory(t *testing.T) {
ctx := context.Background()
t.Run("nil writer returns nil", func(t *testing.T) {
svc := &Service{stateHistoryStore: nil}
result, err := svc.GetLastSavedRuleStateHistory(ctx, "r1")
require.NoError(t, err)
assert.Nil(t, result)
})
t.Run("delegates to writer", func(t *testing.T) {
expected := []alertmanagertypes.RuleStateHistory{
{RuleID: "r1", State: "firing", Fingerprint: 123},
}
writer := &fakeStateHistoryStore{getResult: expected}
svc := &Service{stateHistoryStore: writer}
result, err := svc.GetLastSavedRuleStateHistory(ctx, "r1")
require.NoError(t, err)
assert.Equal(t, expected, result)
})
}
func TestGetRuleStats(t *testing.T) {
ctx := context.Background()
t.Run("aggregates current and past period stats", func(t *testing.T) {
currentSeries := &alertmanagertypes.Series{Points: []alertmanagertypes.Point{{Timestamp: 1000, Value: 5}}}
currentResSeries := &alertmanagertypes.Series{Points: []alertmanagertypes.Point{{Timestamp: 1000, Value: 120}}}
store := &fakeStateHistoryStore{
totalTriggers: 10,
triggersSeries: currentSeries,
avgResolutionTime: 300.5,
avgResTimeSeries: currentResSeries,
}
svc := &Service{stateHistoryStore: store}
params := &alertmanagertypes.QueryRuleStateHistory{
Start: 1000,
End: 90000000, // ~1 day
}
result, err := svc.GetRuleStats(ctx, "org-1", "rule-1", params)
require.NoError(t, err)
assert.Equal(t, uint64(10), result.TotalCurrentTriggers)
assert.Equal(t, uint64(10), result.TotalPastTriggers)
assert.Equal(t, currentSeries, result.CurrentTriggersSeries)
assert.Equal(t, 300.5, result.CurrentAvgResolutionTime)
})
t.Run("period shifting for duration >= 1 day", func(t *testing.T) {
store := &fakeStateHistoryStore{}
svc := &Service{stateHistoryStore: store}
// 2-day window: Start=0, End=172800000 (2 days in millis)
params := &alertmanagertypes.QueryRuleStateHistory{
Start: 0,
End: 172800000,
}
_, err := svc.GetRuleStats(ctx, "org-1", "rule-1", params)
require.NoError(t, err)
// First 4 calls are current period, next 4 are past period.
// For 2 days: ceil(172800000/86400000) = 2, shift = 2*86400000 = 172800000
require.GreaterOrEqual(t, len(store.statsCalls), 8)
pastParams := store.statsCalls[4] // first past period call
assert.Equal(t, int64(-172800000), pastParams.Start)
assert.Equal(t, int64(0), pastParams.End)
})
t.Run("period shifting for duration < 1 day", func(t *testing.T) {
store := &fakeStateHistoryStore{}
svc := &Service{stateHistoryStore: store}
// 1-hour window
params := &alertmanagertypes.QueryRuleStateHistory{
Start: 100000000,
End: 103600000, // 3600000ms = 1 hour
}
_, err := svc.GetRuleStats(ctx, "org-1", "rule-1", params)
require.NoError(t, err)
// For < 1 day: shift by exactly 1 day (86400000ms)
require.GreaterOrEqual(t, len(store.statsCalls), 8)
pastParams := store.statsCalls[4]
assert.Equal(t, int64(100000000-86400000), pastParams.Start)
assert.Equal(t, int64(103600000-86400000), pastParams.End)
})
t.Run("NaN and Inf avg resolution times are zeroed", func(t *testing.T) {
for _, val := range []float64{math.NaN(), math.Inf(1), math.Inf(-1)} {
store := &fakeStateHistoryStore{
avgResolutionTime: val,
}
svc := &Service{stateHistoryStore: store}
result, err := svc.GetRuleStats(ctx, "org-1", "rule-1", &alertmanagertypes.QueryRuleStateHistory{
Start: 0, End: 100000000,
})
require.NoError(t, err)
assert.Equal(t, float64(0), result.CurrentAvgResolutionTime)
assert.Equal(t, float64(0), result.PastAvgResolutionTime)
}
})
}

View File

@@ -0,0 +1,150 @@
package signozalertmanager
import (
"log/slog"
"strings"
"sync"
"github.com/expr-lang/expr"
"github.com/prometheus/common/model"
)
// convertLabelSetToEnv converts a flat label set with dotted keys into a nested map
// structure for expr-lang evaluation. When both a leaf and a deeper nested path exist
// (e.g. "foo" and "foo.bar"), the nested structure takes precedence.
func convertLabelSetToEnv(labelSet model.LabelSet) map[string]interface{} {
env := make(map[string]interface{})
for lk, lv := range labelSet {
key := strings.TrimSpace(string(lk))
value := string(lv)
if strings.Contains(key, ".") {
parts := strings.Split(key, ".")
current := env
for i, raw := range parts {
part := strings.TrimSpace(raw)
last := i == len(parts)-1
if last {
if _, isMap := current[part].(map[string]interface{}); isMap {
break
}
current[part] = value
break
}
if nextMap, ok := current[part].(map[string]interface{}); ok {
current = nextMap
continue
}
newMap := make(map[string]interface{})
current[part] = newMap
current = newMap
}
continue
}
if _, isMap := env[key].(map[string]interface{}); isMap {
continue
}
env[key] = value
}
return env
}
// evaluateExpr compiles and runs an expr-lang expression against the given label set.
func evaluateExpr(expression string, labelSet model.LabelSet) (bool, error) {
env := convertLabelSetToEnv(labelSet)
program, err := expr.Compile(expression, expr.Env(env))
if err != nil {
return false, err
}
output, err := expr.Run(program, env)
if err != nil {
return false, err
}
if boolVal, ok := output.(bool); ok {
return boolVal, nil
}
return false, nil
}
// activeMaintenanceExpr holds an active maintenance's scoping criteria.
// Muting logic: (ruleIDs match OR ruleIDs empty) AND (expression match OR expression empty).
type activeMaintenanceExpr struct {
ruleIDs []string
expression string
}
// MaintenanceExprMuter implements types.Muter for expression-based maintenance scoping.
// It evaluates expr-lang expressions against alert labels to determine if an alert
// should be muted (suppressed) during a maintenance window.
type MaintenanceExprMuter struct {
mu sync.RWMutex
expressions []activeMaintenanceExpr
logger *slog.Logger
}
// NewMaintenanceExprMuter creates a new MaintenanceExprMuter.
func NewMaintenanceExprMuter(logger *slog.Logger) *MaintenanceExprMuter {
return &MaintenanceExprMuter{
logger: logger,
}
}
// Mutes returns true if the given label set matches any active maintenance entry.
// Each entry uses AND logic: (ruleIDs match OR empty) AND (expression match OR empty).
// Empty ruleIDs means all rules are in scope. Empty expression means all labels match.
func (m *MaintenanceExprMuter) Mutes(labels model.LabelSet) bool {
m.mu.RLock()
defer m.mu.RUnlock()
for _, ae := range m.expressions {
// Check rule scope: empty ruleIDs means all rules match.
ruleMatch := len(ae.ruleIDs) == 0
if !ruleMatch {
alertRuleID := string(labels["ruleId"])
for _, rid := range ae.ruleIDs {
if rid == alertRuleID {
ruleMatch = true
break
}
}
}
if !ruleMatch {
continue
}
// Check expression scope: empty expression means all labels match.
if ae.expression == "" {
return true
}
matched, err := evaluateExpr(ae.expression, labels)
if err != nil {
m.logger.Error("failed to evaluate maintenance expression",
"expression", ae.expression,
"error", err)
continue
}
if matched {
return true
}
}
return false
}
// SetActiveExpressions updates the list of active maintenance expressions.
func (m *MaintenanceExprMuter) SetActiveExpressions(exprs []activeMaintenanceExpr) {
m.mu.Lock()
defer m.mu.Unlock()
m.expressions = exprs
}

View File

@@ -10,12 +10,14 @@ import (
amConfig "github.com/prometheus/alertmanager/config"
"github.com/SigNoz/signoz/pkg/alertmanager"
"github.com/SigNoz/signoz/pkg/alertmanager/alertmanagerstore/clickhousealertmanagerstore"
"github.com/SigNoz/signoz/pkg/alertmanager/alertmanagerstore/sqlalertmanagerstore"
"github.com/SigNoz/signoz/pkg/alertmanager/nfmanager"
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/factory"
"github.com/SigNoz/signoz/pkg/modules/organization"
"github.com/SigNoz/signoz/pkg/sqlstore"
"github.com/SigNoz/signoz/pkg/telemetrystore"
"github.com/SigNoz/signoz/pkg/types"
"github.com/SigNoz/signoz/pkg/types/alertmanagertypes"
"github.com/SigNoz/signoz/pkg/types/authtypes"
@@ -23,25 +25,34 @@ import (
)
type provider struct {
service *alertmanager.Service
config alertmanager.Config
settings factory.ScopedProviderSettings
configStore alertmanagertypes.ConfigStore
stateStore alertmanagertypes.StateStore
notificationManager nfmanager.NotificationManager
stopC chan struct{}
service *alertmanager.Service
config alertmanager.Config
settings factory.ScopedProviderSettings
configStore alertmanagertypes.ConfigStore
stateStore alertmanagertypes.StateStore
notificationManager nfmanager.NotificationManager
maintenanceStore alertmanagertypes.MaintenanceStore
maintenanceExprMuter *MaintenanceExprMuter
orgGetter organization.Getter
stopC chan struct{}
}
func NewFactory(sqlstore sqlstore.SQLStore, orgGetter organization.Getter, notificationManager nfmanager.NotificationManager) factory.ProviderFactory[alertmanager.Alertmanager, alertmanager.Config] {
func NewFactory(sqlstore sqlstore.SQLStore, orgGetter organization.Getter, notificationManager nfmanager.NotificationManager, telemetryStore telemetrystore.TelemetryStore) factory.ProviderFactory[alertmanager.Alertmanager, alertmanager.Config] {
return factory.NewProviderFactory(factory.MustNewName("signoz"), func(ctx context.Context, settings factory.ProviderSettings, config alertmanager.Config) (alertmanager.Alertmanager, error) {
return New(ctx, settings, config, sqlstore, orgGetter, notificationManager)
return New(ctx, settings, config, sqlstore, orgGetter, notificationManager, telemetryStore)
})
}
func New(ctx context.Context, providerSettings factory.ProviderSettings, config alertmanager.Config, sqlstore sqlstore.SQLStore, orgGetter organization.Getter, notificationManager nfmanager.NotificationManager) (*provider, error) {
func New(ctx context.Context, providerSettings factory.ProviderSettings, config alertmanager.Config, sqlstore sqlstore.SQLStore, orgGetter organization.Getter, notificationManager nfmanager.NotificationManager, telemetryStore telemetrystore.TelemetryStore) (*provider, error) {
settings := factory.NewScopedProviderSettings(providerSettings, "github.com/SigNoz/signoz/pkg/alertmanager/signozalertmanager")
configStore := sqlalertmanagerstore.NewConfigStore(sqlstore)
stateStore := sqlalertmanagerstore.NewStateStore(sqlstore)
maintenanceExprMuter := NewMaintenanceExprMuter(settings.Logger())
var stateHistoryStore alertmanagertypes.StateHistoryStore
if telemetryStore != nil {
stateHistoryStore = clickhousealertmanagerstore.NewStateHistoryStore(telemetryStore.ClickhouseDB())
}
p := &provider{
service: alertmanager.New(
@@ -52,13 +63,18 @@ func New(ctx context.Context, providerSettings factory.ProviderSettings, config
configStore,
orgGetter,
notificationManager,
maintenanceExprMuter,
stateHistoryStore,
),
settings: settings,
config: config,
configStore: configStore,
stateStore: stateStore,
notificationManager: notificationManager,
stopC: make(chan struct{}),
settings: settings,
config: config,
configStore: configStore,
stateStore: stateStore,
notificationManager: notificationManager,
maintenanceStore: sqlalertmanagerstore.NewMaintenanceStore(sqlstore),
maintenanceExprMuter: maintenanceExprMuter,
orgGetter: orgGetter,
stopC: make(chan struct{}),
}
return p, nil
@@ -70,16 +86,28 @@ func (provider *provider) Start(ctx context.Context) error {
return err
}
ticker := time.NewTicker(provider.config.Signoz.PollInterval)
defer ticker.Stop()
// Initial maintenance sync before entering the ticker loop.
provider.syncMaintenance(ctx, provider.maintenanceExprMuter)
// Start background sweep for stale alerts in state history tracking.
provider.service.StartStateHistorySweep(ctx)
serverTicker := time.NewTicker(provider.config.Signoz.PollInterval)
defer serverTicker.Stop()
maintenanceTicker := time.NewTicker(maintenanceSyncInterval)
defer maintenanceTicker.Stop()
for {
select {
case <-provider.stopC:
return nil
case <-ticker.C:
case <-serverTicker.C:
if err := provider.service.SyncServers(ctx); err != nil {
provider.settings.Logger().ErrorContext(ctx, "failed to sync alertmanager servers", "error", err)
}
case <-maintenanceTicker.C:
provider.syncMaintenance(ctx, provider.maintenanceExprMuter)
}
}
}
@@ -561,3 +589,89 @@ func (provider *provider) DeleteAllInhibitRulesByRuleId(ctx context.Context, org
return provider.configStore.Set(ctx, config)
}
func (provider *provider) GetAllPlannedMaintenance(ctx context.Context, orgID string) ([]*alertmanagertypes.GettablePlannedMaintenance, error) {
return provider.maintenanceStore.GetAllPlannedMaintenance(ctx, orgID)
}
func (provider *provider) GetPlannedMaintenanceByID(ctx context.Context, id valuer.UUID) (*alertmanagertypes.GettablePlannedMaintenance, error) {
return provider.maintenanceStore.GetPlannedMaintenanceByID(ctx, id)
}
func (provider *provider) CreatePlannedMaintenance(ctx context.Context, maintenance alertmanagertypes.GettablePlannedMaintenance) (valuer.UUID, error) {
return provider.maintenanceStore.CreatePlannedMaintenance(ctx, maintenance)
}
func (provider *provider) EditPlannedMaintenance(ctx context.Context, maintenance alertmanagertypes.GettablePlannedMaintenance, id valuer.UUID) error {
return provider.maintenanceStore.EditPlannedMaintenance(ctx, maintenance, id)
}
func (provider *provider) DeletePlannedMaintenance(ctx context.Context, id valuer.UUID) error {
return provider.maintenanceStore.DeletePlannedMaintenance(ctx, id)
}
func (provider *provider) RecordRuleStateHistory(ctx context.Context, orgID string, entries []alertmanagertypes.RuleStateHistory) error {
return provider.service.RecordRuleStateHistory(ctx, orgID, entries)
}
func (provider *provider) GetLastSavedRuleStateHistory(ctx context.Context, ruleID string) ([]alertmanagertypes.RuleStateHistory, error) {
return provider.service.GetLastSavedRuleStateHistory(ctx, ruleID)
}
func (provider *provider) GetRuleStateHistoryTimeline(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStateTimeline, error) {
return provider.service.GetRuleStateHistoryTimeline(ctx, orgID, ruleID, params)
}
func (provider *provider) GetRuleStateHistoryTopContributors(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateHistoryContributor, error) {
return provider.service.GetRuleStateHistoryTopContributors(ctx, orgID, ruleID, params)
}
func (provider *provider) GetOverallStateTransitions(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) ([]alertmanagertypes.RuleStateTransition, error) {
return provider.service.GetOverallStateTransitions(ctx, orgID, ruleID, params)
}
func (provider *provider) GetRuleStats(ctx context.Context, orgID string, ruleID string, params *alertmanagertypes.QueryRuleStateHistory) (*alertmanagertypes.RuleStats, error) {
return provider.service.GetRuleStats(ctx, orgID, ruleID, params)
}
const (
maintenanceSyncInterval = 30 * time.Second
)
// syncMaintenance checks planned maintenance windows and updates the given
// MaintenanceExprMuter with active maintenance entries. The muter is injected
// into the notification pipeline as a MuteStage, suppressing notifications
// while allowing rules to continue evaluating (preserving state history).
func (provider *provider) syncMaintenance(ctx context.Context, muter *MaintenanceExprMuter) {
orgs, err := provider.orgGetter.ListByOwnedKeyRange(ctx)
if err != nil {
provider.settings.Logger().ErrorContext(ctx, "failed to list orgs for maintenance sync", "error", err)
return
}
now := time.Now()
var activeExprs []activeMaintenanceExpr
for _, org := range orgs {
orgID := org.ID.StringValue()
maintenanceList, err := provider.maintenanceStore.GetAllPlannedMaintenance(ctx, orgID)
if err != nil {
provider.settings.Logger().ErrorContext(ctx, "failed to get planned maintenance for sync", "orgID", orgID, "error", err)
continue
}
for _, maint := range maintenanceList {
_, active := maint.CurrentWindowEndTime(now)
if !active {
continue
}
activeExprs = append(activeExprs, activeMaintenanceExpr{
ruleIDs: maint.RuleIDs,
expression: maint.Expression,
})
}
}
muter.SetActiveExpressions(activeExprs)
}

View File

@@ -0,0 +1,216 @@
package signozalertmanager
import (
"log/slog"
"testing"
"github.com/prometheus/common/model"
"github.com/stretchr/testify/assert"
)
func TestMaintenanceExprMuter(t *testing.T) {
logger := slog.New(slog.DiscardHandler)
tests := []struct {
name string
exprs []activeMaintenanceExpr
labels model.LabelSet
want bool
}{
// --- no maintenance ---
{
name: "no expressions - not muted",
exprs: nil,
labels: model.LabelSet{"env": "prod"},
want: false,
},
// --- expression only (ruleIDs empty = all rules) ---
{
name: "expression only - matching",
exprs: []activeMaintenanceExpr{
{expression: `env == "prod"`},
},
labels: model.LabelSet{"env": "prod"},
want: true,
},
{
name: "expression only - non-matching",
exprs: []activeMaintenanceExpr{
{expression: `env == "prod"`},
},
labels: model.LabelSet{"env": "staging"},
want: false,
},
{
name: "expression only - matches regardless of ruleId label",
exprs: []activeMaintenanceExpr{
{expression: `env == "prod"`},
},
labels: model.LabelSet{"env": "prod", "ruleId": "any-rule"},
want: true,
},
// --- ruleIDs only (expression empty = all labels) ---
{
name: "ruleIDs only - matching rule",
exprs: []activeMaintenanceExpr{
{ruleIDs: []string{"rule-1", "rule-2"}},
},
labels: model.LabelSet{"ruleId": "rule-1", "env": "prod"},
want: true,
},
{
name: "ruleIDs only - non-matching rule",
exprs: []activeMaintenanceExpr{
{ruleIDs: []string{"rule-1", "rule-2"}},
},
labels: model.LabelSet{"ruleId": "rule-3", "env": "prod"},
want: false,
},
{
name: "ruleIDs only - no ruleId label on alert",
exprs: []activeMaintenanceExpr{
{ruleIDs: []string{"rule-1"}},
},
labels: model.LabelSet{"env": "prod"},
want: false,
},
// --- ruleIDs AND expression ---
{
name: "ruleIDs AND expression - both match",
exprs: []activeMaintenanceExpr{
{ruleIDs: []string{"rule-1"}, expression: `severity == "critical"`},
},
labels: model.LabelSet{"ruleId": "rule-1", "severity": "critical"},
want: true,
},
{
name: "ruleIDs AND expression - rule matches, expression does not",
exprs: []activeMaintenanceExpr{
{ruleIDs: []string{"rule-1"}, expression: `severity == "critical"`},
},
labels: model.LabelSet{"ruleId": "rule-1", "severity": "warning"},
want: false,
},
{
name: "ruleIDs AND expression - expression matches, rule does not",
exprs: []activeMaintenanceExpr{
{ruleIDs: []string{"rule-1"}, expression: `severity == "critical"`},
},
labels: model.LabelSet{"ruleId": "rule-999", "severity": "critical"},
want: false,
},
{
name: "ruleIDs AND expression - neither matches",
exprs: []activeMaintenanceExpr{
{ruleIDs: []string{"rule-1"}, expression: `severity == "critical"`},
},
labels: model.LabelSet{"ruleId": "rule-999", "severity": "warning"},
want: false,
},
// --- catch-all (both empty) ---
{
name: "catch-all - empty ruleIDs and empty expression mutes everything",
exprs: []activeMaintenanceExpr{
{},
},
labels: model.LabelSet{"ruleId": "any-rule", "env": "anything"},
want: true,
},
// --- multiple expressions ---
{
name: "multiple entries - first matches",
exprs: []activeMaintenanceExpr{
{expression: `env == "prod"`},
{expression: `env == "staging"`},
},
labels: model.LabelSet{"env": "prod"},
want: true,
},
{
name: "multiple entries - second matches",
exprs: []activeMaintenanceExpr{
{expression: `env == "staging"`},
{expression: `env == "prod"`},
},
labels: model.LabelSet{"env": "prod"},
want: true,
},
{
name: "multiple entries - none match",
exprs: []activeMaintenanceExpr{
{expression: `env == "staging"`},
{expression: `env == "dev"`},
},
labels: model.LabelSet{"env": "prod"},
want: false,
},
{
name: "multiple entries - ruleIDs entry matches, expression entry does not",
exprs: []activeMaintenanceExpr{
{ruleIDs: []string{"rule-1"}},
{expression: `env == "staging"`},
},
labels: model.LabelSet{"ruleId": "rule-1", "env": "prod"},
want: true,
},
// --- complex expressions ---
{
name: "complex expression with AND",
exprs: []activeMaintenanceExpr{
{expression: `severity == "critical" && env == "prod"`},
},
labels: model.LabelSet{"severity": "critical", "env": "prod"},
want: true,
},
{
name: "complex expression with AND - partial match",
exprs: []activeMaintenanceExpr{
{expression: `severity == "critical" && env == "prod"`},
},
labels: model.LabelSet{"severity": "warning", "env": "prod"},
want: false,
},
{
name: "expression with OR logic",
exprs: []activeMaintenanceExpr{
{expression: `env == "prod" || env == "staging"`},
},
labels: model.LabelSet{"env": "staging"},
want: true,
},
{
name: "expression with nested label (dotted key)",
exprs: []activeMaintenanceExpr{
{expression: `labels.env == "prod"`},
},
labels: model.LabelSet{"labels.env": "prod"},
want: true,
},
// --- ruleId as expression (user can also match ruleId via expression) ---
{
name: "expression matching specific ruleId label",
exprs: []activeMaintenanceExpr{
{expression: `ruleId == "rule-1"`},
},
labels: model.LabelSet{"ruleId": "rule-1", "env": "prod"},
want: true,
},
{
name: "expression matching specific ruleId label - non-matching",
exprs: []activeMaintenanceExpr{
{expression: `ruleId == "rule-1"`},
},
labels: model.LabelSet{"ruleId": "rule-3", "env": "prod"},
want: false,
},
}
for _, tc := range tests {
t.Run(tc.name, func(t *testing.T) {
muter := NewMaintenanceExprMuter(logger)
muter.SetActiveExpressions(tc.exprs)
got := muter.Mutes(tc.labels)
assert.Equal(t, tc.want, got)
})
}
}

View File

@@ -0,0 +1,263 @@
package alertmanager
import (
"encoding/json"
"strconv"
"sync"
"time"
"github.com/SigNoz/signoz/pkg/types/alertmanagertypes"
"github.com/prometheus/alertmanager/types"
"github.com/prometheus/common/model"
)
// trackedAlert represents the last known state of a single alert series.
type trackedAlert struct {
state string // "firing" or "inactive"
labels string // JSON labels
ruleName string
value float64
lastSeen time.Time
}
// ruleOverallState tracks the overall state of a rule across all its alert series.
type ruleOverallState struct {
state string // "firing" or "inactive"
}
// stateTracker maintains per-org, per-rule, per-fingerprint alert state
// to detect state transitions when PutAlerts is called.
type stateTracker struct {
mu sync.Mutex
alerts map[string]map[string]map[uint64]*trackedAlert // orgID → ruleID → fingerprint → state
overallState map[string]map[string]*ruleOverallState // orgID → ruleID → overall state
}
func newStateTracker() *stateTracker {
return &stateTracker{
alerts: make(map[string]map[string]map[uint64]*trackedAlert),
overallState: make(map[string]map[string]*ruleOverallState),
}
}
// processAlerts detects state transitions from incoming alerts and returns
// RuleStateHistory entries for transitions only.
func (t *stateTracker) processAlerts(orgID string, alerts []*types.Alert, now time.Time) []alertmanagertypes.RuleStateHistory {
t.mu.Lock()
defer t.mu.Unlock()
if _, ok := t.alerts[orgID]; !ok {
t.alerts[orgID] = make(map[string]map[uint64]*trackedAlert)
}
if _, ok := t.overallState[orgID]; !ok {
t.overallState[orgID] = make(map[string]*ruleOverallState)
}
var entries []alertmanagertypes.RuleStateHistory
// Track which rules were affected in this batch for overall_state computation.
affectedRules := make(map[string]bool)
for _, alert := range alerts {
ruleID := string(alert.Labels[model.LabelName("ruleId")])
if ruleID == "" {
continue
}
fp := uint64(alert.Fingerprint())
ruleName := string(alert.Labels[model.LabelName("alertname")])
labelsJSON := labelsToJSON(alert.Labels)
value := valueFromAnnotations(alert.Annotations)
var newState string
if !alert.EndsAt.IsZero() && !alert.EndsAt.After(now) {
newState = "inactive"
} else {
newState = "firing"
}
if _, ok := t.alerts[orgID][ruleID]; !ok {
t.alerts[orgID][ruleID] = make(map[uint64]*trackedAlert)
}
tracked, exists := t.alerts[orgID][ruleID][fp]
if !exists {
// First time seeing this alert.
t.alerts[orgID][ruleID][fp] = &trackedAlert{
state: newState,
labels: labelsJSON,
ruleName: ruleName,
value: value,
lastSeen: now,
}
if newState == "firing" {
// New firing alert — record transition.
entries = append(entries, alertmanagertypes.RuleStateHistory{
OrgID: orgID,
RuleID: ruleID,
RuleName: ruleName,
State: "firing",
StateChanged: true,
UnixMilli: now.UnixMilli(),
Labels: labelsJSON,
Fingerprint: fp,
Value: value,
})
affectedRules[ruleID] = true
}
// Not found + resolved: no-op (we didn't track it firing).
continue
}
// Alert exists in tracker — check for transition.
tracked.lastSeen = now
tracked.value = value
tracked.labels = labelsJSON
if tracked.state != newState {
// State transition detected.
tracked.state = newState
entries = append(entries, alertmanagertypes.RuleStateHistory{
OrgID: orgID,
RuleID: ruleID,
RuleName: ruleName,
State: newState,
StateChanged: true,
UnixMilli: now.UnixMilli(),
Labels: labelsJSON,
Fingerprint: fp,
Value: value,
})
affectedRules[ruleID] = true
}
// Same state — no transition, nothing to record.
}
// Compute overall_state for affected rules and set on entries.
for ruleID := range affectedRules {
currentOverall := t.computeOverallState(orgID, ruleID)
prevOverall, hasPrev := t.overallState[orgID][ruleID]
overallChanged := !hasPrev || prevOverall.state != currentOverall
if !hasPrev {
t.overallState[orgID][ruleID] = &ruleOverallState{state: currentOverall}
} else {
prevOverall.state = currentOverall
}
// Set overall_state on all entries for this rule.
for i := range entries {
if entries[i].RuleID == ruleID {
entries[i].OverallState = currentOverall
entries[i].OverallStateChanged = overallChanged
}
}
}
return entries
}
// computeOverallState returns "firing" if any tracked alert for the rule is firing.
func (t *stateTracker) computeOverallState(orgID, ruleID string) string {
ruleAlerts, ok := t.alerts[orgID][ruleID]
if !ok {
return "inactive"
}
for _, a := range ruleAlerts {
if a.state == "firing" {
return "firing"
}
}
return "inactive"
}
// sweepStale finds alerts that haven't been updated within staleTimeout and
// records them as resolved. Returns transition entries grouped by orgID.
func (t *stateTracker) sweepStale(staleTimeout time.Duration, now time.Time) map[string][]alertmanagertypes.RuleStateHistory {
t.mu.Lock()
defer t.mu.Unlock()
result := make(map[string][]alertmanagertypes.RuleStateHistory)
affectedRules := make(map[string]map[string]bool) // orgID → ruleID → true
for orgID, rules := range t.alerts {
for ruleID, fingerprints := range rules {
for fp, tracked := range fingerprints {
if tracked.state != "firing" {
continue
}
if now.Sub(tracked.lastSeen) <= staleTimeout {
continue
}
// Stale firing alert — mark as resolved.
tracked.state = "inactive"
result[orgID] = append(result[orgID], alertmanagertypes.RuleStateHistory{
OrgID: orgID,
RuleID: ruleID,
RuleName: tracked.ruleName,
State: "inactive",
StateChanged: true,
UnixMilli: now.UnixMilli(),
Labels: tracked.labels,
Fingerprint: fp,
Value: tracked.value,
})
if affectedRules[orgID] == nil {
affectedRules[orgID] = make(map[string]bool)
}
affectedRules[orgID][ruleID] = true
}
}
}
// Compute overall_state for affected rules.
for orgID, rules := range affectedRules {
for ruleID := range rules {
currentOverall := t.computeOverallState(orgID, ruleID)
prevOverall, hasPrev := t.overallState[orgID][ruleID]
overallChanged := !hasPrev || prevOverall.state != currentOverall
if hasPrev {
prevOverall.state = currentOverall
}
for i := range result[orgID] {
if result[orgID][i].RuleID == ruleID {
result[orgID][i].OverallState = currentOverall
result[orgID][i].OverallStateChanged = overallChanged
}
}
}
}
return result
}
// labelsToJSON converts a model.LabelSet to a JSON string.
func labelsToJSON(ls model.LabelSet) string {
m := make(map[string]string, len(ls))
for k, v := range ls {
m[string(k)] = string(v)
}
b, err := json.Marshal(m)
if err != nil {
return "{}"
}
return string(b)
}
// valueFromAnnotations extracts the metric value from alert annotations.
func valueFromAnnotations(annotations model.LabelSet) float64 {
valStr := string(annotations[model.LabelName("value")])
if valStr == "" {
return 0
}
v, err := strconv.ParseFloat(valStr, 64)
if err != nil {
return 0
}
return v
}

View File

@@ -0,0 +1,328 @@
package alertmanager
import (
"testing"
"time"
"github.com/prometheus/alertmanager/types"
"github.com/prometheus/common/model"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func makeAlert(ruleID, alertname string, firing bool, now time.Time, extraLabels map[string]string) *types.Alert {
labels := model.LabelSet{
"ruleId": model.LabelValue(ruleID),
"alertname": model.LabelValue(alertname),
}
for k, v := range extraLabels {
labels[model.LabelName(k)] = model.LabelValue(v)
}
alert := &types.Alert{
Alert: model.Alert{
Labels: labels,
Annotations: model.LabelSet{"value": "42.5"},
StartsAt: now.Add(-1 * time.Minute),
},
UpdatedAt: now,
}
if firing {
alert.EndsAt = now.Add(5 * time.Minute) // future = firing
} else {
alert.EndsAt = now.Add(-10 * time.Second) // past = resolved
}
return alert
}
func TestProcessAlerts_NewFiringAlert(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
alerts := []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, map[string]string{"host": "server-1"}),
}
entries := tracker.processAlerts("org-1", alerts, now)
require.Len(t, entries, 1)
assert.Equal(t, "firing", entries[0].State)
assert.Equal(t, "rule-1", entries[0].RuleID)
assert.Equal(t, "HighCPU", entries[0].RuleName)
assert.Equal(t, "org-1", entries[0].OrgID)
assert.Equal(t, true, entries[0].StateChanged)
assert.Equal(t, 42.5, entries[0].Value)
assert.Equal(t, now.UnixMilli(), entries[0].UnixMilli)
assert.Equal(t, "firing", entries[0].OverallState)
assert.Equal(t, true, entries[0].OverallStateChanged)
}
func TestProcessAlerts_StillFiringNoTransition(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
alerts := []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, map[string]string{"host": "server-1"}),
}
// First call: new firing.
entries := tracker.processAlerts("org-1", alerts, now)
require.Len(t, entries, 1)
// Second call: still firing — no transition.
entries = tracker.processAlerts("org-1", alerts, now.Add(1*time.Minute))
assert.Empty(t, entries)
}
func TestProcessAlerts_FiringThenResolved(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
// First: fire the alert.
firingAlerts := []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, map[string]string{"host": "server-1"}),
}
entries := tracker.processAlerts("org-1", firingAlerts, now)
require.Len(t, entries, 1)
assert.Equal(t, "firing", entries[0].State)
// Second: resolve the alert.
resolvedAlerts := []*types.Alert{
makeAlert("rule-1", "HighCPU", false, now.Add(5*time.Minute), map[string]string{"host": "server-1"}),
}
entries = tracker.processAlerts("org-1", resolvedAlerts, now.Add(5*time.Minute))
require.Len(t, entries, 1)
assert.Equal(t, "inactive", entries[0].State)
assert.Equal(t, "rule-1", entries[0].RuleID)
assert.Equal(t, "inactive", entries[0].OverallState)
assert.Equal(t, true, entries[0].OverallStateChanged)
}
func TestProcessAlerts_ResolvedWithoutPriorFiring(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
// A resolved alert arriving without prior tracking should produce no entry.
alerts := []*types.Alert{
makeAlert("rule-1", "HighCPU", false, now, map[string]string{"host": "server-1"}),
}
entries := tracker.processAlerts("org-1", alerts, now)
assert.Empty(t, entries)
}
func TestProcessAlerts_ReFiring(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
// Fire.
entries := tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, map[string]string{"host": "server-1"}),
}, now)
require.Len(t, entries, 1)
assert.Equal(t, "firing", entries[0].State)
// Resolve.
entries = tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", false, now.Add(5*time.Minute), map[string]string{"host": "server-1"}),
}, now.Add(5*time.Minute))
require.Len(t, entries, 1)
assert.Equal(t, "inactive", entries[0].State)
// Re-fire.
entries = tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now.Add(10*time.Minute), map[string]string{"host": "server-1"}),
}, now.Add(10*time.Minute))
require.Len(t, entries, 1)
assert.Equal(t, "firing", entries[0].State)
assert.Equal(t, "firing", entries[0].OverallState)
assert.Equal(t, true, entries[0].OverallStateChanged)
}
func TestProcessAlerts_OverallStateComputation(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
// Fire two series for the same rule.
entries := tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, map[string]string{"host": "server-1"}),
makeAlert("rule-1", "HighCPU", true, now, map[string]string{"host": "server-2"}),
}, now)
require.Len(t, entries, 2)
assert.Equal(t, "firing", entries[0].OverallState)
assert.Equal(t, "firing", entries[1].OverallState)
// Resolve only one series — overall should still be "firing".
entries = tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", false, now.Add(5*time.Minute), map[string]string{"host": "server-1"}),
}, now.Add(5*time.Minute))
require.Len(t, entries, 1)
assert.Equal(t, "inactive", entries[0].State)
assert.Equal(t, "firing", entries[0].OverallState)
assert.Equal(t, false, entries[0].OverallStateChanged) // still firing overall
// Resolve the second series — overall should transition to "inactive".
entries = tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", false, now.Add(6*time.Minute), map[string]string{"host": "server-2"}),
}, now.Add(6*time.Minute))
require.Len(t, entries, 1)
assert.Equal(t, "inactive", entries[0].State)
assert.Equal(t, "inactive", entries[0].OverallState)
assert.Equal(t, true, entries[0].OverallStateChanged) // transitioned to inactive
}
func TestProcessAlerts_MultipleRulesIndependent(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
entries := tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, map[string]string{"host": "server-1"}),
makeAlert("rule-2", "HighMem", true, now, map[string]string{"host": "server-1"}),
}, now)
require.Len(t, entries, 2)
// Each rule has its own overall state.
assert.Equal(t, "rule-1", entries[0].RuleID)
assert.Equal(t, "rule-2", entries[1].RuleID)
assert.Equal(t, "firing", entries[0].OverallState)
assert.Equal(t, "firing", entries[1].OverallState)
}
func TestProcessAlerts_AlertWithoutRuleIDSkipped(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
alert := &types.Alert{
Alert: model.Alert{
Labels: model.LabelSet{"alertname": "NoRuleID"},
StartsAt: now.Add(-1 * time.Minute),
EndsAt: now.Add(5 * time.Minute),
},
UpdatedAt: now,
}
entries := tracker.processAlerts("org-1", []*types.Alert{alert}, now)
assert.Empty(t, entries)
}
func TestProcessAlerts_MultipleOrgs(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
// Org 1 fires.
entries1 := tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, nil),
}, now)
require.Len(t, entries1, 1)
assert.Equal(t, "org-1", entries1[0].OrgID)
// Org 2 fires same rule ID — independent tracking.
entries2 := tracker.processAlerts("org-2", []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, nil),
}, now)
require.Len(t, entries2, 1)
assert.Equal(t, "org-2", entries2[0].OrgID)
}
func TestSweepStale_FiringAlertBecomesInactive(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
// Fire an alert.
tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, map[string]string{"host": "server-1"}),
}, now)
// Sweep with staleTimeout = 5 minutes, 10 minutes later.
result := tracker.sweepStale(5*time.Minute, now.Add(10*time.Minute))
require.Len(t, result["org-1"], 1)
assert.Equal(t, "inactive", result["org-1"][0].State)
assert.Equal(t, "rule-1", result["org-1"][0].RuleID)
assert.Equal(t, "inactive", result["org-1"][0].OverallState)
assert.Equal(t, true, result["org-1"][0].OverallStateChanged)
}
func TestSweepStale_RecentAlertNotSwept(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
// Fire an alert.
tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, map[string]string{"host": "server-1"}),
}, now)
// Sweep with staleTimeout = 10 minutes, only 2 minutes later.
result := tracker.sweepStale(10*time.Minute, now.Add(2*time.Minute))
assert.Empty(t, result)
}
func TestSweepStale_InactiveAlertNotSwept(t *testing.T) {
tracker := newStateTracker()
now := time.Now()
// Fire then resolve.
tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", true, now, nil),
}, now)
tracker.processAlerts("org-1", []*types.Alert{
makeAlert("rule-1", "HighCPU", false, now.Add(1*time.Minute), nil),
}, now.Add(1*time.Minute))
// Sweep much later — should produce nothing since alert is already inactive.
result := tracker.sweepStale(5*time.Minute, now.Add(30*time.Minute))
assert.Empty(t, result)
}
func TestLabelsToJSON(t *testing.T) {
ls := model.LabelSet{
"alertname": "HighCPU",
"env": "prod",
}
result := labelsToJSON(ls)
// Parse back and verify.
parsed := labelsFromJSON(result)
require.NotNil(t, parsed)
assert.Equal(t, model.LabelValue("HighCPU"), parsed["alertname"])
assert.Equal(t, model.LabelValue("prod"), parsed["env"])
}
func TestValueFromAnnotations(t *testing.T) {
tests := []struct {
name string
annotations model.LabelSet
want float64
}{
{
name: "valid float",
annotations: model.LabelSet{"value": "42.5"},
want: 42.5,
},
{
name: "empty value",
annotations: model.LabelSet{},
want: 0,
},
{
name: "invalid value",
annotations: model.LabelSet{"value": "not-a-number"},
want: 0,
},
{
name: "scientific notation",
annotations: model.LabelSet{"value": "1.5E+02"},
want: 150,
},
}
for _, tc := range tests {
t.Run(tc.name, func(t *testing.T) {
got := valueFromAnnotations(tc.annotations)
assert.Equal(t, tc.want, got)
})
}
}

View File

@@ -26,6 +26,8 @@ import (
"github.com/SigNoz/signoz/pkg/alertmanager"
"github.com/SigNoz/signoz/pkg/apis/fields"
"github.com/SigNoz/signoz/pkg/http/handler"
"github.com/SigNoz/signoz/pkg/types/alertmanagertypes"
errorsV2 "github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/http/middleware"
"github.com/SigNoz/signoz/pkg/http/render"
@@ -492,18 +494,109 @@ func (aH *APIHandler) Respond(w http.ResponseWriter, data interface{}) {
func (aH *APIHandler) RegisterRoutes(router *mux.Router, am *middleware.AuthZ) {
router.HandleFunc("/api/v1/query_range", am.ViewAccess(aH.queryRangeMetrics)).Methods(http.MethodGet)
router.HandleFunc("/api/v1/query", am.ViewAccess(aH.queryMetrics)).Methods(http.MethodGet)
router.HandleFunc("/api/v1/channels", am.ViewAccess(aH.AlertmanagerAPI.ListChannels)).Methods(http.MethodGet)
router.HandleFunc("/api/v1/channels/{id}", am.ViewAccess(aH.AlertmanagerAPI.GetChannelByID)).Methods(http.MethodGet)
router.HandleFunc("/api/v1/channels/{id}", am.AdminAccess(aH.AlertmanagerAPI.UpdateChannelByID)).Methods(http.MethodPut)
router.HandleFunc("/api/v1/channels/{id}", am.AdminAccess(aH.AlertmanagerAPI.DeleteChannelByID)).Methods(http.MethodDelete)
router.HandleFunc("/api/v1/channels", am.EditAccess(aH.AlertmanagerAPI.CreateChannel)).Methods(http.MethodPost)
router.HandleFunc("/api/v1/testChannel", am.EditAccess(aH.AlertmanagerAPI.TestReceiver)).Methods(http.MethodPost)
router.Handle("/api/v1/channels", handler.New(am.ViewAccess(aH.AlertmanagerAPI.ListChannels), handler.OpenAPIDef{
ID: "ListChannels",
Tags: []string{"channels"},
Summary: "List notification channels",
Description: "Returns all notification channels for the organization.",
Response: make([]*alertmanagertypes.Channel, 0),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
})).Methods(http.MethodGet)
router.Handle("/api/v1/channels/{id}", handler.New(am.ViewAccess(aH.AlertmanagerAPI.GetChannelByID), handler.OpenAPIDef{
ID: "GetChannelByID",
Tags: []string{"channels"},
Summary: "Get a notification channel",
Description: "Returns a single notification channel by ID.",
Response: new(alertmanagertypes.Channel),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusNotFound},
})).Methods(http.MethodGet)
router.Handle("/api/v1/channels/{id}", handler.New(am.AdminAccess(aH.AlertmanagerAPI.UpdateChannelByID), handler.OpenAPIDef{
ID: "UpdateChannelByID",
Tags: []string{"channels"},
Summary: "Update a notification channel",
Description: "Updates a notification channel by ID.",
SuccessStatusCode: http.StatusNoContent,
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusNotFound},
})).Methods(http.MethodPut)
router.Handle("/api/v1/channels/{id}", handler.New(am.AdminAccess(aH.AlertmanagerAPI.DeleteChannelByID), handler.OpenAPIDef{
ID: "DeleteChannelByID",
Tags: []string{"channels"},
Summary: "Delete a notification channel",
Description: "Deletes a notification channel by ID.",
SuccessStatusCode: http.StatusNoContent,
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusNotFound},
})).Methods(http.MethodDelete)
router.Handle("/api/v1/channels", handler.New(am.EditAccess(aH.AlertmanagerAPI.CreateChannel), handler.OpenAPIDef{
ID: "CreateChannel",
Tags: []string{"channels"},
Summary: "Create a notification channel",
Description: "Creates a new notification channel.",
Response: new(alertmanagertypes.Channel),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusCreated,
ErrorStatusCodes: []int{http.StatusBadRequest},
})).Methods(http.MethodPost)
router.Handle("/api/v1/testChannel", handler.New(am.EditAccess(aH.AlertmanagerAPI.TestReceiver), handler.OpenAPIDef{
ID: "TestReceiver",
Tags: []string{"channels"},
Summary: "Test a notification channel",
Description: "Sends a test alert to a receiver configuration.",
SuccessStatusCode: http.StatusNoContent,
ErrorStatusCodes: []int{http.StatusBadRequest},
})).Methods(http.MethodPost)
router.HandleFunc("/api/v1/route_policies", am.ViewAccess(aH.AlertmanagerAPI.GetAllRoutePolicies)).Methods(http.MethodGet)
router.HandleFunc("/api/v1/route_policies/{id}", am.ViewAccess(aH.AlertmanagerAPI.GetRoutePolicyByID)).Methods(http.MethodGet)
router.HandleFunc("/api/v1/route_policies", am.AdminAccess(aH.AlertmanagerAPI.CreateRoutePolicy)).Methods(http.MethodPost)
router.HandleFunc("/api/v1/route_policies/{id}", am.AdminAccess(aH.AlertmanagerAPI.DeleteRoutePolicyByID)).Methods(http.MethodDelete)
router.HandleFunc("/api/v1/route_policies/{id}", am.AdminAccess(aH.AlertmanagerAPI.UpdateRoutePolicy)).Methods(http.MethodPut)
router.Handle("/api/v1/route_policies", handler.New(am.ViewAccess(aH.AlertmanagerAPI.GetAllRoutePolicies), handler.OpenAPIDef{
ID: "GetAllRoutePolicies",
Tags: []string{"route_policies"},
Summary: "List route policies",
Description: "Returns all notification route policies.",
Response: make([]*alertmanagertypes.GettableRoutePolicy, 0),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
})).Methods(http.MethodGet)
router.Handle("/api/v1/route_policies/{id}", handler.New(am.ViewAccess(aH.AlertmanagerAPI.GetRoutePolicyByID), handler.OpenAPIDef{
ID: "GetRoutePolicyByID",
Tags: []string{"route_policies"},
Summary: "Get a route policy",
Description: "Returns a single notification route policy by ID.",
Response: new(alertmanagertypes.GettableRoutePolicy),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusNotFound},
})).Methods(http.MethodGet)
router.Handle("/api/v1/route_policies", handler.New(am.AdminAccess(aH.AlertmanagerAPI.CreateRoutePolicy), handler.OpenAPIDef{
ID: "CreateRoutePolicy",
Tags: []string{"route_policies"},
Summary: "Create a route policy",
Description: "Creates a new notification route policy.",
Request: new(alertmanagertypes.PostableRoutePolicy),
Response: new(alertmanagertypes.GettableRoutePolicy),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusCreated,
ErrorStatusCodes: []int{http.StatusBadRequest},
})).Methods(http.MethodPost)
router.Handle("/api/v1/route_policies/{id}", handler.New(am.AdminAccess(aH.AlertmanagerAPI.DeleteRoutePolicyByID), handler.OpenAPIDef{
ID: "DeleteRoutePolicyByID",
Tags: []string{"route_policies"},
Summary: "Delete a route policy",
Description: "Deletes a notification route policy by ID.",
SuccessStatusCode: http.StatusNoContent,
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusNotFound},
})).Methods(http.MethodDelete)
router.Handle("/api/v1/route_policies/{id}", handler.New(am.AdminAccess(aH.AlertmanagerAPI.UpdateRoutePolicy), handler.OpenAPIDef{
ID: "UpdateRoutePolicy",
Tags: []string{"route_policies"},
Summary: "Update a route policy",
Description: "Updates a notification route policy by ID.",
Request: new(alertmanagertypes.PostableRoutePolicy),
Response: new(alertmanagertypes.GettableRoutePolicy),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusNotFound},
})).Methods(http.MethodPut)
router.HandleFunc("/api/v1/alerts", am.ViewAccess(aH.AlertmanagerAPI.GetAlerts)).Methods(http.MethodGet)
@@ -525,6 +618,103 @@ func (aH *APIHandler) RegisterRoutes(router *mux.Router, am *middleware.AuthZ) {
router.HandleFunc("/api/v1/downtime_schedules/{id}", am.EditAccess(aH.editDowntimeSchedule)).Methods(http.MethodPut)
router.HandleFunc("/api/v1/downtime_schedules/{id}", am.EditAccess(aH.deleteDowntimeSchedule)).Methods(http.MethodDelete)
// V2 downtime schedules (alertmanager-based)
router.Handle("/api/v2/downtime_schedules", handler.New(am.ViewAccess(aH.AlertmanagerAPI.ListDowntimeSchedules), handler.OpenAPIDef{
ID: "ListDowntimeSchedules",
Tags: []string{"downtime_schedules"},
Summary: "List downtime schedules",
Description: "Returns all planned maintenance schedules for the organization. Supports filtering by active and recurring query parameters.",
Response: make([]*alertmanagertypes.GettablePlannedMaintenance, 0),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
})).Methods(http.MethodGet)
router.Handle("/api/v2/downtime_schedules/{id}", handler.New(am.ViewAccess(aH.AlertmanagerAPI.GetDowntimeSchedule), handler.OpenAPIDef{
ID: "GetDowntimeSchedule",
Tags: []string{"downtime_schedules"},
Summary: "Get a downtime schedule",
Description: "Returns a single planned maintenance schedule by ID.",
Response: new(alertmanagertypes.GettablePlannedMaintenance),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusNotFound},
})).Methods(http.MethodGet)
router.Handle("/api/v2/downtime_schedules", handler.New(am.EditAccess(aH.AlertmanagerAPI.CreateDowntimeSchedule), handler.OpenAPIDef{
ID: "CreateDowntimeSchedule",
Tags: []string{"downtime_schedules"},
Summary: "Create a downtime schedule",
Description: "Creates a new planned maintenance schedule.",
Request: new(alertmanagertypes.GettablePlannedMaintenance),
Response: nil,
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest},
})).Methods(http.MethodPost)
router.Handle("/api/v2/downtime_schedules/{id}", handler.New(am.EditAccess(aH.AlertmanagerAPI.EditDowntimeSchedule), handler.OpenAPIDef{
ID: "EditDowntimeSchedule",
Tags: []string{"downtime_schedules"},
Summary: "Update a downtime schedule",
Description: "Updates an existing planned maintenance schedule by ID.",
Request: new(alertmanagertypes.GettablePlannedMaintenance),
Response: nil,
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusNotFound},
})).Methods(http.MethodPut)
router.Handle("/api/v2/downtime_schedules/{id}", handler.New(am.EditAccess(aH.AlertmanagerAPI.DeleteDowntimeSchedule), handler.OpenAPIDef{
ID: "DeleteDowntimeSchedule",
Tags: []string{"downtime_schedules"},
Summary: "Delete a downtime schedule",
Description: "Deletes a planned maintenance schedule by ID.",
SuccessStatusCode: http.StatusNoContent,
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusNotFound},
})).Methods(http.MethodDelete)
// V2 rule state history (alertmanager-based)
router.Handle("/api/v2/rules/{id}/history/timeline", handler.New(am.ViewAccess(aH.AlertmanagerAPI.GetRuleStateHistoryTimeline), handler.OpenAPIDef{
ID: "GetRuleStateHistoryTimeline",
Tags: []string{"rule_state_history"},
Summary: "Get rule state history timeline",
Description: "Returns paginated state history entries for a rule within a time range, with optional state filter and distinct label keys for filter UI.",
Request: new(alertmanagertypes.QueryRuleStateHistory),
Response: new(alertmanagertypes.RuleStateTimeline),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest},
})).Methods(http.MethodPost)
router.Handle("/api/v2/rules/{id}/history/stats", handler.New(am.ViewAccess(aH.AlertmanagerAPI.GetRuleStats), handler.OpenAPIDef{
ID: "GetRuleStats",
Tags: []string{"rule_state_history"},
Summary: "Get rule trigger and resolution statistics",
Description: "Returns trigger counts and average resolution times for a rule, comparing the current time period against a previous period of equal length.",
Request: new(alertmanagertypes.QueryRuleStateHistory),
Response: new(alertmanagertypes.RuleStats),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest},
})).Methods(http.MethodPost)
router.Handle("/api/v2/rules/{id}/history/top_contributors", handler.New(am.ViewAccess(aH.AlertmanagerAPI.GetRuleStateHistoryTopContributors), handler.OpenAPIDef{
ID: "GetRuleStateHistoryTopContributors",
Tags: []string{"rule_state_history"},
Summary: "Get top contributing alert series",
Description: "Returns alert series (by fingerprint) that transitioned to firing most frequently for a rule within a time range, ranked by count.",
Request: new(alertmanagertypes.QueryRuleStateHistory),
Response: make([]alertmanagertypes.RuleStateHistoryContributor, 0),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest},
})).Methods(http.MethodPost)
router.Handle("/api/v2/rules/{id}/history/overall_status", handler.New(am.ViewAccess(aH.AlertmanagerAPI.GetOverallStateTransitions), handler.OpenAPIDef{
ID: "GetOverallStateTransitions",
Tags: []string{"rule_state_history"},
Summary: "Get overall state transition timeline",
Description: "Returns a timeline of contiguous firing and inactive periods for a rule within a time range, with gap-filling between transitions.",
Request: new(alertmanagertypes.QueryRuleStateHistory),
Response: make([]alertmanagertypes.RuleStateTransition, 0),
ResponseContentType: "application/json",
SuccessStatusCode: http.StatusOK,
ErrorStatusCodes: []int{http.StatusBadRequest},
})).Methods(http.MethodPost)
router.HandleFunc("/api/v1/dashboards", am.ViewAccess(aH.List)).Methods(http.MethodGet)
router.HandleFunc("/api/v1/dashboards", am.EditAccess(aH.Signoz.Handlers.Dashboard.Create)).Methods(http.MethodPost)
router.HandleFunc("/api/v1/dashboards/{id}", am.ViewAccess(aH.Get)).Methods(http.MethodGet)

View File

@@ -478,15 +478,14 @@ func (r *BaseRule) RecordRuleStateHistory(ctx context.Context, prevState, curren
}
}
if len(revisedItemsToAdd) > 0 && r.reader != nil {
if len(revisedItemsToAdd) > 0 {
zap.L().Debug("writing rule state history", zap.String("ruleid", r.ID()), zap.Any("revisedItemsToAdd", revisedItemsToAdd))
entries := make([]model.RuleStateHistory, 0, len(revisedItemsToAdd))
for _, item := range revisedItemsToAdd {
entries = append(entries, item)
}
err := r.reader.AddRuleStateHistory(ctx, entries)
if err != nil {
if err := r.reader.AddRuleStateHistory(ctx, entries); err != nil {
zap.L().Error("error while inserting rule state history", zap.Error(err), zap.Any("itemsToAdd", itemsToAdd))
}
}

View File

@@ -34,7 +34,7 @@ func TestNewHandlers(t *testing.T) {
orgGetter := implorganization.NewGetter(implorganization.NewStore(sqlstore), sharder)
notificationManager := nfmanagertest.NewMock()
require.NoError(t, err)
alertmanager, err := signozalertmanager.New(context.TODO(), providerSettings, alertmanager.Config{}, sqlstore, orgGetter, notificationManager)
alertmanager, err := signozalertmanager.New(context.TODO(), providerSettings, alertmanager.Config{}, sqlstore, orgGetter, notificationManager, nil)
require.NoError(t, err)
tokenizer := tokenizertest.NewMockTokenizer(t)
emailing := emailingtest.New()

View File

@@ -34,7 +34,7 @@ func TestNewModules(t *testing.T) {
orgGetter := implorganization.NewGetter(implorganization.NewStore(sqlstore), sharder)
notificationManager := nfmanagertest.NewMock()
require.NoError(t, err)
alertmanager, err := signozalertmanager.New(context.TODO(), providerSettings, alertmanager.Config{}, sqlstore, orgGetter, notificationManager)
alertmanager, err := signozalertmanager.New(context.TODO(), providerSettings, alertmanager.Config{}, sqlstore, orgGetter, notificationManager, nil)
require.NoError(t, err)
tokenizer := tokenizertest.NewMockTokenizer(t)
emailing := emailingtest.New()

View File

@@ -166,6 +166,8 @@ func NewSQLMigrationProviderFactories(
sqlmigration.NewAddAuthzIndexFactory(sqlstore, sqlschema),
sqlmigration.NewMigrateRbacToAuthzFactory(sqlstore),
sqlmigration.NewMigratePublicDashboardsFactory(sqlstore),
sqlmigration.NewCreatePlannedMaintenanceV2Factory(sqlstore),
sqlmigration.NewCreateRuleStateHistoryV2Factory(telemetryStore),
)
}
@@ -191,9 +193,9 @@ func NewNotificationManagerProviderFactories(routeStore alertmanagertypes.RouteS
)
}
func NewAlertmanagerProviderFactories(sqlstore sqlstore.SQLStore, orgGetter organization.Getter, nfManager nfmanager.NotificationManager) factory.NamedMap[factory.ProviderFactory[alertmanager.Alertmanager, alertmanager.Config]] {
func NewAlertmanagerProviderFactories(sqlstore sqlstore.SQLStore, orgGetter organization.Getter, nfManager nfmanager.NotificationManager, telemetryStore telemetrystore.TelemetryStore) factory.NamedMap[factory.ProviderFactory[alertmanager.Alertmanager, alertmanager.Config]] {
return factory.MustNewNamedMap(
signozalertmanager.NewFactory(sqlstore, orgGetter, nfManager),
signozalertmanager.NewFactory(sqlstore, orgGetter, nfManager, telemetryStore),
)
}

View File

@@ -58,7 +58,7 @@ func TestNewProviderFactories(t *testing.T) {
assert.NotPanics(t, func() {
orgGetter := implorganization.NewGetter(implorganization.NewStore(sqlstoretest.New(sqlstore.Config{Provider: "sqlite"}, sqlmock.QueryMatcherEqual)), nil)
notificationManager := nfmanagertest.NewMock()
NewAlertmanagerProviderFactories(sqlstoretest.New(sqlstore.Config{Provider: "sqlite"}, sqlmock.QueryMatcherEqual), orgGetter, notificationManager)
NewAlertmanagerProviderFactories(sqlstoretest.New(sqlstore.Config{Provider: "sqlite"}, sqlmock.QueryMatcherEqual), orgGetter, notificationManager, nil)
})
assert.NotPanics(t, func() {

View File

@@ -311,7 +311,7 @@ func New(
ctx,
providerSettings,
config.Alertmanager,
NewAlertmanagerProviderFactories(sqlstore, orgGetter, nfManager),
NewAlertmanagerProviderFactories(sqlstore, orgGetter, nfManager, telemetrystore),
config.Alertmanager.Provider,
)
if err != nil {

View File

@@ -0,0 +1,45 @@
package sqlmigration
import (
"context"
"github.com/SigNoz/signoz/pkg/factory"
"github.com/SigNoz/signoz/pkg/sqlstore"
"github.com/SigNoz/signoz/pkg/types/alertmanagertypes"
"github.com/uptrace/bun"
"github.com/uptrace/bun/migrate"
)
type createPlannedMaintenanceV2 struct {
sqlstore sqlstore.SQLStore
}
func NewCreatePlannedMaintenanceV2Factory(sqlstore sqlstore.SQLStore) factory.ProviderFactory[SQLMigration, Config] {
return factory.NewProviderFactory(factory.MustNewName("create_planned_maintenance_v2"), func(ctx context.Context, providerSettings factory.ProviderSettings, config Config) (SQLMigration, error) {
return &createPlannedMaintenanceV2{sqlstore: sqlstore}, nil
})
}
func (migration *createPlannedMaintenanceV2) Register(migrations *migrate.Migrations) error {
if err := migrations.Register(migration.Up, migration.Down); err != nil {
return err
}
return nil
}
func (migration *createPlannedMaintenanceV2) Up(ctx context.Context, db *bun.DB) error {
_, err := db.NewCreateTable().
Model((*alertmanagertypes.StorablePlannedMaintenance)(nil)).
IfNotExists().
Exec(ctx)
return err
}
func (migration *createPlannedMaintenanceV2) Down(ctx context.Context, db *bun.DB) error {
_, err := db.NewDropTable().
Model((*alertmanagertypes.StorablePlannedMaintenance)(nil)).
IfExists().
Exec(ctx)
return err
}

View File

@@ -0,0 +1,80 @@
package sqlmigration
import (
"context"
"github.com/SigNoz/signoz/pkg/factory"
"github.com/SigNoz/signoz/pkg/telemetrystore"
"github.com/uptrace/bun"
"github.com/uptrace/bun/migrate"
)
type createRuleStateHistoryV2 struct {
telemetryStore telemetrystore.TelemetryStore
}
func NewCreateRuleStateHistoryV2Factory(telemetryStore telemetrystore.TelemetryStore) factory.ProviderFactory[SQLMigration, Config] {
return factory.NewProviderFactory(factory.MustNewName("create_rule_state_history_v2"), func(ctx context.Context, providerSettings factory.ProviderSettings, config Config) (SQLMigration, error) {
return &createRuleStateHistoryV2{telemetryStore: telemetryStore}, nil
})
}
func (migration *createRuleStateHistoryV2) Register(migrations *migrate.Migrations) error {
if err := migrations.Register(migration.Up, migration.Down); err != nil {
return err
}
return nil
}
func (migration *createRuleStateHistoryV2) Up(ctx context.Context, db *bun.DB) error {
// Create the local MergeTree table.
if err := migration.telemetryStore.ClickhouseDB().Exec(ctx, `
CREATE TABLE IF NOT EXISTS signoz_analytics.rule_state_history_v2
(
org_id LowCardinality(String),
rule_id String,
rule_name String,
fingerprint UInt64,
labels String,
state LowCardinality(String),
state_changed Bool DEFAULT true,
value Float64,
unix_milli Int64,
overall_state LowCardinality(String),
overall_state_changed Bool
)
ENGINE = MergeTree()
PARTITION BY toDate(unix_milli / 1000)
ORDER BY (org_id, rule_id, fingerprint, unix_milli)
TTL toDate(unix_milli / 1000) + INTERVAL 90 DAY
`); err != nil {
return err
}
// Create the distributed table.
if err := migration.telemetryStore.ClickhouseDB().Exec(ctx, `
CREATE TABLE IF NOT EXISTS signoz_analytics.distributed_rule_state_history_v2
AS signoz_analytics.rule_state_history_v2
ENGINE = Distributed('cluster', 'signoz_analytics', 'rule_state_history_v2', cityHash64(rule_id))
`); err != nil {
return err
}
return nil
}
func (migration *createRuleStateHistoryV2) Down(ctx context.Context, db *bun.DB) error {
if err := migration.telemetryStore.ClickhouseDB().Exec(ctx, `
DROP TABLE IF EXISTS signoz_analytics.distributed_rule_state_history_v2
`); err != nil {
return err
}
if err := migration.telemetryStore.ClickhouseDB().Exec(ctx, `
DROP TABLE IF EXISTS signoz_analytics.rule_state_history_v2
`); err != nil {
return err
}
return nil
}

View File

@@ -0,0 +1,445 @@
package alertmanagertypes
import (
"context"
"encoding/json"
"log/slog"
"time"
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/types"
"github.com/SigNoz/signoz/pkg/valuer"
"github.com/expr-lang/expr"
"github.com/uptrace/bun"
)
var (
ErrCodeInvalidPlannedMaintenancePayload = errors.MustNewCode("invalid_planned_maintenance_payload")
)
type StorablePlannedMaintenance struct {
bun.BaseModel `bun:"table:planned_maintenance_v2"`
types.Identifiable
types.TimeAuditable
types.UserAuditable
Name string `bun:"name,type:text,notnull"`
Description string `bun:"description,type:text"`
Schedule *Schedule `bun:"schedule,type:text,notnull"`
RuleIDs string `bun:"rule_ids,type:text"`
Expression string `bun:"expression,type:text"`
OrgID string `bun:"org_id,type:text"`
}
type GettablePlannedMaintenance struct {
Id string `json:"id"`
Name string `json:"name"`
Description string `json:"description"`
Schedule *Schedule `json:"schedule"`
RuleIDs []string `json:"ruleIds,omitempty"`
Expression string `json:"expression,omitempty"`
CreatedAt time.Time `json:"createdAt"`
CreatedBy string `json:"createdBy"`
UpdatedAt time.Time `json:"updatedAt"`
UpdatedBy string `json:"updatedBy"`
Status string `json:"status"`
Kind string `json:"kind"`
}
func (m *GettablePlannedMaintenance) IsActive(now time.Time) bool {
loc, err := time.LoadLocation(m.Schedule.Timezone)
if err != nil {
return false
}
currentTime := now.In(loc)
// fixed schedule
if !m.Schedule.StartTime.IsZero() && !m.Schedule.EndTime.IsZero() {
startTime := m.Schedule.StartTime.In(loc)
endTime := m.Schedule.EndTime.In(loc)
if currentTime.Equal(startTime) || currentTime.Equal(endTime) ||
(currentTime.After(startTime) && currentTime.Before(endTime)) {
return true
}
}
// recurring schedule
if m.Schedule.Recurrence != nil {
start := m.Schedule.Recurrence.StartTime
// Make sure the recurrence has started
if currentTime.Before(start.In(loc)) {
return false
}
// Check if recurrence has expired
if m.Schedule.Recurrence.EndTime != nil {
endTime := *m.Schedule.Recurrence.EndTime
if !endTime.IsZero() && currentTime.After(endTime.In(loc)) {
return false
}
}
switch m.Schedule.Recurrence.RepeatType {
case RepeatTypeDaily:
return m.checkDaily(currentTime, m.Schedule.Recurrence, loc)
case RepeatTypeWeekly:
return m.checkWeekly(currentTime, m.Schedule.Recurrence, loc)
case RepeatTypeMonthly:
return m.checkMonthly(currentTime, m.Schedule.Recurrence, loc)
}
}
return false
}
// checkDaily rebases the recurrence start to today (or yesterday if needed)
// and returns true if currentTime is within [candidate, candidate+Duration].
func (m *GettablePlannedMaintenance) checkDaily(currentTime time.Time, rec *Recurrence, loc *time.Location) bool {
candidate := time.Date(
currentTime.Year(), currentTime.Month(), currentTime.Day(),
rec.StartTime.Hour(), rec.StartTime.Minute(), 0, 0,
loc,
)
if candidate.After(currentTime) {
candidate = candidate.AddDate(0, 0, -1)
}
return currentTime.Sub(candidate) <= time.Duration(rec.Duration)
}
// checkWeekly finds the most recent allowed occurrence by rebasing the recurrence's
// time-of-day onto the allowed weekday. It does this for each allowed day and returns true
// if the current time falls within the candidate window.
func (m *GettablePlannedMaintenance) checkWeekly(currentTime time.Time, rec *Recurrence, loc *time.Location) bool {
// If no days specified, treat as every day (like daily).
if len(rec.RepeatOn) == 0 {
return m.checkDaily(currentTime, rec, loc)
}
for _, day := range rec.RepeatOn {
allowedDay, ok := RepeatOnAllMap[day]
if !ok {
continue // skip invalid days
}
// Compute the day difference: allowedDay - current weekday.
delta := int(allowedDay) - int(currentTime.Weekday())
// Build a candidate occurrence by rebasing today's date to the allowed weekday.
candidate := time.Date(
currentTime.Year(), currentTime.Month(), currentTime.Day(),
rec.StartTime.Hour(), rec.StartTime.Minute(), 0, 0,
loc,
).AddDate(0, 0, delta)
// If the candidate is in the future, subtract 7 days.
if candidate.After(currentTime) {
candidate = candidate.AddDate(0, 0, -7)
}
if currentTime.Sub(candidate) <= time.Duration(rec.Duration) {
return true
}
}
return false
}
// checkMonthly rebases the candidate occurrence using the recurrence's day-of-month.
// If the candidate for the current month is in the future, it uses the previous month.
func (m *GettablePlannedMaintenance) checkMonthly(currentTime time.Time, rec *Recurrence, loc *time.Location) bool {
refDay := rec.StartTime.Day()
year, month, _ := currentTime.Date()
lastDay := time.Date(year, month+1, 0, 0, 0, 0, 0, loc).Day()
day := refDay
if refDay > lastDay {
day = lastDay
}
candidate := time.Date(year, month, day,
rec.StartTime.Hour(), rec.StartTime.Minute(), rec.StartTime.Second(), rec.StartTime.Nanosecond(),
loc,
)
if candidate.After(currentTime) {
// Use previous month.
candidate = candidate.AddDate(0, -1, 0)
y, m, _ := candidate.Date()
lastDayPrev := time.Date(y, m+1, 0, 0, 0, 0, 0, loc).Day()
if refDay > lastDayPrev {
candidate = time.Date(y, m, lastDayPrev,
rec.StartTime.Hour(), rec.StartTime.Minute(), rec.StartTime.Second(), rec.StartTime.Nanosecond(),
loc,
)
} else {
candidate = time.Date(y, m, refDay,
rec.StartTime.Hour(), rec.StartTime.Minute(), rec.StartTime.Second(), rec.StartTime.Nanosecond(),
loc,
)
}
}
return currentTime.Sub(candidate) <= time.Duration(rec.Duration)
}
// CurrentWindowEndTime returns the end time of the current active maintenance window.
// Returns zero time and false if the maintenance is not currently active.
func (m *GettablePlannedMaintenance) CurrentWindowEndTime(now time.Time) (time.Time, bool) {
loc, err := time.LoadLocation(m.Schedule.Timezone)
if err != nil {
return time.Time{}, false
}
currentTime := now.In(loc)
// fixed schedule
if !m.Schedule.StartTime.IsZero() && !m.Schedule.EndTime.IsZero() {
startTime := m.Schedule.StartTime.In(loc)
endTime := m.Schedule.EndTime.In(loc)
if currentTime.Equal(startTime) || currentTime.Equal(endTime) ||
(currentTime.After(startTime) && currentTime.Before(endTime)) {
return endTime, true
}
}
// recurring schedule
if m.Schedule.Recurrence != nil {
start := m.Schedule.Recurrence.StartTime
if currentTime.Before(start.In(loc)) {
return time.Time{}, false
}
if m.Schedule.Recurrence.EndTime != nil {
endTime := *m.Schedule.Recurrence.EndTime
if !endTime.IsZero() && currentTime.After(endTime.In(loc)) {
return time.Time{}, false
}
}
var candidate time.Time
var active bool
switch m.Schedule.Recurrence.RepeatType {
case RepeatTypeDaily:
candidate, active = m.currentDailyWindowEnd(currentTime, m.Schedule.Recurrence, loc)
case RepeatTypeWeekly:
candidate, active = m.currentWeeklyWindowEnd(currentTime, m.Schedule.Recurrence, loc)
case RepeatTypeMonthly:
candidate, active = m.currentMonthlyWindowEnd(currentTime, m.Schedule.Recurrence, loc)
}
if active {
return candidate, true
}
}
return time.Time{}, false
}
func (m *GettablePlannedMaintenance) currentDailyWindowEnd(currentTime time.Time, rec *Recurrence, loc *time.Location) (time.Time, bool) {
candidate := time.Date(
currentTime.Year(), currentTime.Month(), currentTime.Day(),
rec.StartTime.Hour(), rec.StartTime.Minute(), 0, 0,
loc,
)
if candidate.After(currentTime) {
candidate = candidate.AddDate(0, 0, -1)
}
endTime := candidate.Add(time.Duration(rec.Duration))
if currentTime.Before(endTime) || currentTime.Equal(endTime) {
return endTime, true
}
return time.Time{}, false
}
func (m *GettablePlannedMaintenance) currentWeeklyWindowEnd(currentTime time.Time, rec *Recurrence, loc *time.Location) (time.Time, bool) {
if len(rec.RepeatOn) == 0 {
return m.currentDailyWindowEnd(currentTime, rec, loc)
}
for _, day := range rec.RepeatOn {
allowedDay, ok := RepeatOnAllMap[day]
if !ok {
continue
}
delta := int(allowedDay) - int(currentTime.Weekday())
candidate := time.Date(
currentTime.Year(), currentTime.Month(), currentTime.Day(),
rec.StartTime.Hour(), rec.StartTime.Minute(), 0, 0,
loc,
).AddDate(0, 0, delta)
if candidate.After(currentTime) {
candidate = candidate.AddDate(0, 0, -7)
}
endTime := candidate.Add(time.Duration(rec.Duration))
if currentTime.Before(endTime) || currentTime.Equal(endTime) {
return endTime, true
}
}
return time.Time{}, false
}
func (m *GettablePlannedMaintenance) currentMonthlyWindowEnd(currentTime time.Time, rec *Recurrence, loc *time.Location) (time.Time, bool) {
refDay := rec.StartTime.Day()
year, month, _ := currentTime.Date()
lastDay := time.Date(year, month+1, 0, 0, 0, 0, 0, loc).Day()
day := refDay
if refDay > lastDay {
day = lastDay
}
candidate := time.Date(year, month, day,
rec.StartTime.Hour(), rec.StartTime.Minute(), rec.StartTime.Second(), rec.StartTime.Nanosecond(),
loc,
)
if candidate.After(currentTime) {
candidate = candidate.AddDate(0, -1, 0)
y, m, _ := candidate.Date()
lastDayPrev := time.Date(y, m+1, 0, 0, 0, 0, 0, loc).Day()
if refDay > lastDayPrev {
candidate = time.Date(y, m, lastDayPrev,
rec.StartTime.Hour(), rec.StartTime.Minute(), rec.StartTime.Second(), rec.StartTime.Nanosecond(),
loc,
)
} else {
candidate = time.Date(y, m, refDay,
rec.StartTime.Hour(), rec.StartTime.Minute(), rec.StartTime.Second(), rec.StartTime.Nanosecond(),
loc,
)
}
}
endTime := candidate.Add(time.Duration(rec.Duration))
if currentTime.Before(endTime) || currentTime.Equal(endTime) {
return endTime, true
}
return time.Time{}, false
}
func (m *GettablePlannedMaintenance) IsUpcoming() bool {
loc, err := time.LoadLocation(m.Schedule.Timezone)
if err != nil {
return false
}
now := time.Now().In(loc)
if !m.Schedule.StartTime.IsZero() && !m.Schedule.EndTime.IsZero() {
return now.Before(m.Schedule.StartTime)
}
if m.Schedule.Recurrence != nil {
return now.Before(m.Schedule.Recurrence.StartTime)
}
return false
}
func (m *GettablePlannedMaintenance) IsRecurring() bool {
return m.Schedule.Recurrence != nil
}
func (m *GettablePlannedMaintenance) Validate() error {
if m.Name == "" {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidPlannedMaintenancePayload, "missing name in the payload")
}
if m.Schedule == nil {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidPlannedMaintenancePayload, "missing schedule in the payload")
}
if m.Schedule.Timezone == "" {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidPlannedMaintenancePayload, "missing timezone in the payload")
}
_, err := time.LoadLocation(m.Schedule.Timezone)
if err != nil {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidPlannedMaintenancePayload, "invalid timezone in the payload")
}
if !m.Schedule.StartTime.IsZero() && !m.Schedule.EndTime.IsZero() {
if m.Schedule.StartTime.After(m.Schedule.EndTime) {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidPlannedMaintenancePayload, "start time cannot be after end time")
}
}
if m.Schedule.Recurrence != nil {
if m.Schedule.Recurrence.RepeatType == "" {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidPlannedMaintenancePayload, "missing repeat type in the payload")
}
if m.Schedule.Recurrence.Duration == 0 {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidPlannedMaintenancePayload, "missing duration in the payload")
}
if m.Schedule.Recurrence.EndTime != nil && m.Schedule.Recurrence.EndTime.Before(m.Schedule.Recurrence.StartTime) {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidPlannedMaintenancePayload, "end time cannot be before start time")
}
}
if m.Expression != "" {
if _, err := expr.Compile(m.Expression); err != nil {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidPlannedMaintenancePayload, "invalid expression: %v", err)
}
}
return nil
}
func (m GettablePlannedMaintenance) MarshalJSON() ([]byte, error) {
now := time.Now().In(time.FixedZone(m.Schedule.Timezone, 0))
var status string
if m.IsActive(now) {
status = "active"
} else if m.IsUpcoming() {
status = "upcoming"
} else {
status = "expired"
}
var kind string
if !m.Schedule.StartTime.IsZero() && !m.Schedule.EndTime.IsZero() && m.Schedule.EndTime.After(m.Schedule.StartTime) {
kind = "fixed"
} else {
kind = "recurring"
}
return json.Marshal(struct {
Id string `json:"id"`
Name string `json:"name"`
Description string `json:"description"`
Schedule *Schedule `json:"schedule"`
RuleIDs []string `json:"ruleIds,omitempty"`
Expression string `json:"expression,omitempty"`
CreatedAt time.Time `json:"createdAt"`
CreatedBy string `json:"createdBy"`
UpdatedAt time.Time `json:"updatedAt"`
UpdatedBy string `json:"updatedBy"`
Status string `json:"status"`
Kind string `json:"kind"`
}{
Id: m.Id,
Name: m.Name,
Description: m.Description,
Schedule: m.Schedule,
RuleIDs: m.RuleIDs,
Expression: m.Expression,
CreatedAt: m.CreatedAt,
CreatedBy: m.CreatedBy,
UpdatedAt: m.UpdatedAt,
UpdatedBy: m.UpdatedBy,
Status: status,
Kind: kind,
})
}
// ConvertStorableToGettable converts a StorablePlannedMaintenance to GettablePlannedMaintenance.
func ConvertStorableToGettable(s *StorablePlannedMaintenance) *GettablePlannedMaintenance {
var ruleIDs []string
if s.RuleIDs != "" {
if err := json.Unmarshal([]byte(s.RuleIDs), &ruleIDs); err != nil {
slog.Error("failed to unmarshal rule_ids from DB", "error", err, "raw", s.RuleIDs)
}
}
return &GettablePlannedMaintenance{
Id: s.ID.StringValue(),
Name: s.Name,
Description: s.Description,
Schedule: s.Schedule,
RuleIDs: ruleIDs,
Expression: s.Expression,
CreatedAt: s.CreatedAt,
UpdatedAt: s.UpdatedAt,
CreatedBy: s.CreatedBy,
UpdatedBy: s.UpdatedBy,
}
}
type MaintenanceStore interface {
CreatePlannedMaintenance(context.Context, GettablePlannedMaintenance) (valuer.UUID, error)
DeletePlannedMaintenance(context.Context, valuer.UUID) error
GetPlannedMaintenanceByID(context.Context, valuer.UUID) (*GettablePlannedMaintenance, error)
EditPlannedMaintenance(context.Context, GettablePlannedMaintenance, valuer.UUID) error
GetAllPlannedMaintenance(context.Context, string) ([]*GettablePlannedMaintenance, error)
}

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,86 @@
package alertmanagertypes
import (
"database/sql/driver"
"encoding/json"
"time"
"github.com/SigNoz/signoz/pkg/errors"
)
type RepeatType string
const (
RepeatTypeDaily RepeatType = "daily"
RepeatTypeWeekly RepeatType = "weekly"
RepeatTypeMonthly RepeatType = "monthly"
)
type RepeatOn string
const (
RepeatOnSunday RepeatOn = "sunday"
RepeatOnMonday RepeatOn = "monday"
RepeatOnTuesday RepeatOn = "tuesday"
RepeatOnWednesday RepeatOn = "wednesday"
RepeatOnThursday RepeatOn = "thursday"
RepeatOnFriday RepeatOn = "friday"
RepeatOnSaturday RepeatOn = "saturday"
)
var RepeatOnAllMap = map[RepeatOn]time.Weekday{
RepeatOnSunday: time.Sunday,
RepeatOnMonday: time.Monday,
RepeatOnTuesday: time.Tuesday,
RepeatOnWednesday: time.Wednesday,
RepeatOnThursday: time.Thursday,
RepeatOnFriday: time.Friday,
RepeatOnSaturday: time.Saturday,
}
type Duration time.Duration
func (d Duration) MarshalJSON() ([]byte, error) {
return json.Marshal(time.Duration(d).String())
}
func (d *Duration) UnmarshalJSON(b []byte) error {
var v interface{}
if err := json.Unmarshal(b, &v); err != nil {
return err
}
switch value := v.(type) {
case float64:
*d = Duration(time.Duration(value))
return nil
case string:
tmp, err := time.ParseDuration(value)
if err != nil {
return err
}
*d = Duration(tmp)
return nil
default:
return errors.New(errors.TypeInvalidInput, errors.CodeInvalidInput, "invalid duration")
}
}
type Recurrence struct {
StartTime time.Time `json:"startTime"`
EndTime *time.Time `json:"endTime,omitempty"`
Duration Duration `json:"duration"`
RepeatType RepeatType `json:"repeatType"`
RepeatOn []RepeatOn `json:"repeatOn"`
}
func (r *Recurrence) Scan(src interface{}) error {
if data, ok := src.([]byte); ok {
return json.Unmarshal(data, r)
}
return nil
}
func (r *Recurrence) Value() (driver.Value, error) {
return json.Marshal(r)
}

View File

@@ -0,0 +1,132 @@
package alertmanagertypes
import (
"database/sql/driver"
"encoding/json"
"time"
)
type Schedule struct {
Timezone string `json:"timezone"`
StartTime time.Time `json:"startTime,omitempty"`
EndTime time.Time `json:"endTime,omitempty"`
Recurrence *Recurrence `json:"recurrence"`
}
func (s *Schedule) Scan(src interface{}) error {
if data, ok := src.([]byte); ok {
return json.Unmarshal(data, s)
}
return nil
}
func (s *Schedule) Value() (driver.Value, error) {
return json.Marshal(s)
}
func (s Schedule) MarshalJSON() ([]byte, error) {
loc, err := time.LoadLocation(s.Timezone)
if err != nil {
return nil, err
}
var startTime, endTime time.Time
if !s.StartTime.IsZero() {
startTime = time.Date(s.StartTime.Year(), s.StartTime.Month(), s.StartTime.Day(), s.StartTime.Hour(), s.StartTime.Minute(), s.StartTime.Second(), s.StartTime.Nanosecond(), loc)
}
if !s.EndTime.IsZero() {
endTime = time.Date(s.EndTime.Year(), s.EndTime.Month(), s.EndTime.Day(), s.EndTime.Hour(), s.EndTime.Minute(), s.EndTime.Second(), s.EndTime.Nanosecond(), loc)
}
var recurrence *Recurrence
if s.Recurrence != nil {
recStartTime := time.Date(s.Recurrence.StartTime.Year(), s.Recurrence.StartTime.Month(), s.Recurrence.StartTime.Day(), s.Recurrence.StartTime.Hour(), s.Recurrence.StartTime.Minute(), s.Recurrence.StartTime.Second(), s.Recurrence.StartTime.Nanosecond(), loc)
var recEndTime *time.Time
if s.Recurrence.EndTime != nil {
end := time.Date(s.Recurrence.EndTime.Year(), s.Recurrence.EndTime.Month(), s.Recurrence.EndTime.Day(), s.Recurrence.EndTime.Hour(), s.Recurrence.EndTime.Minute(), s.Recurrence.EndTime.Second(), s.Recurrence.EndTime.Nanosecond(), loc)
recEndTime = &end
}
recurrence = &Recurrence{
StartTime: recStartTime,
EndTime: recEndTime,
Duration: s.Recurrence.Duration,
RepeatType: s.Recurrence.RepeatType,
RepeatOn: s.Recurrence.RepeatOn,
}
}
return json.Marshal(&struct {
Timezone string `json:"timezone"`
StartTime string `json:"startTime"`
EndTime string `json:"endTime"`
Recurrence *Recurrence `json:"recurrence,omitempty"`
}{
Timezone: s.Timezone,
StartTime: startTime.Format(time.RFC3339),
EndTime: endTime.Format(time.RFC3339),
Recurrence: recurrence,
})
}
func (s *Schedule) UnmarshalJSON(data []byte) error {
aux := &struct {
Timezone string `json:"timezone"`
StartTime string `json:"startTime"`
EndTime string `json:"endTime"`
Recurrence *Recurrence `json:"recurrence,omitempty"`
}{}
if err := json.Unmarshal(data, aux); err != nil {
return err
}
loc, err := time.LoadLocation(aux.Timezone)
if err != nil {
return err
}
var startTime time.Time
if aux.StartTime != "" {
startTime, err = time.Parse(time.RFC3339, aux.StartTime)
if err != nil {
return err
}
s.StartTime = time.Date(startTime.Year(), startTime.Month(), startTime.Day(), startTime.Hour(), startTime.Minute(), startTime.Second(), startTime.Nanosecond(), loc)
}
var endTime time.Time
if aux.EndTime != "" {
endTime, err = time.Parse(time.RFC3339, aux.EndTime)
if err != nil {
return err
}
s.EndTime = time.Date(endTime.Year(), endTime.Month(), endTime.Day(), endTime.Hour(), endTime.Minute(), endTime.Second(), endTime.Nanosecond(), loc)
}
s.Timezone = aux.Timezone
if aux.Recurrence != nil {
recStartTime, err := time.Parse(time.RFC3339, aux.Recurrence.StartTime.Format(time.RFC3339))
if err != nil {
return err
}
var recEndTime *time.Time
if aux.Recurrence.EndTime != nil {
end, err := time.Parse(time.RFC3339, aux.Recurrence.EndTime.Format(time.RFC3339))
if err != nil {
return err
}
endConverted := time.Date(end.Year(), end.Month(), end.Day(), end.Hour(), end.Minute(), end.Second(), end.Nanosecond(), loc)
recEndTime = &endConverted
}
s.Recurrence = &Recurrence{
StartTime: time.Date(recStartTime.Year(), recStartTime.Month(), recStartTime.Day(), recStartTime.Hour(), recStartTime.Minute(), recStartTime.Second(), recStartTime.Nanosecond(), loc),
EndTime: recEndTime,
Duration: aux.Recurrence.Duration,
RepeatType: aux.Recurrence.RepeatType,
RepeatOn: aux.Recurrence.RepeatOn,
}
}
return nil
}

View File

@@ -0,0 +1,137 @@
package alertmanagertypes
import (
"context"
"github.com/SigNoz/signoz/pkg/errors"
"github.com/SigNoz/signoz/pkg/valuer"
)
var (
ErrCodeInvalidStateHistoryQuery = errors.MustNewCode("invalid_state_history_query")
)
// AlertState represents the state of an alert series (firing, inactive, muted, no_data)
// or the overall state of a rule (firing, inactive).
type AlertState struct {
valuer.String
}
var (
AlertStateFiring = AlertState{valuer.NewString("firing")}
AlertStateInactive = AlertState{valuer.NewString("inactive")}
AlertStateMuted = AlertState{valuer.NewString("muted")}
AlertStateNoData = AlertState{valuer.NewString("no_data")}
)
// SortOrder represents the sort direction for query results.
type SortOrder struct {
valuer.String
}
var (
SortOrderAsc = SortOrder{valuer.NewString("asc")}
SortOrderDesc = SortOrder{valuer.NewString("desc")}
)
// RuleStateHistory represents a single state transition entry stored in ClickHouse.
// Only transitions are recorded, not every evaluation.
type RuleStateHistory struct {
OrgID string `json:"orgId"`
RuleID string `json:"ruleId"`
RuleName string `json:"ruleName"`
OverallState string `json:"overallState"` // aggregate rule state: "firing" if any series fires
OverallStateChanged bool `json:"overallStateChanged"` // true if this entry changed the overall state
State string `json:"state"` // per-series state: firing, inactive, muted, no_data
StateChanged bool `json:"stateChanged"` // always true in v2 (only transitions stored)
UnixMilli int64 `json:"unixMilli"`
Labels string `json:"labels"` // JSON-encoded label set
Fingerprint uint64 `json:"fingerprint"` // hash of the full label set
Value float64 `json:"value"`
}
// QueryRuleStateHistory is the request body for all v2 state history API endpoints.
type QueryRuleStateHistory struct {
Start int64 `json:"start"` // unix millis, required
End int64 `json:"end"` // unix millis, required
State AlertState `json:"state"` // optional filter: firing, inactive, muted
Offset int64 `json:"offset"`
Limit int64 `json:"limit"`
Order SortOrder `json:"order"`
}
func (q *QueryRuleStateHistory) Validate() error {
if q.Start == 0 || q.End == 0 {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidStateHistoryQuery, "start and end are required")
}
if q.Offset < 0 || q.Limit < 0 {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidStateHistoryQuery, "offset and limit must be greater than or equal to 0")
}
if q.Order.StringValue() != SortOrderAsc.StringValue() && q.Order.StringValue() != SortOrderDesc.StringValue() {
return errors.Newf(errors.TypeInvalidInput, ErrCodeInvalidStateHistoryQuery, "order must be asc or desc")
}
return nil
}
// RuleStateTimeline is the paginated response for the timeline endpoint.
type RuleStateTimeline struct {
Items []RuleStateHistory `json:"items"`
Total uint64 `json:"total"`
Labels map[string][]string `json:"labels"` // distinct label keys/values for filter UI
}
// RuleStateHistoryContributor is an alert series ranked by firing frequency.
type RuleStateHistoryContributor struct {
Fingerprint uint64 `json:"fingerprint"`
Labels string `json:"labels"` // JSON-encoded label set
Count uint64 `json:"count"`
}
// RuleStateTransition represents a contiguous time period during which a rule
// was in a particular overall state (firing or inactive).
type RuleStateTransition struct {
State AlertState `json:"state"`
Start int64 `json:"start"`
End int64 `json:"end"`
}
// RuleStats compares trigger counts and avg resolution times between the current
// time period and a previous period of equal length.
type RuleStats struct {
TotalCurrentTriggers uint64 `json:"totalCurrentTriggers"`
TotalPastTriggers uint64 `json:"totalPastTriggers"`
CurrentTriggersSeries *Series `json:"currentTriggersSeries"`
PastTriggersSeries *Series `json:"pastTriggersSeries"`
CurrentAvgResolutionTime float64 `json:"currentAvgResolutionTime"`
PastAvgResolutionTime float64 `json:"pastAvgResolutionTime"`
CurrentAvgResolutionTimeSeries *Series `json:"currentAvgResolutionTimeSeries"`
PastAvgResolutionTimeSeries *Series `json:"pastAvgResolutionTimeSeries"`
}
type Series struct {
Labels map[string]string `json:"labels"`
Points []Point `json:"values"`
}
type Point struct {
Timestamp int64 `json:"timestamp"`
Value float64 `json:"value"`
}
// StateHistoryStore provides read and write access to rule state history in ClickHouse.
type StateHistoryStore interface {
WriteRuleStateHistory(ctx context.Context, entries []RuleStateHistory) error
// GetLastSavedRuleStateHistory returns the most recent transition per fingerprint,
// used to restore in-memory state after restart.
GetLastSavedRuleStateHistory(ctx context.Context, ruleID string) ([]RuleStateHistory, error)
GetRuleStateHistoryTimeline(ctx context.Context, orgID string, ruleID string, params *QueryRuleStateHistory) (*RuleStateTimeline, error)
GetRuleStateHistoryTopContributors(ctx context.Context, orgID string, ruleID string, params *QueryRuleStateHistory) ([]RuleStateHistoryContributor, error)
// GetOverallStateTransitions returns firing/inactive periods with gap-filling.
GetOverallStateTransitions(ctx context.Context, orgID string, ruleID string, params *QueryRuleStateHistory) ([]RuleStateTransition, error)
GetTotalTriggers(ctx context.Context, orgID string, ruleID string, params *QueryRuleStateHistory) (uint64, error)
GetTriggersByInterval(ctx context.Context, orgID string, ruleID string, params *QueryRuleStateHistory) (*Series, error)
// GetAvgResolutionTime returns avg seconds between firing and next resolution.
GetAvgResolutionTime(ctx context.Context, orgID string, ruleID string, params *QueryRuleStateHistory) (float64, error)
GetAvgResolutionTimeByInterval(ctx context.Context, orgID string, ruleID string, params *QueryRuleStateHistory) (*Series, error)
}

View File

@@ -34,9 +34,9 @@ var (
// StatsRequest represents the payload accepted by the metrics stats endpoint.
type StatsRequest struct {
Filter *qbtypes.Filter `json:"filter,omitempty"`
Start int64 `json:"start" required:"true"`
End int64 `json:"end" required:"true"`
Limit int `json:"limit" required:"true"`
Start int64 `json:"start"`
End int64 `json:"end"`
Limit int `json:"limit"`
Offset int `json:"offset"`
OrderBy *qbtypes.OrderBy `json:"orderBy,omitempty"`
}
@@ -96,26 +96,26 @@ func (req *StatsRequest) UnmarshalJSON(data []byte) error {
// Stat represents the summary information returned per metric.
type Stat struct {
MetricName string `json:"metricName" required:"true"`
Description string `json:"description" required:"true"`
MetricType metrictypes.Type `json:"type" required:"true" enum:"gauge,sum,histogram,summary,exponentialhistogram"`
MetricUnit string `json:"unit" required:"true"`
TimeSeries uint64 `json:"timeseries" required:"true"`
Samples uint64 `json:"samples" required:"true"`
MetricName string `json:"metricName"`
Description string `json:"description"`
MetricType metrictypes.Type `json:"type"`
MetricUnit string `json:"unit"`
TimeSeries uint64 `json:"timeseries"`
Samples uint64 `json:"samples"`
}
// StatsResponse represents the aggregated metrics statistics.
type StatsResponse struct {
Metrics []Stat `json:"metrics" required:"true" nullable:"true"`
Total uint64 `json:"total" required:"true"`
Metrics []Stat `json:"metrics"`
Total uint64 `json:"total"`
}
type MetricMetadata struct {
Description string `json:"description" required:"true"`
MetricType metrictypes.Type `json:"type" required:"true" enum:"gauge,sum,histogram,summary,exponentialhistogram"`
MetricUnit string `json:"unit" required:"true"`
Temporality metrictypes.Temporality `json:"temporality" required:"true" enum:"delta,cumulative,unspecified"`
IsMonotonic bool `json:"isMonotonic" required:"true"`
Description string `json:"description"`
MetricType metrictypes.Type `json:"type"`
MetricUnit string `json:"unit"`
Temporality metrictypes.Temporality `json:"temporality"`
IsMonotonic bool `json:"isMonotonic"`
}
// MarshalBinary implements cachetypes.Cacheable interface
@@ -130,21 +130,21 @@ func (m *MetricMetadata) UnmarshalBinary(data []byte) error {
// UpdateMetricMetadataRequest represents the payload for updating metric metadata.
type UpdateMetricMetadataRequest struct {
MetricName string `json:"metricName" required:"true"`
Type metrictypes.Type `json:"type" required:"true" enum:"gauge,sum,histogram,summary,exponentialhistogram"`
Description string `json:"description" required:"true"`
Unit string `json:"unit" required:"true"`
Temporality metrictypes.Temporality `json:"temporality" required:"true" enum:"delta,cumulative,unspecified"`
IsMonotonic bool `json:"isMonotonic" required:"true"`
MetricName string `json:"metricName"`
Type metrictypes.Type `json:"type"`
Description string `json:"description"`
Unit string `json:"unit"`
Temporality metrictypes.Temporality `json:"temporality"`
IsMonotonic bool `json:"isMonotonic"`
}
// TreemapRequest represents the payload for the metrics treemap endpoint.
type TreemapRequest struct {
Filter *qbtypes.Filter `json:"filter,omitempty"`
Start int64 `json:"start" required:"true"`
End int64 `json:"end" required:"true"`
Limit int `json:"limit" required:"true"`
Mode TreemapMode `json:"mode" required:"true" enum:"timeseries,samples"`
Start int64 `json:"start"`
End int64 `json:"end"`
Limit int `json:"limit"`
Mode TreemapMode `json:"mode"`
}
// Validate enforces basic constraints on TreemapRequest.
@@ -210,52 +210,52 @@ func (req *TreemapRequest) UnmarshalJSON(data []byte) error {
// TreemapEntry represents each node in the treemap response.
type TreemapEntry struct {
MetricName string `json:"metricName" required:"true"`
Percentage float64 `json:"percentage" required:"true"`
TotalValue uint64 `json:"totalValue" required:"true"`
MetricName string `json:"metricName"`
Percentage float64 `json:"percentage"`
TotalValue uint64 `json:"totalValue"`
}
// TreemapResponse is the output structure for the treemap endpoint.
type TreemapResponse struct {
TimeSeries []TreemapEntry `json:"timeseries" required:"true" nullable:"true"`
Samples []TreemapEntry `json:"samples" required:"true" nullable:"true"`
TimeSeries []TreemapEntry `json:"timeseries"`
Samples []TreemapEntry `json:"samples"`
}
// MetricAlert represents an alert associated with a metric.
type MetricAlert struct {
AlertName string `json:"alertName" required:"true"`
AlertID string `json:"alertId" required:"true"`
AlertName string `json:"alertName"`
AlertID string `json:"alertId"`
}
// MetricAlertsResponse represents the response for metric alerts endpoint.
type MetricAlertsResponse struct {
Alerts []MetricAlert `json:"alerts" required:"true" nullable:"true"`
Alerts []MetricAlert `json:"alerts"`
}
// MetricDashboard represents a dashboard/widget referencing a metric.
type MetricDashboard struct {
DashboardName string `json:"dashboardName" required:"true"`
DashboardID string `json:"dashboardId" required:"true"`
WidgetID string `json:"widgetId" required:"true"`
WidgetName string `json:"widgetName" required:"true"`
DashboardName string `json:"dashboardName"`
DashboardID string `json:"dashboardId"`
WidgetID string `json:"widgetId"`
WidgetName string `json:"widgetName"`
}
// MetricDashboardsResponse represents the response for metric dashboards endpoint.
type MetricDashboardsResponse struct {
Dashboards []MetricDashboard `json:"dashboards" required:"true" nullable:"true"`
Dashboards []MetricDashboard `json:"dashboards"`
}
// MetricHighlightsResponse is the output structure for the metric highlights endpoint.
type MetricHighlightsResponse struct {
DataPoints uint64 `json:"dataPoints" required:"true"`
LastReceived uint64 `json:"lastReceived" required:"true"`
TotalTimeSeries uint64 `json:"totalTimeSeries" required:"true"`
ActiveTimeSeries uint64 `json:"activeTimeSeries" required:"true"`
DataPoints uint64 `json:"dataPoints"`
LastReceived uint64 `json:"lastReceived"`
TotalTimeSeries uint64 `json:"totalTimeSeries"`
ActiveTimeSeries uint64 `json:"activeTimeSeries"`
}
// MetricAttributesRequest represents the payload for the metric attributes endpoint.
type MetricAttributesRequest struct {
MetricName string `json:"metricName" required:"true"`
MetricName string `json:"metricName"`
Start *int64 `json:"start,omitempty"`
End *int64 `json:"end,omitempty"`
}
@@ -292,17 +292,17 @@ func (req *MetricAttributesRequest) UnmarshalJSON(data []byte) error {
// MetricAttribute represents a single attribute with its values and count.
type MetricAttribute struct {
Key string `json:"key" required:"true"`
Values []string `json:"values" required:"true" nullable:"true"`
ValueCount uint64 `json:"valueCount" required:"true"`
Key string `json:"key"`
Values []string `json:"values"`
ValueCount uint64 `json:"valueCount"`
}
// MetricAttributesResponse is the output structure for the metric attributes endpoint.
type MetricAttributesResponse struct {
Attributes []MetricAttribute `json:"attributes" required:"true" nullable:"true"`
TotalKeys int64 `json:"totalKeys" required:"true"`
Attributes []MetricAttribute `json:"attributes"`
TotalKeys int64 `json:"totalKeys"`
}
type MetricNameParams struct {
MetricName string `query:"metricName" required:"true"`
MetricName string `query:"metricName"`
}

View File

@@ -130,7 +130,7 @@ var (
SumType = Type{valuer.NewString("sum")}
HistogramType = Type{valuer.NewString("histogram")}
SummaryType = Type{valuer.NewString("summary")}
ExpHistogramType = Type{valuer.NewString("exponentialhistogram")}
ExpHistogramType = Type{valuer.NewString("exponential_histogram")}
UnspecifiedType = Type{valuer.NewString("")}
)