mirror of
https://github.com/SigNoz/signoz.git
synced 2026-04-27 14:10:30 +01:00
Compare commits
123 Commits
chore/dril
...
infraM/v2_
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
3bec0df0ad | ||
|
|
520e92049c | ||
|
|
92d297ac9d | ||
|
|
eff29aefba | ||
|
|
ca73453c9e | ||
|
|
1d836d674f | ||
|
|
83724b0cde | ||
|
|
3050e37ec7 | ||
|
|
bdbaa32485 | ||
|
|
55b2215025 | ||
|
|
e90378e618 | ||
|
|
4592b78f48 | ||
|
|
d81c99feae | ||
|
|
65a456ff9e | ||
|
|
264577b673 | ||
|
|
b35c6676f9 | ||
|
|
c78c9a42db | ||
|
|
1095caa123 | ||
|
|
9043b49762 | ||
|
|
d4084a7494 | ||
|
|
27c564b3bf | ||
|
|
f02c491828 | ||
|
|
3d53b8f77f | ||
|
|
dffe94fec4 | ||
|
|
b7d4f18aae | ||
|
|
9ad2ec428a | ||
|
|
c9360fcf13 | ||
|
|
b5ab45db20 | ||
|
|
08f76aca78 | ||
|
|
983d4fe4f2 | ||
|
|
833af794c3 | ||
|
|
21b51d1fcc | ||
|
|
56f22682c8 | ||
|
|
9c8359940c | ||
|
|
4050880275 | ||
|
|
5e775f64f2 | ||
|
|
0189f23f46 | ||
|
|
49a36d4e3d | ||
|
|
9407d658ab | ||
|
|
5035712485 | ||
|
|
bab17c3615 | ||
|
|
37b44f4db9 | ||
|
|
99dd6e5f1e | ||
|
|
9c7131fa6a | ||
|
|
ad889a2e1d | ||
|
|
a4f6d0cbf5 | ||
|
|
589bed7c16 | ||
|
|
93843a1f48 | ||
|
|
88c43108fc | ||
|
|
ed4cf540e8 | ||
|
|
9e2dfa9033 | ||
|
|
d98d5d68ee | ||
|
|
2cb1c3b73b | ||
|
|
ae7ca497ad | ||
|
|
a579916961 | ||
|
|
4a16d56abf | ||
|
|
642b5ac3f0 | ||
|
|
a12112619c | ||
|
|
014785f1bc | ||
|
|
58ee797b10 | ||
|
|
82d236742f | ||
|
|
397e1ad5be | ||
|
|
8d6b25ca9b | ||
|
|
5fa6bd8b8d | ||
|
|
bd9977483b | ||
|
|
50fbdfeeef | ||
|
|
e2b1b73e87 | ||
|
|
cb9f3fd3e5 | ||
|
|
232acc343d | ||
|
|
2025afdccc | ||
|
|
d2f4d4af93 | ||
|
|
47ff7bbb8e | ||
|
|
724071c5dc | ||
|
|
4d24979358 | ||
|
|
042943b10a | ||
|
|
48a9be7ec8 | ||
|
|
a9504b2120 | ||
|
|
8755887c4a | ||
|
|
4cb4662b3a | ||
|
|
e6900dabc8 | ||
|
|
c1ba389b63 | ||
|
|
3a1f40234f | ||
|
|
2e4891fa63 | ||
|
|
04ebc0bec7 | ||
|
|
271f9b81ed | ||
|
|
6fa815c294 | ||
|
|
63ec518efb | ||
|
|
c4ca20dd90 | ||
|
|
e56cc4222b | ||
|
|
07d2944d7c | ||
|
|
dea01ae36a | ||
|
|
62ea5b54e2 | ||
|
|
e549a7e42f | ||
|
|
90e2ebb11f | ||
|
|
61baa1be7a | ||
|
|
b946fa665f | ||
|
|
2e049556e4 | ||
|
|
492a5e70d7 | ||
|
|
ba1f2771e8 | ||
|
|
7458fb4855 | ||
|
|
5f55f3938b | ||
|
|
3e8102485c | ||
|
|
861c682ea5 | ||
|
|
c8e5895dff | ||
|
|
82d72e7edb | ||
|
|
a3f8ecaaf1 | ||
|
|
19aada656c | ||
|
|
b21bb4280f | ||
|
|
bc0a4fdb5c | ||
|
|
37fb0e9254 | ||
|
|
aecfa1a174 | ||
|
|
b869d23d94 | ||
|
|
6ee3d44f76 | ||
|
|
462e554107 | ||
|
|
66afa73e6f | ||
|
|
54c604bcf4 | ||
|
|
c1be02ba54 | ||
|
|
d3c7ba8f45 | ||
|
|
039c4a0496 | ||
|
|
51a94b6bbc | ||
|
|
bbfbb94f52 | ||
|
|
d1eb9ef16f | ||
|
|
3db00f8bc3 |
@@ -2474,6 +2474,164 @@ components:
|
||||
- requiredMetricsCheck
|
||||
- endTimeBeforeRetention
|
||||
type: object
|
||||
InframonitoringtypesNodeCondition:
|
||||
enum:
|
||||
- ready
|
||||
- not_ready
|
||||
- ""
|
||||
type: string
|
||||
InframonitoringtypesNodeRecord:
|
||||
properties:
|
||||
condition:
|
||||
$ref: '#/components/schemas/InframonitoringtypesNodeCondition'
|
||||
meta:
|
||||
additionalProperties: {}
|
||||
nullable: true
|
||||
type: object
|
||||
nodeCPU:
|
||||
format: double
|
||||
type: number
|
||||
nodeCPUAllocatable:
|
||||
format: double
|
||||
type: number
|
||||
nodeMemory:
|
||||
format: double
|
||||
type: number
|
||||
nodeMemoryAllocatable:
|
||||
format: double
|
||||
type: number
|
||||
nodeName:
|
||||
type: string
|
||||
notReadyNodeCount:
|
||||
type: integer
|
||||
readyNodeCount:
|
||||
type: integer
|
||||
required:
|
||||
- nodeName
|
||||
- condition
|
||||
- readyNodeCount
|
||||
- notReadyNodeCount
|
||||
- nodeCPU
|
||||
- nodeCPUAllocatable
|
||||
- nodeMemory
|
||||
- nodeMemoryAllocatable
|
||||
- meta
|
||||
type: object
|
||||
InframonitoringtypesNodes:
|
||||
properties:
|
||||
endTimeBeforeRetention:
|
||||
type: boolean
|
||||
records:
|
||||
items:
|
||||
$ref: '#/components/schemas/InframonitoringtypesNodeRecord'
|
||||
nullable: true
|
||||
type: array
|
||||
requiredMetricsCheck:
|
||||
$ref: '#/components/schemas/InframonitoringtypesRequiredMetricsCheck'
|
||||
total:
|
||||
type: integer
|
||||
type:
|
||||
$ref: '#/components/schemas/InframonitoringtypesResponseType'
|
||||
warning:
|
||||
$ref: '#/components/schemas/Querybuildertypesv5QueryWarnData'
|
||||
required:
|
||||
- type
|
||||
- records
|
||||
- total
|
||||
- requiredMetricsCheck
|
||||
- endTimeBeforeRetention
|
||||
type: object
|
||||
InframonitoringtypesPodPhase:
|
||||
enum:
|
||||
- pending
|
||||
- running
|
||||
- succeeded
|
||||
- failed
|
||||
- unknown
|
||||
- ""
|
||||
type: string
|
||||
InframonitoringtypesPodRecord:
|
||||
properties:
|
||||
failedPodCount:
|
||||
type: integer
|
||||
meta:
|
||||
additionalProperties: {}
|
||||
nullable: true
|
||||
type: object
|
||||
pendingPodCount:
|
||||
type: integer
|
||||
podAge:
|
||||
format: int64
|
||||
type: integer
|
||||
podCPU:
|
||||
format: double
|
||||
type: number
|
||||
podCPULimit:
|
||||
format: double
|
||||
type: number
|
||||
podCPURequest:
|
||||
format: double
|
||||
type: number
|
||||
podMemory:
|
||||
format: double
|
||||
type: number
|
||||
podMemoryLimit:
|
||||
format: double
|
||||
type: number
|
||||
podMemoryRequest:
|
||||
format: double
|
||||
type: number
|
||||
podPhase:
|
||||
$ref: '#/components/schemas/InframonitoringtypesPodPhase'
|
||||
podUID:
|
||||
type: string
|
||||
runningPodCount:
|
||||
type: integer
|
||||
succeededPodCount:
|
||||
type: integer
|
||||
unknownPodCount:
|
||||
type: integer
|
||||
required:
|
||||
- podUID
|
||||
- podCPU
|
||||
- podCPURequest
|
||||
- podCPULimit
|
||||
- podMemory
|
||||
- podMemoryRequest
|
||||
- podMemoryLimit
|
||||
- podPhase
|
||||
- pendingPodCount
|
||||
- runningPodCount
|
||||
- succeededPodCount
|
||||
- failedPodCount
|
||||
- unknownPodCount
|
||||
- podAge
|
||||
- meta
|
||||
type: object
|
||||
InframonitoringtypesPods:
|
||||
properties:
|
||||
endTimeBeforeRetention:
|
||||
type: boolean
|
||||
records:
|
||||
items:
|
||||
$ref: '#/components/schemas/InframonitoringtypesPodRecord'
|
||||
nullable: true
|
||||
type: array
|
||||
requiredMetricsCheck:
|
||||
$ref: '#/components/schemas/InframonitoringtypesRequiredMetricsCheck'
|
||||
total:
|
||||
type: integer
|
||||
type:
|
||||
$ref: '#/components/schemas/InframonitoringtypesResponseType'
|
||||
warning:
|
||||
$ref: '#/components/schemas/Querybuildertypesv5QueryWarnData'
|
||||
required:
|
||||
- type
|
||||
- records
|
||||
- total
|
||||
- requiredMetricsCheck
|
||||
- endTimeBeforeRetention
|
||||
type: object
|
||||
InframonitoringtypesPostableHosts:
|
||||
properties:
|
||||
end:
|
||||
@@ -2500,6 +2658,58 @@ components:
|
||||
- end
|
||||
- limit
|
||||
type: object
|
||||
InframonitoringtypesPostableNodes:
|
||||
properties:
|
||||
end:
|
||||
format: int64
|
||||
type: integer
|
||||
filter:
|
||||
$ref: '#/components/schemas/Querybuildertypesv5Filter'
|
||||
groupBy:
|
||||
items:
|
||||
$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'
|
||||
nullable: true
|
||||
type: array
|
||||
limit:
|
||||
type: integer
|
||||
offset:
|
||||
type: integer
|
||||
orderBy:
|
||||
$ref: '#/components/schemas/Querybuildertypesv5OrderBy'
|
||||
start:
|
||||
format: int64
|
||||
type: integer
|
||||
required:
|
||||
- start
|
||||
- end
|
||||
- limit
|
||||
type: object
|
||||
InframonitoringtypesPostablePods:
|
||||
properties:
|
||||
end:
|
||||
format: int64
|
||||
type: integer
|
||||
filter:
|
||||
$ref: '#/components/schemas/Querybuildertypesv5Filter'
|
||||
groupBy:
|
||||
items:
|
||||
$ref: '#/components/schemas/Querybuildertypesv5GroupByKey'
|
||||
nullable: true
|
||||
type: array
|
||||
limit:
|
||||
type: integer
|
||||
offset:
|
||||
type: integer
|
||||
orderBy:
|
||||
$ref: '#/components/schemas/Querybuildertypesv5OrderBy'
|
||||
start:
|
||||
format: int64
|
||||
type: integer
|
||||
required:
|
||||
- start
|
||||
- end
|
||||
- limit
|
||||
type: object
|
||||
InframonitoringtypesRequiredMetricsCheck:
|
||||
properties:
|
||||
missingMetrics:
|
||||
@@ -10227,7 +10437,9 @@ paths:
|
||||
five metrics, and pagination via offset/limit. The response type is ''list''
|
||||
for the default host.name grouping or ''grouped_list'' for custom groupBy
|
||||
keys. Also reports missing required metrics and whether the requested time
|
||||
range falls before the data retention boundary.'
|
||||
range falls before the data retention boundary. Numeric metric fields (cpu,
|
||||
memory, wait, load15, diskUsage) return -1 as a sentinel when no data is available
|
||||
for that field; frontends should render ''—'' rather than the literal value.'
|
||||
operationId: ListHosts
|
||||
requestBody:
|
||||
content:
|
||||
@@ -10281,6 +10493,151 @@ paths:
|
||||
summary: List Hosts for Infra Monitoring
|
||||
tags:
|
||||
- inframonitoring
|
||||
/api/v2/infra_monitoring/nodes:
|
||||
post:
|
||||
deprecated: false
|
||||
description: 'Returns a paginated list of Kubernetes nodes with key metrics:
|
||||
CPU usage, CPU allocatable, memory working set, memory allocatable, and per-group
|
||||
readyNodeCount / notReadyNodeCount derived from each node''s latest k8s.node.condition_ready
|
||||
value in the window. Each node includes metadata attributes (k8s.node.uid,
|
||||
k8s.cluster.name). The response type is ''list'' for the default k8s.node.name
|
||||
grouping (each row is one node with its current condition string: ready /
|
||||
not_ready / '''') or ''grouped_list'' for custom groupBy keys (each row aggregates
|
||||
nodes in the group with readyNodeCount and notReadyNodeCount; condition stays
|
||||
empty). Supports filtering via a filter expression, custom groupBy, ordering
|
||||
by cpu / cpu_allocatable / memory / memory_allocatable, and pagination via
|
||||
offset/limit. Also reports missing required metrics and whether the requested
|
||||
time range falls before the data retention boundary. Numeric metric fields
|
||||
(nodeCPU, nodeCPUAllocatable, nodeMemory, nodeMemoryAllocatable) return -1
|
||||
as a sentinel when no data is available for that field; frontends should render
|
||||
''—'' rather than the literal value.'
|
||||
operationId: ListNodes
|
||||
requestBody:
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
$ref: '#/components/schemas/InframonitoringtypesPostableNodes'
|
||||
responses:
|
||||
"200":
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
properties:
|
||||
data:
|
||||
$ref: '#/components/schemas/InframonitoringtypesNodes'
|
||||
status:
|
||||
type: string
|
||||
required:
|
||||
- status
|
||||
- data
|
||||
type: object
|
||||
description: OK
|
||||
"400":
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
$ref: '#/components/schemas/RenderErrorResponse'
|
||||
description: Bad Request
|
||||
"401":
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
$ref: '#/components/schemas/RenderErrorResponse'
|
||||
description: Unauthorized
|
||||
"403":
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
$ref: '#/components/schemas/RenderErrorResponse'
|
||||
description: Forbidden
|
||||
"500":
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
$ref: '#/components/schemas/RenderErrorResponse'
|
||||
description: Internal Server Error
|
||||
security:
|
||||
- api_key:
|
||||
- VIEWER
|
||||
- tokenizer:
|
||||
- VIEWER
|
||||
summary: List Nodes for Infra Monitoring
|
||||
tags:
|
||||
- inframonitoring
|
||||
/api/v2/infra_monitoring/pods:
|
||||
post:
|
||||
deprecated: false
|
||||
description: 'Returns a paginated list of Kubernetes pods with key metrics:
|
||||
CPU usage, CPU request/limit utilization, memory working set, memory request/limit
|
||||
utilization, current pod phase (pending/running/succeeded/failed/unknown),
|
||||
and pod age (ms since start time). Each pod includes metadata attributes (namespace,
|
||||
node, workload owner such as deployment/statefulset/daemonset/job/cronjob,
|
||||
cluster). Supports filtering via a filter expression, custom groupBy to aggregate
|
||||
pods by any attribute, ordering by any of the six metrics (cpu, cpu_request,
|
||||
cpu_limit, memory, memory_request, memory_limit), and pagination via offset/limit.
|
||||
The response type is ''list'' for the default k8s.pod.uid grouping (each row
|
||||
is one pod with its current phase) or ''grouped_list'' for custom groupBy
|
||||
keys (each row aggregates pods in the group with per-phase counts: pendingPodCount,
|
||||
runningPodCount, succeededPodCount, failedPodCount, unknownPodCount derived
|
||||
from each pod''s latest phase in the window). Also reports missing required
|
||||
metrics and whether the requested time range falls before the data retention
|
||||
boundary. Numeric metric fields (podCPU, podCPURequest, podCPULimit, podMemory,
|
||||
podMemoryRequest, podMemoryLimit, podAge) return -1 as a sentinel when no
|
||||
data is available for that field; frontends should render ''—'' rather than
|
||||
the literal value.'
|
||||
operationId: ListPods
|
||||
requestBody:
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
$ref: '#/components/schemas/InframonitoringtypesPostablePods'
|
||||
responses:
|
||||
"200":
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
properties:
|
||||
data:
|
||||
$ref: '#/components/schemas/InframonitoringtypesPods'
|
||||
status:
|
||||
type: string
|
||||
required:
|
||||
- status
|
||||
- data
|
||||
type: object
|
||||
description: OK
|
||||
"400":
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
$ref: '#/components/schemas/RenderErrorResponse'
|
||||
description: Bad Request
|
||||
"401":
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
$ref: '#/components/schemas/RenderErrorResponse'
|
||||
description: Unauthorized
|
||||
"403":
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
$ref: '#/components/schemas/RenderErrorResponse'
|
||||
description: Forbidden
|
||||
"500":
|
||||
content:
|
||||
application/json:
|
||||
schema:
|
||||
$ref: '#/components/schemas/RenderErrorResponse'
|
||||
description: Internal Server Error
|
||||
security:
|
||||
- api_key:
|
||||
- VIEWER
|
||||
- tokenizer:
|
||||
- VIEWER
|
||||
summary: List Pods for Infra Monitoring
|
||||
tags:
|
||||
- inframonitoring
|
||||
/api/v2/livez:
|
||||
get:
|
||||
deprecated: false
|
||||
|
||||
@@ -13,7 +13,11 @@ import type {
|
||||
|
||||
import type {
|
||||
InframonitoringtypesPostableHostsDTO,
|
||||
InframonitoringtypesPostableNodesDTO,
|
||||
InframonitoringtypesPostablePodsDTO,
|
||||
ListHosts200,
|
||||
ListNodes200,
|
||||
ListPods200,
|
||||
RenderErrorResponseDTO,
|
||||
} from '../sigNoz.schemas';
|
||||
|
||||
@@ -21,7 +25,7 @@ import { GeneratedAPIInstance } from '../../../generatedAPIInstance';
|
||||
import type { ErrorType, BodyType } from '../../../generatedAPIInstance';
|
||||
|
||||
/**
|
||||
* Returns a paginated list of hosts with key infrastructure metrics: CPU usage (%), memory usage (%), I/O wait (%), disk usage (%), and 15-minute load average. Each host includes its current status (active/inactive based on metrics reported in the last 10 minutes) and metadata attributes (e.g., os.type). Supports filtering via a filter expression, filtering by host status, custom groupBy to aggregate hosts by any attribute, ordering by any of the five metrics, and pagination via offset/limit. The response type is 'list' for the default host.name grouping or 'grouped_list' for custom groupBy keys. Also reports missing required metrics and whether the requested time range falls before the data retention boundary.
|
||||
* Returns a paginated list of hosts with key infrastructure metrics: CPU usage (%), memory usage (%), I/O wait (%), disk usage (%), and 15-minute load average. Each host includes its current status (active/inactive based on metrics reported in the last 10 minutes) and metadata attributes (e.g., os.type). Supports filtering via a filter expression, filtering by host status, custom groupBy to aggregate hosts by any attribute, ordering by any of the five metrics, and pagination via offset/limit. The response type is 'list' for the default host.name grouping or 'grouped_list' for custom groupBy keys. Also reports missing required metrics and whether the requested time range falls before the data retention boundary. Numeric metric fields (cpu, memory, wait, load15, diskUsage) return -1 as a sentinel when no data is available for that field; frontends should render '—' rather than the literal value.
|
||||
* @summary List Hosts for Infra Monitoring
|
||||
*/
|
||||
export const listHosts = (
|
||||
@@ -104,3 +108,171 @@ export const useListHosts = <
|
||||
|
||||
return useMutation(mutationOptions);
|
||||
};
|
||||
/**
|
||||
* Returns a paginated list of Kubernetes nodes with key metrics: CPU usage, CPU allocatable, memory working set, memory allocatable, and per-group readyNodeCount / notReadyNodeCount derived from each node's latest k8s.node.condition_ready value in the window. Each node includes metadata attributes (k8s.node.uid, k8s.cluster.name). The response type is 'list' for the default k8s.node.name grouping (each row is one node with its current condition string: ready / not_ready / '') or 'grouped_list' for custom groupBy keys (each row aggregates nodes in the group with readyNodeCount and notReadyNodeCount; condition stays empty). Supports filtering via a filter expression, custom groupBy, ordering by cpu / cpu_allocatable / memory / memory_allocatable, and pagination via offset/limit. Also reports missing required metrics and whether the requested time range falls before the data retention boundary. Numeric metric fields (nodeCPU, nodeCPUAllocatable, nodeMemory, nodeMemoryAllocatable) return -1 as a sentinel when no data is available for that field; frontends should render '—' rather than the literal value.
|
||||
* @summary List Nodes for Infra Monitoring
|
||||
*/
|
||||
export const listNodes = (
|
||||
inframonitoringtypesPostableNodesDTO: BodyType<InframonitoringtypesPostableNodesDTO>,
|
||||
signal?: AbortSignal,
|
||||
) => {
|
||||
return GeneratedAPIInstance<ListNodes200>({
|
||||
url: `/api/v2/infra_monitoring/nodes`,
|
||||
method: 'POST',
|
||||
headers: { 'Content-Type': 'application/json' },
|
||||
data: inframonitoringtypesPostableNodesDTO,
|
||||
signal,
|
||||
});
|
||||
};
|
||||
|
||||
export const getListNodesMutationOptions = <
|
||||
TError = ErrorType<RenderErrorResponseDTO>,
|
||||
TContext = unknown,
|
||||
>(options?: {
|
||||
mutation?: UseMutationOptions<
|
||||
Awaited<ReturnType<typeof listNodes>>,
|
||||
TError,
|
||||
{ data: BodyType<InframonitoringtypesPostableNodesDTO> },
|
||||
TContext
|
||||
>;
|
||||
}): UseMutationOptions<
|
||||
Awaited<ReturnType<typeof listNodes>>,
|
||||
TError,
|
||||
{ data: BodyType<InframonitoringtypesPostableNodesDTO> },
|
||||
TContext
|
||||
> => {
|
||||
const mutationKey = ['listNodes'];
|
||||
const { mutation: mutationOptions } = options
|
||||
? options.mutation &&
|
||||
'mutationKey' in options.mutation &&
|
||||
options.mutation.mutationKey
|
||||
? options
|
||||
: { ...options, mutation: { ...options.mutation, mutationKey } }
|
||||
: { mutation: { mutationKey } };
|
||||
|
||||
const mutationFn: MutationFunction<
|
||||
Awaited<ReturnType<typeof listNodes>>,
|
||||
{ data: BodyType<InframonitoringtypesPostableNodesDTO> }
|
||||
> = (props) => {
|
||||
const { data } = props ?? {};
|
||||
|
||||
return listNodes(data);
|
||||
};
|
||||
|
||||
return { mutationFn, ...mutationOptions };
|
||||
};
|
||||
|
||||
export type ListNodesMutationResult = NonNullable<
|
||||
Awaited<ReturnType<typeof listNodes>>
|
||||
>;
|
||||
export type ListNodesMutationBody =
|
||||
BodyType<InframonitoringtypesPostableNodesDTO>;
|
||||
export type ListNodesMutationError = ErrorType<RenderErrorResponseDTO>;
|
||||
|
||||
/**
|
||||
* @summary List Nodes for Infra Monitoring
|
||||
*/
|
||||
export const useListNodes = <
|
||||
TError = ErrorType<RenderErrorResponseDTO>,
|
||||
TContext = unknown,
|
||||
>(options?: {
|
||||
mutation?: UseMutationOptions<
|
||||
Awaited<ReturnType<typeof listNodes>>,
|
||||
TError,
|
||||
{ data: BodyType<InframonitoringtypesPostableNodesDTO> },
|
||||
TContext
|
||||
>;
|
||||
}): UseMutationResult<
|
||||
Awaited<ReturnType<typeof listNodes>>,
|
||||
TError,
|
||||
{ data: BodyType<InframonitoringtypesPostableNodesDTO> },
|
||||
TContext
|
||||
> => {
|
||||
const mutationOptions = getListNodesMutationOptions(options);
|
||||
|
||||
return useMutation(mutationOptions);
|
||||
};
|
||||
/**
|
||||
* Returns a paginated list of Kubernetes pods with key metrics: CPU usage, CPU request/limit utilization, memory working set, memory request/limit utilization, current pod phase (pending/running/succeeded/failed/unknown), and pod age (ms since start time). Each pod includes metadata attributes (namespace, node, workload owner such as deployment/statefulset/daemonset/job/cronjob, cluster). Supports filtering via a filter expression, custom groupBy to aggregate pods by any attribute, ordering by any of the six metrics (cpu, cpu_request, cpu_limit, memory, memory_request, memory_limit), and pagination via offset/limit. The response type is 'list' for the default k8s.pod.uid grouping (each row is one pod with its current phase) or 'grouped_list' for custom groupBy keys (each row aggregates pods in the group with per-phase counts: pendingPodCount, runningPodCount, succeededPodCount, failedPodCount, unknownPodCount derived from each pod's latest phase in the window). Also reports missing required metrics and whether the requested time range falls before the data retention boundary. Numeric metric fields (podCPU, podCPURequest, podCPULimit, podMemory, podMemoryRequest, podMemoryLimit, podAge) return -1 as a sentinel when no data is available for that field; frontends should render '—' rather than the literal value.
|
||||
* @summary List Pods for Infra Monitoring
|
||||
*/
|
||||
export const listPods = (
|
||||
inframonitoringtypesPostablePodsDTO: BodyType<InframonitoringtypesPostablePodsDTO>,
|
||||
signal?: AbortSignal,
|
||||
) => {
|
||||
return GeneratedAPIInstance<ListPods200>({
|
||||
url: `/api/v2/infra_monitoring/pods`,
|
||||
method: 'POST',
|
||||
headers: { 'Content-Type': 'application/json' },
|
||||
data: inframonitoringtypesPostablePodsDTO,
|
||||
signal,
|
||||
});
|
||||
};
|
||||
|
||||
export const getListPodsMutationOptions = <
|
||||
TError = ErrorType<RenderErrorResponseDTO>,
|
||||
TContext = unknown,
|
||||
>(options?: {
|
||||
mutation?: UseMutationOptions<
|
||||
Awaited<ReturnType<typeof listPods>>,
|
||||
TError,
|
||||
{ data: BodyType<InframonitoringtypesPostablePodsDTO> },
|
||||
TContext
|
||||
>;
|
||||
}): UseMutationOptions<
|
||||
Awaited<ReturnType<typeof listPods>>,
|
||||
TError,
|
||||
{ data: BodyType<InframonitoringtypesPostablePodsDTO> },
|
||||
TContext
|
||||
> => {
|
||||
const mutationKey = ['listPods'];
|
||||
const { mutation: mutationOptions } = options
|
||||
? options.mutation &&
|
||||
'mutationKey' in options.mutation &&
|
||||
options.mutation.mutationKey
|
||||
? options
|
||||
: { ...options, mutation: { ...options.mutation, mutationKey } }
|
||||
: { mutation: { mutationKey } };
|
||||
|
||||
const mutationFn: MutationFunction<
|
||||
Awaited<ReturnType<typeof listPods>>,
|
||||
{ data: BodyType<InframonitoringtypesPostablePodsDTO> }
|
||||
> = (props) => {
|
||||
const { data } = props ?? {};
|
||||
|
||||
return listPods(data);
|
||||
};
|
||||
|
||||
return { mutationFn, ...mutationOptions };
|
||||
};
|
||||
|
||||
export type ListPodsMutationResult = NonNullable<
|
||||
Awaited<ReturnType<typeof listPods>>
|
||||
>;
|
||||
export type ListPodsMutationBody =
|
||||
BodyType<InframonitoringtypesPostablePodsDTO>;
|
||||
export type ListPodsMutationError = ErrorType<RenderErrorResponseDTO>;
|
||||
|
||||
/**
|
||||
* @summary List Pods for Infra Monitoring
|
||||
*/
|
||||
export const useListPods = <
|
||||
TError = ErrorType<RenderErrorResponseDTO>,
|
||||
TContext = unknown,
|
||||
>(options?: {
|
||||
mutation?: UseMutationOptions<
|
||||
Awaited<ReturnType<typeof listPods>>,
|
||||
TError,
|
||||
{ data: BodyType<InframonitoringtypesPostablePodsDTO> },
|
||||
TContext
|
||||
>;
|
||||
}): UseMutationResult<
|
||||
Awaited<ReturnType<typeof listPods>>,
|
||||
TError,
|
||||
{ data: BodyType<InframonitoringtypesPostablePodsDTO> },
|
||||
TContext
|
||||
> => {
|
||||
const mutationOptions = getListPodsMutationOptions(options);
|
||||
|
||||
return useMutation(mutationOptions);
|
||||
};
|
||||
|
||||
@@ -3243,6 +3243,180 @@ export interface InframonitoringtypesHostsDTO {
|
||||
warning?: Querybuildertypesv5QueryWarnDataDTO;
|
||||
}
|
||||
|
||||
export enum InframonitoringtypesNodeConditionDTO {
|
||||
ready = 'ready',
|
||||
not_ready = 'not_ready',
|
||||
'' = '',
|
||||
}
|
||||
/**
|
||||
* @nullable
|
||||
*/
|
||||
export type InframonitoringtypesNodeRecordDTOMeta = {
|
||||
[key: string]: unknown;
|
||||
} | null;
|
||||
|
||||
export interface InframonitoringtypesNodeRecordDTO {
|
||||
condition: InframonitoringtypesNodeConditionDTO;
|
||||
/**
|
||||
* @type object
|
||||
* @nullable true
|
||||
*/
|
||||
meta: InframonitoringtypesNodeRecordDTOMeta;
|
||||
/**
|
||||
* @type number
|
||||
* @format double
|
||||
*/
|
||||
nodeCPU: number;
|
||||
/**
|
||||
* @type number
|
||||
* @format double
|
||||
*/
|
||||
nodeCPUAllocatable: number;
|
||||
/**
|
||||
* @type number
|
||||
* @format double
|
||||
*/
|
||||
nodeMemory: number;
|
||||
/**
|
||||
* @type number
|
||||
* @format double
|
||||
*/
|
||||
nodeMemoryAllocatable: number;
|
||||
/**
|
||||
* @type string
|
||||
*/
|
||||
nodeName: string;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
notReadyNodeCount: number;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
readyNodeCount: number;
|
||||
}
|
||||
|
||||
export interface InframonitoringtypesNodesDTO {
|
||||
/**
|
||||
* @type boolean
|
||||
*/
|
||||
endTimeBeforeRetention: boolean;
|
||||
/**
|
||||
* @type array
|
||||
* @nullable true
|
||||
*/
|
||||
records: InframonitoringtypesNodeRecordDTO[] | null;
|
||||
requiredMetricsCheck: InframonitoringtypesRequiredMetricsCheckDTO;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
total: number;
|
||||
type: InframonitoringtypesResponseTypeDTO;
|
||||
warning?: Querybuildertypesv5QueryWarnDataDTO;
|
||||
}
|
||||
|
||||
export enum InframonitoringtypesPodPhaseDTO {
|
||||
pending = 'pending',
|
||||
running = 'running',
|
||||
succeeded = 'succeeded',
|
||||
failed = 'failed',
|
||||
unknown = 'unknown',
|
||||
'' = '',
|
||||
}
|
||||
/**
|
||||
* @nullable
|
||||
*/
|
||||
export type InframonitoringtypesPodRecordDTOMeta = {
|
||||
[key: string]: unknown;
|
||||
} | null;
|
||||
|
||||
export interface InframonitoringtypesPodRecordDTO {
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
failedPodCount: number;
|
||||
/**
|
||||
* @type object
|
||||
* @nullable true
|
||||
*/
|
||||
meta: InframonitoringtypesPodRecordDTOMeta;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
pendingPodCount: number;
|
||||
/**
|
||||
* @type integer
|
||||
* @format int64
|
||||
*/
|
||||
podAge: number;
|
||||
/**
|
||||
* @type number
|
||||
* @format double
|
||||
*/
|
||||
podCPU: number;
|
||||
/**
|
||||
* @type number
|
||||
* @format double
|
||||
*/
|
||||
podCPULimit: number;
|
||||
/**
|
||||
* @type number
|
||||
* @format double
|
||||
*/
|
||||
podCPURequest: number;
|
||||
/**
|
||||
* @type number
|
||||
* @format double
|
||||
*/
|
||||
podMemory: number;
|
||||
/**
|
||||
* @type number
|
||||
* @format double
|
||||
*/
|
||||
podMemoryLimit: number;
|
||||
/**
|
||||
* @type number
|
||||
* @format double
|
||||
*/
|
||||
podMemoryRequest: number;
|
||||
podPhase: InframonitoringtypesPodPhaseDTO;
|
||||
/**
|
||||
* @type string
|
||||
*/
|
||||
podUID: string;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
runningPodCount: number;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
succeededPodCount: number;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
unknownPodCount: number;
|
||||
}
|
||||
|
||||
export interface InframonitoringtypesPodsDTO {
|
||||
/**
|
||||
* @type boolean
|
||||
*/
|
||||
endTimeBeforeRetention: boolean;
|
||||
/**
|
||||
* @type array
|
||||
* @nullable true
|
||||
*/
|
||||
records: InframonitoringtypesPodRecordDTO[] | null;
|
||||
requiredMetricsCheck: InframonitoringtypesRequiredMetricsCheckDTO;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
total: number;
|
||||
type: InframonitoringtypesResponseTypeDTO;
|
||||
warning?: Querybuildertypesv5QueryWarnDataDTO;
|
||||
}
|
||||
|
||||
export interface InframonitoringtypesPostableHostsDTO {
|
||||
/**
|
||||
* @type integer
|
||||
@@ -3271,6 +3445,62 @@ export interface InframonitoringtypesPostableHostsDTO {
|
||||
start: number;
|
||||
}
|
||||
|
||||
export interface InframonitoringtypesPostableNodesDTO {
|
||||
/**
|
||||
* @type integer
|
||||
* @format int64
|
||||
*/
|
||||
end: number;
|
||||
filter?: Querybuildertypesv5FilterDTO;
|
||||
/**
|
||||
* @type array
|
||||
* @nullable true
|
||||
*/
|
||||
groupBy?: Querybuildertypesv5GroupByKeyDTO[] | null;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
limit: number;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
offset?: number;
|
||||
orderBy?: Querybuildertypesv5OrderByDTO;
|
||||
/**
|
||||
* @type integer
|
||||
* @format int64
|
||||
*/
|
||||
start: number;
|
||||
}
|
||||
|
||||
export interface InframonitoringtypesPostablePodsDTO {
|
||||
/**
|
||||
* @type integer
|
||||
* @format int64
|
||||
*/
|
||||
end: number;
|
||||
filter?: Querybuildertypesv5FilterDTO;
|
||||
/**
|
||||
* @type array
|
||||
* @nullable true
|
||||
*/
|
||||
groupBy?: Querybuildertypesv5GroupByKeyDTO[] | null;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
limit: number;
|
||||
/**
|
||||
* @type integer
|
||||
*/
|
||||
offset?: number;
|
||||
orderBy?: Querybuildertypesv5OrderByDTO;
|
||||
/**
|
||||
* @type integer
|
||||
* @format int64
|
||||
*/
|
||||
start: number;
|
||||
}
|
||||
|
||||
export interface InframonitoringtypesRequiredMetricsCheckDTO {
|
||||
/**
|
||||
* @type array
|
||||
@@ -7107,6 +7337,22 @@ export type ListHosts200 = {
|
||||
status: string;
|
||||
};
|
||||
|
||||
export type ListNodes200 = {
|
||||
data: InframonitoringtypesNodesDTO;
|
||||
/**
|
||||
* @type string
|
||||
*/
|
||||
status: string;
|
||||
};
|
||||
|
||||
export type ListPods200 = {
|
||||
data: InframonitoringtypesPodsDTO;
|
||||
/**
|
||||
* @type string
|
||||
*/
|
||||
status: string;
|
||||
};
|
||||
|
||||
export type Livez200 = {
|
||||
data: FactoryResponseDTO;
|
||||
/**
|
||||
|
||||
@@ -139,8 +139,8 @@ function ChartPreview({
|
||||
if (startTime && endTime && startTime !== endTime) {
|
||||
dispatch(
|
||||
UpdateTimeInterval('custom', [
|
||||
parseInt(getTimeString(startTime), 10),
|
||||
parseInt(getTimeString(endTime), 10),
|
||||
Number.parseInt(getTimeString(startTime), 10),
|
||||
Number.parseInt(getTimeString(endTime), 10),
|
||||
]),
|
||||
);
|
||||
}
|
||||
|
||||
@@ -370,7 +370,7 @@ function FormAlertRules({
|
||||
// onQueryCategoryChange handles changes to query category
|
||||
// in state as well as sets additional defaults
|
||||
const onQueryCategoryChange = (val: EQueryType): void => {
|
||||
const element = document.getElementById('top');
|
||||
const element = document.querySelector('#top');
|
||||
if (element) {
|
||||
element.scrollIntoView({ behavior: 'smooth' });
|
||||
}
|
||||
|
||||
@@ -279,7 +279,7 @@ function Explorer(): JSX.Element {
|
||||
[],
|
||||
);
|
||||
|
||||
const [warning, setWarning] = useState<Warning | undefined>(undefined);
|
||||
const [warning, setWarning] = useState<Warning | undefined>();
|
||||
|
||||
const oneChartPerQueryDisabledTooltip = useMemo(() => {
|
||||
if (splitedQueries.length <= 1) {
|
||||
@@ -291,7 +291,7 @@ function Explorer(): JSX.Element {
|
||||
if (disableOneChartPerQuery) {
|
||||
return 'One chart per query cannot be disabled for multiple queries with different units.';
|
||||
}
|
||||
return undefined;
|
||||
return;
|
||||
}, [disableOneChartPerQuery, splitedQueries.length, units.length]);
|
||||
|
||||
// Show the y axis unit selector if -
|
||||
|
||||
@@ -217,7 +217,7 @@ function Inspect({
|
||||
);
|
||||
}
|
||||
|
||||
if (!inspectMetricsTimeSeries.length) {
|
||||
if (inspectMetricsTimeSeries.length === 0) {
|
||||
return renderFallback(
|
||||
'inspect-metrics-empty',
|
||||
<Empty description="No time series found for this metric to inspect." />,
|
||||
|
||||
@@ -254,10 +254,10 @@ export function useInspectMetrics(
|
||||
const valuesMap = new Map<number, number>();
|
||||
|
||||
series.values.forEach(({ timestamp, value }) => {
|
||||
valuesMap.set(timestamp, parseFloat(value));
|
||||
valuesMap.set(timestamp, Number.parseFloat(value));
|
||||
});
|
||||
|
||||
return timestamps.map((timestamp) => valuesMap.get(timestamp) ?? NaN);
|
||||
return timestamps.map((timestamp) => valuesMap.get(timestamp) ?? Number.NaN);
|
||||
});
|
||||
|
||||
const rawData = [timestamps, ...timeseriesArray];
|
||||
@@ -271,7 +271,7 @@ export function useInspectMetrics(
|
||||
labels.add(label);
|
||||
});
|
||||
});
|
||||
return Array.from(labels);
|
||||
return [...labels];
|
||||
}, [inspectMetricsData]);
|
||||
|
||||
const reset = useCallback(() => {
|
||||
|
||||
@@ -18,7 +18,7 @@ describe('RunQueryBtn', () => {
|
||||
);
|
||||
});
|
||||
|
||||
test('renders run state and triggers on click', async () => {
|
||||
it('renders run state and triggers on click', async () => {
|
||||
const user = userEvent.setup();
|
||||
const onRun = jest.fn();
|
||||
const onCancel = jest.fn();
|
||||
@@ -35,7 +35,7 @@ describe('RunQueryBtn', () => {
|
||||
expect(onRun).toHaveBeenCalledTimes(1);
|
||||
});
|
||||
|
||||
test('shows cancel state and calls handleCancelQuery', async () => {
|
||||
it('shows cancel state and calls handleCancelQuery', async () => {
|
||||
const user = userEvent.setup();
|
||||
const onRun = jest.fn();
|
||||
const onCancel = jest.fn();
|
||||
@@ -51,19 +51,19 @@ describe('RunQueryBtn', () => {
|
||||
expect(onCancel).toHaveBeenCalledTimes(1);
|
||||
});
|
||||
|
||||
test('disabled when disabled prop is true', () => {
|
||||
it('disabled when disabled prop is true', () => {
|
||||
render(<RunQueryBtn disabled />);
|
||||
expect(screen.getByRole('button', { name: /run query/i })).toBeDisabled();
|
||||
});
|
||||
|
||||
test('disabled when no props provided', () => {
|
||||
it('disabled when no props provided', () => {
|
||||
render(<RunQueryBtn />);
|
||||
expect(
|
||||
screen.getByRole('button', { name: /run query/i }),
|
||||
).toBeInTheDocument();
|
||||
});
|
||||
|
||||
test('shows Command + CornerDownLeft on mac', () => {
|
||||
it('shows Command + CornerDownLeft on mac', () => {
|
||||
const { container } = render(
|
||||
<RunQueryBtn
|
||||
onStageRunQuery={jest.fn()}
|
||||
@@ -77,7 +77,7 @@ describe('RunQueryBtn', () => {
|
||||
).toBeInTheDocument();
|
||||
});
|
||||
|
||||
test('shows ChevronUp + CornerDownLeft on non-mac', () => {
|
||||
it('shows ChevronUp + CornerDownLeft on non-mac', () => {
|
||||
(getUserOperatingSystem as jest.Mock).mockReturnValue(
|
||||
UserOperatingSystem.WINDOWS,
|
||||
);
|
||||
@@ -95,7 +95,7 @@ describe('RunQueryBtn', () => {
|
||||
).toBeInTheDocument();
|
||||
});
|
||||
|
||||
test('renders custom label when provided', () => {
|
||||
it('renders custom label when provided', () => {
|
||||
render(
|
||||
<RunQueryBtn
|
||||
onStageRunQuery={jest.fn()}
|
||||
|
||||
@@ -115,8 +115,8 @@ export const useGetQueryRange: UseGetQueryRange = (
|
||||
|
||||
const updatedQuery = updateBarStepInterval(
|
||||
requestData.query,
|
||||
requestData.start ? requestData.start * 1e3 : parseInt(start, 10) * 1e3,
|
||||
requestData.end ? requestData.end * 1e3 : parseInt(end, 10) * 1e3,
|
||||
requestData.start ? requestData.start * 1e3 : Number.parseInt(start, 10) * 1e3,
|
||||
requestData.end ? requestData.end * 1e3 : Number.parseInt(end, 10) * 1e3,
|
||||
);
|
||||
|
||||
return {
|
||||
|
||||
@@ -98,7 +98,7 @@ function LogsExplorer(): JSX.Element {
|
||||
setIsLoadingQueries(false);
|
||||
}, [queryClient]);
|
||||
|
||||
const [warning, setWarning] = useState<Warning | undefined>(undefined);
|
||||
const [warning, setWarning] = useState<Warning | undefined>();
|
||||
|
||||
const handleChangeSelectedView = useCallback(
|
||||
(view: ExplorerViews, querySearchParameters?: ICurrentQueryData): void => {
|
||||
|
||||
@@ -101,7 +101,7 @@ function TracesExplorer(): JSX.Element {
|
||||
getExplorerViewFromUrl(searchParams, panelTypesFromUrl),
|
||||
);
|
||||
|
||||
const [warning, setWarning] = useState<Warning | undefined>(undefined);
|
||||
const [warning, setWarning] = useState<Warning | undefined>();
|
||||
const [isOpen, setOpen] = useState<boolean>(true);
|
||||
|
||||
const defaultQuery = useMemo(
|
||||
|
||||
@@ -16,7 +16,7 @@ func (provider *provider) addInfraMonitoringRoutes(router *mux.Router) error {
|
||||
ID: "ListHosts",
|
||||
Tags: []string{"inframonitoring"},
|
||||
Summary: "List Hosts for Infra Monitoring",
|
||||
Description: "Returns a paginated list of hosts with key infrastructure metrics: CPU usage (%), memory usage (%), I/O wait (%), disk usage (%), and 15-minute load average. Each host includes its current status (active/inactive based on metrics reported in the last 10 minutes) and metadata attributes (e.g., os.type). Supports filtering via a filter expression, filtering by host status, custom groupBy to aggregate hosts by any attribute, ordering by any of the five metrics, and pagination via offset/limit. The response type is 'list' for the default host.name grouping or 'grouped_list' for custom groupBy keys. Also reports missing required metrics and whether the requested time range falls before the data retention boundary.",
|
||||
Description: "Returns a paginated list of hosts with key infrastructure metrics: CPU usage (%), memory usage (%), I/O wait (%), disk usage (%), and 15-minute load average. Each host includes its current status (active/inactive based on metrics reported in the last 10 minutes) and metadata attributes (e.g., os.type). Supports filtering via a filter expression, filtering by host status, custom groupBy to aggregate hosts by any attribute, ordering by any of the five metrics, and pagination via offset/limit. The response type is 'list' for the default host.name grouping or 'grouped_list' for custom groupBy keys. Also reports missing required metrics and whether the requested time range falls before the data retention boundary. Numeric metric fields (cpu, memory, wait, load15, diskUsage) return -1 as a sentinel when no data is available for that field; frontends should render '—' rather than the literal value.",
|
||||
Request: new(inframonitoringtypes.PostableHosts),
|
||||
RequestContentType: "application/json",
|
||||
Response: new(inframonitoringtypes.Hosts),
|
||||
@@ -29,5 +29,43 @@ func (provider *provider) addInfraMonitoringRoutes(router *mux.Router) error {
|
||||
return err
|
||||
}
|
||||
|
||||
if err := router.Handle("/api/v2/infra_monitoring/pods", handler.New(
|
||||
provider.authZ.ViewAccess(provider.infraMonitoringHandler.ListPods),
|
||||
handler.OpenAPIDef{
|
||||
ID: "ListPods",
|
||||
Tags: []string{"inframonitoring"},
|
||||
Summary: "List Pods for Infra Monitoring",
|
||||
Description: "Returns a paginated list of Kubernetes pods with key metrics: CPU usage, CPU request/limit utilization, memory working set, memory request/limit utilization, current pod phase (pending/running/succeeded/failed/unknown), and pod age (ms since start time). Each pod includes metadata attributes (namespace, node, workload owner such as deployment/statefulset/daemonset/job/cronjob, cluster). Supports filtering via a filter expression, custom groupBy to aggregate pods by any attribute, ordering by any of the six metrics (cpu, cpu_request, cpu_limit, memory, memory_request, memory_limit), and pagination via offset/limit. The response type is 'list' for the default k8s.pod.uid grouping (each row is one pod with its current phase) or 'grouped_list' for custom groupBy keys (each row aggregates pods in the group with per-phase counts: pendingPodCount, runningPodCount, succeededPodCount, failedPodCount, unknownPodCount derived from each pod's latest phase in the window). Also reports missing required metrics and whether the requested time range falls before the data retention boundary. Numeric metric fields (podCPU, podCPURequest, podCPULimit, podMemory, podMemoryRequest, podMemoryLimit, podAge) return -1 as a sentinel when no data is available for that field; frontends should render '—' rather than the literal value.",
|
||||
Request: new(inframonitoringtypes.PostablePods),
|
||||
RequestContentType: "application/json",
|
||||
Response: new(inframonitoringtypes.Pods),
|
||||
ResponseContentType: "application/json",
|
||||
SuccessStatusCode: http.StatusOK,
|
||||
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusUnauthorized},
|
||||
Deprecated: false,
|
||||
SecuritySchemes: newSecuritySchemes(types.RoleViewer),
|
||||
})).Methods(http.MethodPost).GetError(); err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
if err := router.Handle("/api/v2/infra_monitoring/nodes", handler.New(
|
||||
provider.authZ.ViewAccess(provider.infraMonitoringHandler.ListNodes),
|
||||
handler.OpenAPIDef{
|
||||
ID: "ListNodes",
|
||||
Tags: []string{"inframonitoring"},
|
||||
Summary: "List Nodes for Infra Monitoring",
|
||||
Description: "Returns a paginated list of Kubernetes nodes with key metrics: CPU usage, CPU allocatable, memory working set, memory allocatable, and per-group readyNodeCount / notReadyNodeCount derived from each node's latest k8s.node.condition_ready value in the window. Each node includes metadata attributes (k8s.node.uid, k8s.cluster.name). The response type is 'list' for the default k8s.node.name grouping (each row is one node with its current condition string: ready / not_ready / '') or 'grouped_list' for custom groupBy keys (each row aggregates nodes in the group with readyNodeCount and notReadyNodeCount; condition stays empty). Supports filtering via a filter expression, custom groupBy, ordering by cpu / cpu_allocatable / memory / memory_allocatable, and pagination via offset/limit. Also reports missing required metrics and whether the requested time range falls before the data retention boundary. Numeric metric fields (nodeCPU, nodeCPUAllocatable, nodeMemory, nodeMemoryAllocatable) return -1 as a sentinel when no data is available for that field; frontends should render '—' rather than the literal value.",
|
||||
Request: new(inframonitoringtypes.PostableNodes),
|
||||
RequestContentType: "application/json",
|
||||
Response: new(inframonitoringtypes.Nodes),
|
||||
ResponseContentType: "application/json",
|
||||
SuccessStatusCode: http.StatusOK,
|
||||
ErrorStatusCodes: []int{http.StatusBadRequest, http.StatusUnauthorized},
|
||||
Deprecated: false,
|
||||
SecuritySchemes: newSecuritySchemes(types.RoleViewer),
|
||||
})).Methods(http.MethodPost).GetError(); err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
@@ -45,3 +45,51 @@ func (h *handler) ListHosts(rw http.ResponseWriter, req *http.Request) {
|
||||
|
||||
render.Success(rw, http.StatusOK, result)
|
||||
}
|
||||
|
||||
func (h *handler) ListPods(rw http.ResponseWriter, req *http.Request) {
|
||||
claims, err := authtypes.ClaimsFromContext(req.Context())
|
||||
if err != nil {
|
||||
render.Error(rw, err)
|
||||
return
|
||||
}
|
||||
|
||||
orgID := valuer.MustNewUUID(claims.OrgID)
|
||||
|
||||
var parsedReq inframonitoringtypes.PostablePods
|
||||
if err := binding.JSON.BindBody(req.Body, &parsedReq); err != nil {
|
||||
render.Error(rw, err)
|
||||
return
|
||||
}
|
||||
|
||||
result, err := h.module.ListPods(req.Context(), orgID, &parsedReq)
|
||||
if err != nil {
|
||||
render.Error(rw, err)
|
||||
return
|
||||
}
|
||||
|
||||
render.Success(rw, http.StatusOK, result)
|
||||
}
|
||||
|
||||
func (h *handler) ListNodes(rw http.ResponseWriter, req *http.Request) {
|
||||
claims, err := authtypes.ClaimsFromContext(req.Context())
|
||||
if err != nil {
|
||||
render.Error(rw, err)
|
||||
return
|
||||
}
|
||||
|
||||
orgID := valuer.MustNewUUID(claims.OrgID)
|
||||
|
||||
var parsedReq inframonitoringtypes.PostableNodes
|
||||
if err := binding.JSON.BindBody(req.Body, &parsedReq); err != nil {
|
||||
render.Error(rw, err)
|
||||
return
|
||||
}
|
||||
|
||||
result, err := h.module.ListNodes(req.Context(), orgID, &parsedReq)
|
||||
if err != nil {
|
||||
render.Error(rw, err)
|
||||
return
|
||||
}
|
||||
|
||||
render.Success(rw, http.StatusOK, result)
|
||||
}
|
||||
|
||||
@@ -14,3 +14,18 @@ type groupHostStatusCounts struct {
|
||||
Active int
|
||||
Inactive int
|
||||
}
|
||||
|
||||
// podPhaseCounts holds per-group pod counts bucketed by latest phase in window.
|
||||
type podPhaseCounts struct {
|
||||
Pending int
|
||||
Running int
|
||||
Succeeded int
|
||||
Failed int
|
||||
Unknown int
|
||||
}
|
||||
|
||||
// nodeConditionCounts holds per-group node counts bucketed by latest condition_ready in window.
|
||||
type nodeConditionCounts struct {
|
||||
Ready int
|
||||
NotReady int
|
||||
}
|
||||
|
||||
@@ -159,3 +159,169 @@ func (m *module) ListHosts(ctx context.Context, orgID valuer.UUID, req *inframon
|
||||
|
||||
return resp, nil
|
||||
}
|
||||
|
||||
func (m *module) ListPods(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostablePods) (*inframonitoringtypes.Pods, error) {
|
||||
if err := req.Validate(); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
resp := &inframonitoringtypes.Pods{}
|
||||
|
||||
if req.OrderBy == nil {
|
||||
req.OrderBy = &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: inframonitoringtypes.PodsOrderByCPU,
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirectionDesc,
|
||||
}
|
||||
}
|
||||
|
||||
if len(req.GroupBy) == 0 {
|
||||
req.GroupBy = []qbtypes.GroupByKey{podUIDGroupByKey}
|
||||
resp.Type = inframonitoringtypes.ResponseTypeList
|
||||
} else {
|
||||
resp.Type = inframonitoringtypes.ResponseTypeGroupedList
|
||||
}
|
||||
|
||||
missingMetrics, minFirstReportedUnixMilli, err := m.getMetricsExistenceAndEarliestTime(ctx, podsTableMetricNamesList)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
if len(missingMetrics) > 0 {
|
||||
resp.RequiredMetricsCheck = inframonitoringtypes.RequiredMetricsCheck{MissingMetrics: missingMetrics}
|
||||
resp.Records = []inframonitoringtypes.PodRecord{}
|
||||
resp.Total = 0
|
||||
return resp, nil
|
||||
}
|
||||
if req.End < int64(minFirstReportedUnixMilli) {
|
||||
resp.EndTimeBeforeRetention = true
|
||||
resp.Records = []inframonitoringtypes.PodRecord{}
|
||||
resp.Total = 0
|
||||
return resp, nil
|
||||
}
|
||||
resp.RequiredMetricsCheck = inframonitoringtypes.RequiredMetricsCheck{MissingMetrics: []string{}}
|
||||
|
||||
metadataMap, err := m.getPodsTableMetadata(ctx, req)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
resp.Total = len(metadataMap)
|
||||
|
||||
pageGroups, err := m.getTopPodGroups(ctx, orgID, req, metadataMap)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
if len(pageGroups) == 0 {
|
||||
resp.Records = []inframonitoringtypes.PodRecord{}
|
||||
return resp, nil
|
||||
}
|
||||
|
||||
filterExpr := ""
|
||||
if req.Filter != nil {
|
||||
filterExpr = req.Filter.Expression
|
||||
}
|
||||
|
||||
fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newPodsTableListQuery())
|
||||
queryResp, err := m.querier.QueryRange(ctx, orgID, fullQueryReq)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
phaseCounts, err := m.getPerGroupPodPhaseCounts(ctx, req, pageGroups)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
isPodUIDInGroupBy := isKeyInGroupByAttrs(req.GroupBy, podUIDAttrKey)
|
||||
resp.Records = buildPodRecords(isPodUIDInGroupBy, queryResp, pageGroups, req.GroupBy, metadataMap, phaseCounts, req.End)
|
||||
resp.Warning = queryResp.Warning
|
||||
|
||||
return resp, nil
|
||||
}
|
||||
|
||||
func (m *module) ListNodes(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableNodes) (*inframonitoringtypes.Nodes, error) {
|
||||
if err := req.Validate(); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
resp := &inframonitoringtypes.Nodes{}
|
||||
|
||||
if req.OrderBy == nil {
|
||||
req.OrderBy = &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: inframonitoringtypes.NodesOrderByCPU,
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirectionDesc,
|
||||
}
|
||||
}
|
||||
|
||||
if len(req.GroupBy) == 0 {
|
||||
req.GroupBy = []qbtypes.GroupByKey{nodeNameGroupByKey}
|
||||
resp.Type = inframonitoringtypes.ResponseTypeList
|
||||
} else {
|
||||
resp.Type = inframonitoringtypes.ResponseTypeGroupedList
|
||||
}
|
||||
|
||||
missingMetrics, minFirstReportedUnixMilli, err := m.getMetricsExistenceAndEarliestTime(ctx, nodesTableMetricNamesList)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
if len(missingMetrics) > 0 {
|
||||
resp.RequiredMetricsCheck = inframonitoringtypes.RequiredMetricsCheck{MissingMetrics: missingMetrics}
|
||||
resp.Records = []inframonitoringtypes.NodeRecord{}
|
||||
resp.Total = 0
|
||||
return resp, nil
|
||||
}
|
||||
if req.End < int64(minFirstReportedUnixMilli) {
|
||||
resp.EndTimeBeforeRetention = true
|
||||
resp.Records = []inframonitoringtypes.NodeRecord{}
|
||||
resp.Total = 0
|
||||
return resp, nil
|
||||
}
|
||||
resp.RequiredMetricsCheck = inframonitoringtypes.RequiredMetricsCheck{MissingMetrics: []string{}}
|
||||
|
||||
metadataMap, err := m.getNodesTableMetadata(ctx, req)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
resp.Total = len(metadataMap)
|
||||
|
||||
pageGroups, err := m.getTopNodeGroups(ctx, orgID, req, metadataMap)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
if len(pageGroups) == 0 {
|
||||
resp.Records = []inframonitoringtypes.NodeRecord{}
|
||||
return resp, nil
|
||||
}
|
||||
|
||||
filterExpr := ""
|
||||
if req.Filter != nil {
|
||||
filterExpr = req.Filter.Expression
|
||||
}
|
||||
|
||||
fullQueryReq := buildFullQueryRequest(req.Start, req.End, filterExpr, req.GroupBy, pageGroups, m.newNodesTableListQuery())
|
||||
queryResp, err := m.querier.QueryRange(ctx, orgID, fullQueryReq)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
conditionCounts, err := m.getPerGroupNodeConditionCounts(ctx, req, pageGroups)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
isNodeNameInGroupBy := isKeyInGroupByAttrs(req.GroupBy, nodeNameAttrKey)
|
||||
resp.Records = buildNodeRecords(isNodeNameInGroupBy, queryResp, pageGroups, req.GroupBy, metadataMap, conditionCounts)
|
||||
resp.Warning = queryResp.Warning
|
||||
|
||||
return resp, nil
|
||||
}
|
||||
|
||||
299
pkg/modules/inframonitoring/implinframonitoring/nodes.go
Normal file
299
pkg/modules/inframonitoring/implinframonitoring/nodes.go
Normal file
@@ -0,0 +1,299 @@
|
||||
package implinframonitoring
|
||||
|
||||
import (
|
||||
"context"
|
||||
"fmt"
|
||||
"slices"
|
||||
"strings"
|
||||
|
||||
"github.com/SigNoz/signoz/pkg/querybuilder"
|
||||
"github.com/SigNoz/signoz/pkg/telemetrymetrics"
|
||||
"github.com/SigNoz/signoz/pkg/types/inframonitoringtypes"
|
||||
"github.com/SigNoz/signoz/pkg/types/metrictypes"
|
||||
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
|
||||
"github.com/SigNoz/signoz/pkg/valuer"
|
||||
"github.com/huandu/go-sqlbuilder"
|
||||
)
|
||||
|
||||
// buildNodeRecords assembles the page records. Condition counts come from
|
||||
// conditionCounts in both modes. In list mode (isNodeNameInGroupBy=true) each
|
||||
// group is one node, so exactly one count is 1; Condition is derived from
|
||||
// which one. In grouped_list mode Condition stays NodeConditionNone.
|
||||
func buildNodeRecords(
|
||||
isNodeNameInGroupBy bool,
|
||||
resp *qbtypes.QueryRangeResponse,
|
||||
pageGroups []map[string]string,
|
||||
groupBy []qbtypes.GroupByKey,
|
||||
metadataMap map[string]map[string]string,
|
||||
conditionCounts map[string]nodeConditionCounts,
|
||||
) []inframonitoringtypes.NodeRecord {
|
||||
metricsMap := parseFullQueryResponse(resp, groupBy)
|
||||
|
||||
records := make([]inframonitoringtypes.NodeRecord, 0, len(pageGroups))
|
||||
for _, labels := range pageGroups {
|
||||
compositeKey := compositeKeyFromLabels(labels, groupBy)
|
||||
nodeName := labels[nodeNameAttrKey]
|
||||
|
||||
record := inframonitoringtypes.NodeRecord{ // initialize with default values
|
||||
NodeName: nodeName,
|
||||
Condition: inframonitoringtypes.NodeConditionNone,
|
||||
NodeCPU: -1,
|
||||
NodeCPUAllocatable: -1,
|
||||
NodeMemory: -1,
|
||||
NodeMemoryAllocatable: -1,
|
||||
Meta: map[string]any{},
|
||||
}
|
||||
|
||||
if metrics, ok := metricsMap[compositeKey]; ok {
|
||||
if v, exists := metrics["A"]; exists {
|
||||
record.NodeCPU = v
|
||||
}
|
||||
if v, exists := metrics["B"]; exists {
|
||||
record.NodeCPUAllocatable = v
|
||||
}
|
||||
if v, exists := metrics["C"]; exists {
|
||||
record.NodeMemory = v
|
||||
}
|
||||
if v, exists := metrics["D"]; exists {
|
||||
record.NodeMemoryAllocatable = v
|
||||
}
|
||||
}
|
||||
|
||||
if conditionCountsForGroup, ok := conditionCounts[compositeKey]; ok {
|
||||
record.ReadyNodesCount = conditionCountsForGroup.Ready
|
||||
record.NotReadyNodesCount = conditionCountsForGroup.NotReady
|
||||
|
||||
// In list mode each group is one node; the count==1 bucket identifies the condition.
|
||||
if isNodeNameInGroupBy {
|
||||
switch {
|
||||
case conditionCountsForGroup.Ready == 1:
|
||||
record.Condition = inframonitoringtypes.NodeConditionReady
|
||||
case conditionCountsForGroup.NotReady == 1:
|
||||
record.Condition = inframonitoringtypes.NodeConditionNotReady
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
if attrs, ok := metadataMap[compositeKey]; ok {
|
||||
for k, v := range attrs {
|
||||
record.Meta[k] = v
|
||||
}
|
||||
}
|
||||
|
||||
records = append(records, record)
|
||||
}
|
||||
return records
|
||||
}
|
||||
|
||||
func (m *module) getTopNodeGroups(
|
||||
ctx context.Context,
|
||||
orgID valuer.UUID,
|
||||
req *inframonitoringtypes.PostableNodes,
|
||||
metadataMap map[string]map[string]string,
|
||||
) ([]map[string]string, error) {
|
||||
orderByKey := req.OrderBy.Key.Name
|
||||
queryNamesForOrderBy := orderByToNodesQueryNames[orderByKey]
|
||||
rankingQueryName := queryNamesForOrderBy[len(queryNamesForOrderBy)-1]
|
||||
|
||||
topReq := &qbtypes.QueryRangeRequest{
|
||||
Start: uint64(req.Start),
|
||||
End: uint64(req.End),
|
||||
RequestType: qbtypes.RequestTypeScalar,
|
||||
CompositeQuery: qbtypes.CompositeQuery{
|
||||
Queries: make([]qbtypes.QueryEnvelope, 0, len(queryNamesForOrderBy)),
|
||||
},
|
||||
}
|
||||
|
||||
for _, envelope := range m.newNodesTableListQuery().CompositeQuery.Queries {
|
||||
if !slices.Contains(queryNamesForOrderBy, envelope.GetQueryName()) {
|
||||
continue
|
||||
}
|
||||
copied := envelope
|
||||
if copied.Type == qbtypes.QueryTypeBuilder {
|
||||
existingExpr := ""
|
||||
if f := copied.GetFilter(); f != nil {
|
||||
existingExpr = f.Expression
|
||||
}
|
||||
reqFilterExpr := ""
|
||||
if req.Filter != nil {
|
||||
reqFilterExpr = req.Filter.Expression
|
||||
}
|
||||
merged := mergeFilterExpressions(existingExpr, reqFilterExpr)
|
||||
copied.SetFilter(&qbtypes.Filter{Expression: merged})
|
||||
copied.SetGroupBy(req.GroupBy)
|
||||
}
|
||||
topReq.CompositeQuery.Queries = append(topReq.CompositeQuery.Queries, copied)
|
||||
}
|
||||
|
||||
resp, err := m.querier.QueryRange(ctx, orgID, topReq)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
allMetricGroups := parseAndSortGroups(resp, rankingQueryName, req.GroupBy, req.OrderBy.Direction)
|
||||
return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), nil
|
||||
}
|
||||
|
||||
func (m *module) getNodesTableMetadata(ctx context.Context, req *inframonitoringtypes.PostableNodes) (map[string]map[string]string, error) {
|
||||
var nonGroupByAttrs []string
|
||||
for _, key := range nodeAttrKeysForMetadata {
|
||||
if !isKeyInGroupByAttrs(req.GroupBy, key) {
|
||||
nonGroupByAttrs = append(nonGroupByAttrs, key)
|
||||
}
|
||||
}
|
||||
return m.getMetadata(ctx, nodesTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)
|
||||
}
|
||||
|
||||
// getPerGroupNodeConditionCounts computes per-group node counts bucketed by each
|
||||
// node's latest condition_ready value (0 / 1) in the requested window.
|
||||
// Pipeline:
|
||||
//
|
||||
// timeSeriesFPs: fp ↔ (node_name, groupBy cols) from the time_series table.
|
||||
// User filter + page-groups filter applied here.
|
||||
// latestConditionPerNode: INNER JOIN samples × timeSeriesFPs, collapsed to
|
||||
// the latest condition value per node via argMax(value, unix_milli).
|
||||
// countNodesPerCondition: per-group uniqExactIf into ready/not_ready buckets.
|
||||
//
|
||||
// Groups absent from the result map have implicit zero counts (caller default).
|
||||
func (m *module) getPerGroupNodeConditionCounts(
|
||||
ctx context.Context,
|
||||
req *inframonitoringtypes.PostableNodes,
|
||||
pageGroups []map[string]string,
|
||||
) (map[string]nodeConditionCounts, error) {
|
||||
if len(pageGroups) == 0 || len(req.GroupBy) == 0 {
|
||||
return map[string]nodeConditionCounts{}, nil
|
||||
}
|
||||
|
||||
// Merged filter expression (user filter + page-groups IN clauses).
|
||||
reqFilterExpr := ""
|
||||
if req.Filter != nil {
|
||||
reqFilterExpr = req.Filter.Expression
|
||||
}
|
||||
pageGroupsFilterExpr := buildPageGroupsFilterExpr(pageGroups)
|
||||
filterExpr := mergeFilterExpressions(reqFilterExpr, pageGroupsFilterExpr)
|
||||
|
||||
// Resolve tables. Same convention as pods.
|
||||
adjustedStart, adjustedEnd, _, localTimeSeriesTable := telemetrymetrics.WhichTSTableToUse(
|
||||
uint64(req.Start), uint64(req.End), nil,
|
||||
)
|
||||
samplesTable := telemetrymetrics.WhichSamplesTableToUse(
|
||||
uint64(req.Start), uint64(req.End),
|
||||
metrictypes.UnspecifiedType, metrictypes.TimeAggregationUnspecified, nil,
|
||||
)
|
||||
valueCol := telemetrymetrics.ValueColumnForSamplesTable(samplesTable)
|
||||
|
||||
// ----- timeSeriesFPs -----
|
||||
timeSeriesFPs := sqlbuilder.NewSelectBuilder()
|
||||
timeSeriesFPsSelectCols := []string{
|
||||
"fingerprint",
|
||||
fmt.Sprintf("JSONExtractString(labels, %s) AS node_name", timeSeriesFPs.Var(nodeNameAttrKey)),
|
||||
}
|
||||
for _, key := range req.GroupBy {
|
||||
timeSeriesFPsSelectCols = append(timeSeriesFPsSelectCols,
|
||||
fmt.Sprintf("JSONExtractString(labels, %s) AS %s", timeSeriesFPs.Var(key.Name), quoteIdentifier(key.Name)),
|
||||
)
|
||||
}
|
||||
timeSeriesFPs.Select(timeSeriesFPsSelectCols...)
|
||||
timeSeriesFPs.From(fmt.Sprintf("%s.%s", telemetrymetrics.DBName, localTimeSeriesTable))
|
||||
timeSeriesFPs.Where(
|
||||
timeSeriesFPs.E("metric_name", nodeConditionMetricName),
|
||||
timeSeriesFPs.GE("unix_milli", adjustedStart),
|
||||
timeSeriesFPs.L("unix_milli", adjustedEnd),
|
||||
)
|
||||
if filterExpr != "" {
|
||||
filterClause, err := m.buildFilterClause(ctx, &qbtypes.Filter{Expression: filterExpr}, req.Start, req.End)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
if filterClause != nil {
|
||||
timeSeriesFPs.AddWhereClause(filterClause)
|
||||
}
|
||||
}
|
||||
timeSeriesFPsGroupBy := []string{"fingerprint", "node_name"}
|
||||
for _, key := range req.GroupBy {
|
||||
timeSeriesFPsGroupBy = append(timeSeriesFPsGroupBy, quoteIdentifier(key.Name))
|
||||
}
|
||||
timeSeriesFPs.GroupBy(timeSeriesFPsGroupBy...)
|
||||
timeSeriesFPsSQL, timeSeriesFPsArgs := timeSeriesFPs.BuildWithFlavor(sqlbuilder.ClickHouse)
|
||||
|
||||
// ----- latestConditionPerNode -----
|
||||
latestConditionPerNode := sqlbuilder.NewSelectBuilder()
|
||||
latestConditionPerNodeSelectCols := []string{"tsfp.node_name AS node_name"}
|
||||
latestConditionPerNodeGroupBy := []string{"node_name"}
|
||||
for _, key := range req.GroupBy {
|
||||
col := quoteIdentifier(key.Name)
|
||||
latestConditionPerNodeSelectCols = append(latestConditionPerNodeSelectCols, fmt.Sprintf("tsfp.%s AS %s", col, col))
|
||||
latestConditionPerNodeGroupBy = append(latestConditionPerNodeGroupBy, col)
|
||||
}
|
||||
latestConditionPerNodeSelectCols = append(latestConditionPerNodeSelectCols,
|
||||
fmt.Sprintf("argMax(samples.%s, samples.unix_milli) AS condition_value", valueCol),
|
||||
)
|
||||
latestConditionPerNode.Select(latestConditionPerNodeSelectCols...)
|
||||
latestConditionPerNode.From(fmt.Sprintf(
|
||||
"%s.%s AS samples INNER JOIN time_series_fps AS tsfp ON samples.fingerprint = tsfp.fingerprint",
|
||||
telemetrymetrics.DBName, samplesTable,
|
||||
))
|
||||
latestConditionPerNode.Where(
|
||||
latestConditionPerNode.E("samples.metric_name", nodeConditionMetricName),
|
||||
latestConditionPerNode.GE("samples.unix_milli", req.Start),
|
||||
latestConditionPerNode.L("samples.unix_milli", req.End),
|
||||
"tsfp.node_name != ''",
|
||||
)
|
||||
latestConditionPerNode.GroupBy(latestConditionPerNodeGroupBy...)
|
||||
latestConditionPerNodeSQL, latestConditionPerNodeArgs := latestConditionPerNode.BuildWithFlavor(sqlbuilder.ClickHouse)
|
||||
|
||||
// ----- countNodesPerCondition (outer SELECT) -----
|
||||
countNodesPerConditionSelectCols := make([]string, 0, len(req.GroupBy)+2)
|
||||
countNodesPerConditionGroupBy := make([]string, 0, len(req.GroupBy))
|
||||
for _, key := range req.GroupBy {
|
||||
col := quoteIdentifier(key.Name)
|
||||
countNodesPerConditionSelectCols = append(countNodesPerConditionSelectCols, col)
|
||||
countNodesPerConditionGroupBy = append(countNodesPerConditionGroupBy, col)
|
||||
}
|
||||
countNodesPerConditionSelectCols = append(countNodesPerConditionSelectCols,
|
||||
fmt.Sprintf("uniqExactIf(node_name, condition_value = %d) AS ready_count", inframonitoringtypes.NodeConditionNumReady),
|
||||
fmt.Sprintf("uniqExactIf(node_name, condition_value = %d) AS not_ready_count", inframonitoringtypes.NodeConditionNumNotReady),
|
||||
)
|
||||
countNodesPerConditionSQL := fmt.Sprintf(
|
||||
"SELECT %s FROM latest_condition_per_node GROUP BY %s",
|
||||
strings.Join(countNodesPerConditionSelectCols, ", "),
|
||||
strings.Join(countNodesPerConditionGroupBy, ", "),
|
||||
)
|
||||
|
||||
// Combine CTEs + outer.
|
||||
cteFragments := []string{
|
||||
fmt.Sprintf("time_series_fps AS (%s)", timeSeriesFPsSQL),
|
||||
fmt.Sprintf("latest_condition_per_node AS (%s)", latestConditionPerNodeSQL),
|
||||
}
|
||||
finalSQL := querybuilder.CombineCTEs(cteFragments) + countNodesPerConditionSQL
|
||||
finalArgs := querybuilder.PrependArgs([][]any{timeSeriesFPsArgs, latestConditionPerNodeArgs}, nil)
|
||||
|
||||
rows, err := m.telemetryStore.ClickhouseDB().Query(ctx, finalSQL, finalArgs...)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
defer rows.Close()
|
||||
|
||||
result := make(map[string]nodeConditionCounts)
|
||||
for rows.Next() {
|
||||
groupVals := make([]string, len(req.GroupBy))
|
||||
scanPtrs := make([]any, 0, len(req.GroupBy)+2)
|
||||
for i := range groupVals {
|
||||
scanPtrs = append(scanPtrs, &groupVals[i])
|
||||
}
|
||||
var ready, notReady uint64
|
||||
scanPtrs = append(scanPtrs, &ready, ¬Ready)
|
||||
|
||||
if err := rows.Scan(scanPtrs...); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
result[compositeKeyFromList(groupVals)] = nodeConditionCounts{
|
||||
Ready: int(ready),
|
||||
NotReady: int(notReady),
|
||||
}
|
||||
}
|
||||
if err := rows.Err(); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
return result, nil
|
||||
}
|
||||
@@ -0,0 +1,136 @@
|
||||
package implinframonitoring
|
||||
|
||||
import (
|
||||
"github.com/SigNoz/signoz/pkg/types/inframonitoringtypes"
|
||||
"github.com/SigNoz/signoz/pkg/types/metrictypes"
|
||||
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
|
||||
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
|
||||
)
|
||||
|
||||
const (
|
||||
nodeNameAttrKey = "k8s.node.name"
|
||||
nodeConditionMetricName = "k8s.node.condition_ready"
|
||||
)
|
||||
|
||||
var nodeNameGroupByKey = qbtypes.GroupByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: nodeNameAttrKey,
|
||||
FieldContext: telemetrytypes.FieldContextResource,
|
||||
FieldDataType: telemetrytypes.FieldDataTypeString,
|
||||
},
|
||||
}
|
||||
|
||||
// nodesTableMetricNamesList drives the existence/retention check.
|
||||
// Includes condition_ready so the response short-circuits cleanly when a
|
||||
// cluster doesn't ship the metric — even though condition_ready isn't part
|
||||
// of the QB composite query (it's queried separately via getPerGroupNodeConditionCounts).
|
||||
var nodesTableMetricNamesList = []string{
|
||||
"k8s.node.cpu.usage",
|
||||
"k8s.node.allocatable_cpu",
|
||||
"k8s.node.memory.working_set",
|
||||
"k8s.node.allocatable_memory",
|
||||
nodeConditionMetricName,
|
||||
}
|
||||
|
||||
var nodeAttrKeysForMetadata = []string{
|
||||
"k8s.node.name",
|
||||
"k8s.node.uid",
|
||||
"k8s.cluster.name",
|
||||
}
|
||||
|
||||
var orderByToNodesQueryNames = map[string][]string{
|
||||
inframonitoringtypes.NodesOrderByCPU: {"A"},
|
||||
inframonitoringtypes.NodesOrderByCPUAllocatable: {"B"},
|
||||
inframonitoringtypes.NodesOrderByMemory: {"C"},
|
||||
inframonitoringtypes.NodesOrderByMemoryAllocatable: {"D"},
|
||||
}
|
||||
|
||||
// newNodesTableListQuery builds the composite QB v5 request for the nodes list.
|
||||
// Node condition is derived separately via getPerGroupNodeConditionCounts (works
|
||||
// for both list and grouped_list modes), so no condition query is included here.
|
||||
func (m *module) newNodesTableListQuery() *qbtypes.QueryRangeRequest {
|
||||
queries := []qbtypes.QueryEnvelope{
|
||||
// Query A: CPU usage
|
||||
{
|
||||
Type: qbtypes.QueryTypeBuilder,
|
||||
Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
|
||||
Name: "A",
|
||||
Signal: telemetrytypes.SignalMetrics,
|
||||
Aggregations: []qbtypes.MetricAggregation{
|
||||
{
|
||||
MetricName: "k8s.node.cpu.usage",
|
||||
TimeAggregation: metrictypes.TimeAggregationAvg,
|
||||
SpaceAggregation: metrictypes.SpaceAggregationSum,
|
||||
ReduceTo: qbtypes.ReduceToAvg,
|
||||
},
|
||||
},
|
||||
GroupBy: []qbtypes.GroupByKey{nodeNameGroupByKey},
|
||||
Disabled: false,
|
||||
},
|
||||
},
|
||||
// Query B: CPU allocatable.
|
||||
// TimeAggregationLatest is the closest v5 equivalent of v1's AnyLast;
|
||||
// allocatable values change rarely so divergence in practice is negligible.
|
||||
// TODO(nikhilmantri0902): switch to AnyLast once v5 supports it.
|
||||
{
|
||||
Type: qbtypes.QueryTypeBuilder,
|
||||
Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
|
||||
Name: "B",
|
||||
Signal: telemetrytypes.SignalMetrics,
|
||||
Aggregations: []qbtypes.MetricAggregation{
|
||||
{
|
||||
MetricName: "k8s.node.allocatable_cpu",
|
||||
TimeAggregation: metrictypes.TimeAggregationLatest,
|
||||
SpaceAggregation: metrictypes.SpaceAggregationSum,
|
||||
ReduceTo: qbtypes.ReduceToAvg,
|
||||
},
|
||||
},
|
||||
GroupBy: []qbtypes.GroupByKey{nodeNameGroupByKey},
|
||||
Disabled: false,
|
||||
},
|
||||
},
|
||||
// Query C: Memory working set
|
||||
{
|
||||
Type: qbtypes.QueryTypeBuilder,
|
||||
Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
|
||||
Name: "C",
|
||||
Signal: telemetrytypes.SignalMetrics,
|
||||
Aggregations: []qbtypes.MetricAggregation{
|
||||
{
|
||||
MetricName: "k8s.node.memory.working_set",
|
||||
TimeAggregation: metrictypes.TimeAggregationAvg,
|
||||
SpaceAggregation: metrictypes.SpaceAggregationSum,
|
||||
ReduceTo: qbtypes.ReduceToAvg,
|
||||
},
|
||||
},
|
||||
GroupBy: []qbtypes.GroupByKey{nodeNameGroupByKey},
|
||||
Disabled: false,
|
||||
},
|
||||
},
|
||||
// Query D: Memory allocatable. Same Latest caveat as Query B.
|
||||
{
|
||||
Type: qbtypes.QueryTypeBuilder,
|
||||
Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
|
||||
Name: "D",
|
||||
Signal: telemetrytypes.SignalMetrics,
|
||||
Aggregations: []qbtypes.MetricAggregation{
|
||||
{
|
||||
MetricName: "k8s.node.allocatable_memory",
|
||||
TimeAggregation: metrictypes.TimeAggregationLatest,
|
||||
SpaceAggregation: metrictypes.SpaceAggregationSum,
|
||||
ReduceTo: qbtypes.ReduceToAvg,
|
||||
},
|
||||
},
|
||||
GroupBy: []qbtypes.GroupByKey{nodeNameGroupByKey},
|
||||
Disabled: false,
|
||||
},
|
||||
},
|
||||
}
|
||||
|
||||
return &qbtypes.QueryRangeRequest{
|
||||
RequestType: qbtypes.RequestTypeScalar,
|
||||
CompositeQuery: qbtypes.CompositeQuery{
|
||||
Queries: queries,
|
||||
},
|
||||
}
|
||||
}
|
||||
334
pkg/modules/inframonitoring/implinframonitoring/pods.go
Normal file
334
pkg/modules/inframonitoring/implinframonitoring/pods.go
Normal file
@@ -0,0 +1,334 @@
|
||||
package implinframonitoring
|
||||
|
||||
import (
|
||||
"context"
|
||||
"fmt"
|
||||
"slices"
|
||||
"strings"
|
||||
"time"
|
||||
|
||||
"github.com/SigNoz/signoz/pkg/querybuilder"
|
||||
"github.com/SigNoz/signoz/pkg/telemetrymetrics"
|
||||
"github.com/SigNoz/signoz/pkg/types/inframonitoringtypes"
|
||||
"github.com/SigNoz/signoz/pkg/types/metrictypes"
|
||||
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
|
||||
"github.com/SigNoz/signoz/pkg/valuer"
|
||||
"github.com/huandu/go-sqlbuilder"
|
||||
)
|
||||
|
||||
// buildPodRecords assembles the page records. Phase counts come from
|
||||
// phaseCounts in both modes. In list mode (isPodUIDInGroupBy=true) each
|
||||
// group is one pod, so exactly one count is 1; PodPhase is derived from
|
||||
// which one. In grouped_list mode PodPhase stays PodPhaseNone.
|
||||
func buildPodRecords(
|
||||
isPodUIDInGroupBy bool,
|
||||
resp *qbtypes.QueryRangeResponse,
|
||||
pageGroups []map[string]string,
|
||||
groupBy []qbtypes.GroupByKey,
|
||||
metadataMap map[string]map[string]string,
|
||||
phaseCounts map[string]podPhaseCounts,
|
||||
reqEnd int64,
|
||||
) []inframonitoringtypes.PodRecord {
|
||||
metricsMap := parseFullQueryResponse(resp, groupBy)
|
||||
|
||||
records := make([]inframonitoringtypes.PodRecord, 0, len(pageGroups))
|
||||
for _, labels := range pageGroups {
|
||||
compositeKey := compositeKeyFromLabels(labels, groupBy)
|
||||
podUID := labels[podUIDAttrKey]
|
||||
|
||||
record := inframonitoringtypes.PodRecord{ // initialize with default values
|
||||
PodUID: podUID,
|
||||
PodPhase: inframonitoringtypes.PodPhaseNone,
|
||||
PodCPU: -1,
|
||||
PodCPURequest: -1,
|
||||
PodCPULimit: -1,
|
||||
PodMemory: -1,
|
||||
PodMemoryRequest: -1,
|
||||
PodMemoryLimit: -1,
|
||||
PodAge: -1,
|
||||
Meta: map[string]any{},
|
||||
}
|
||||
|
||||
if metrics, ok := metricsMap[compositeKey]; ok {
|
||||
if v, exists := metrics["A"]; exists {
|
||||
record.PodCPU = v
|
||||
}
|
||||
if v, exists := metrics["B"]; exists {
|
||||
record.PodCPURequest = v
|
||||
}
|
||||
if v, exists := metrics["C"]; exists {
|
||||
record.PodCPULimit = v
|
||||
}
|
||||
if v, exists := metrics["D"]; exists {
|
||||
record.PodMemory = v
|
||||
}
|
||||
if v, exists := metrics["E"]; exists {
|
||||
record.PodMemoryRequest = v
|
||||
}
|
||||
if v, exists := metrics["F"]; exists {
|
||||
record.PodMemoryLimit = v
|
||||
}
|
||||
}
|
||||
|
||||
if phaseCountsForGroup, ok := phaseCounts[compositeKey]; ok {
|
||||
record.PendingPodCount = phaseCountsForGroup.Pending
|
||||
record.RunningPodCount = phaseCountsForGroup.Running
|
||||
record.SucceededPodCount = phaseCountsForGroup.Succeeded
|
||||
record.FailedPodCount = phaseCountsForGroup.Failed
|
||||
record.UnknownPodCount = phaseCountsForGroup.Unknown
|
||||
|
||||
// In list mode each group is one pod; the count==1 bucket identifies the phase.
|
||||
if isPodUIDInGroupBy {
|
||||
switch {
|
||||
case phaseCountsForGroup.Pending == 1:
|
||||
record.PodPhase = inframonitoringtypes.PodPhasePending
|
||||
case phaseCountsForGroup.Running == 1:
|
||||
record.PodPhase = inframonitoringtypes.PodPhaseRunning
|
||||
case phaseCountsForGroup.Succeeded == 1:
|
||||
record.PodPhase = inframonitoringtypes.PodPhaseSucceeded
|
||||
case phaseCountsForGroup.Failed == 1:
|
||||
record.PodPhase = inframonitoringtypes.PodPhaseFailed
|
||||
case phaseCountsForGroup.Unknown == 1:
|
||||
record.PodPhase = inframonitoringtypes.PodPhaseUnknown
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
if attrs, ok := metadataMap[compositeKey]; ok && isPodUIDInGroupBy {
|
||||
// the condition above ensures we deduce age only if pod uid is in group by because if
|
||||
// it's not in group by then we might have multiple pod uids in the same group and hence then podAge wont make sense
|
||||
if startTimeStr, exists := attrs[podStartTimeAttrKey]; exists && startTimeStr != "" {
|
||||
if t, err := time.Parse(time.RFC3339, startTimeStr); err == nil {
|
||||
startTimeMs := t.UnixMilli()
|
||||
if startTimeMs > 0 {
|
||||
record.PodAge = reqEnd - startTimeMs
|
||||
}
|
||||
}
|
||||
}
|
||||
for k, v := range attrs {
|
||||
record.Meta[k] = v
|
||||
}
|
||||
}
|
||||
|
||||
records = append(records, record)
|
||||
}
|
||||
return records
|
||||
}
|
||||
|
||||
func (m *module) getTopPodGroups(
|
||||
ctx context.Context,
|
||||
orgID valuer.UUID,
|
||||
req *inframonitoringtypes.PostablePods,
|
||||
metadataMap map[string]map[string]string,
|
||||
) ([]map[string]string, error) {
|
||||
orderByKey := req.OrderBy.Key.Name
|
||||
queryNamesForOrderBy := orderByToPodsQueryNames[orderByKey]
|
||||
rankingQueryName := queryNamesForOrderBy[len(queryNamesForOrderBy)-1]
|
||||
|
||||
topReq := &qbtypes.QueryRangeRequest{
|
||||
Start: uint64(req.Start),
|
||||
End: uint64(req.End),
|
||||
RequestType: qbtypes.RequestTypeScalar,
|
||||
CompositeQuery: qbtypes.CompositeQuery{
|
||||
Queries: make([]qbtypes.QueryEnvelope, 0, len(queryNamesForOrderBy)),
|
||||
},
|
||||
}
|
||||
|
||||
for _, envelope := range m.newPodsTableListQuery().CompositeQuery.Queries {
|
||||
if !slices.Contains(queryNamesForOrderBy, envelope.GetQueryName()) {
|
||||
continue
|
||||
}
|
||||
copied := envelope
|
||||
if copied.Type == qbtypes.QueryTypeBuilder {
|
||||
existingExpr := ""
|
||||
if f := copied.GetFilter(); f != nil {
|
||||
existingExpr = f.Expression
|
||||
}
|
||||
reqFilterExpr := ""
|
||||
if req.Filter != nil {
|
||||
reqFilterExpr = req.Filter.Expression
|
||||
}
|
||||
merged := mergeFilterExpressions(existingExpr, reqFilterExpr)
|
||||
copied.SetFilter(&qbtypes.Filter{Expression: merged})
|
||||
copied.SetGroupBy(req.GroupBy)
|
||||
}
|
||||
topReq.CompositeQuery.Queries = append(topReq.CompositeQuery.Queries, copied)
|
||||
}
|
||||
|
||||
resp, err := m.querier.QueryRange(ctx, orgID, topReq)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
allMetricGroups := parseAndSortGroups(resp, rankingQueryName, req.GroupBy, req.OrderBy.Direction)
|
||||
return paginateWithBackfill(allMetricGroups, metadataMap, req.GroupBy, req.Offset, req.Limit), nil
|
||||
}
|
||||
|
||||
func (m *module) getPodsTableMetadata(ctx context.Context, req *inframonitoringtypes.PostablePods) (map[string]map[string]string, error) {
|
||||
var nonGroupByAttrs []string
|
||||
for _, key := range podAttrKeysForMetadata {
|
||||
if !isKeyInGroupByAttrs(req.GroupBy, key) {
|
||||
nonGroupByAttrs = append(nonGroupByAttrs, key)
|
||||
}
|
||||
}
|
||||
return m.getMetadata(ctx, podsTableMetricNamesList, req.GroupBy, nonGroupByAttrs, req.Filter, req.Start, req.End)
|
||||
}
|
||||
|
||||
// getPerGroupPodPhaseCounts computes per-group pod counts bucketed by each
|
||||
// pod's latest phase in the requested window.
|
||||
// Pipeline:
|
||||
//
|
||||
// timeSeriesFPs: fp ↔ (pod_uid, groupBy cols) from the time_series table.
|
||||
// User filter + page-groups filter applied here.
|
||||
// latestPhasePerPod: INNER JOIN samples × timeSeriesFPs, collapsed to
|
||||
// the latest phase per pod via argMax(value, unix_milli).
|
||||
// countPodsPerPhase: per-group uniqExactIf into 5 phase buckets.
|
||||
//
|
||||
// Groups absent from the result map have implicit zero counts (caller default).
|
||||
func (m *module) getPerGroupPodPhaseCounts(
|
||||
ctx context.Context,
|
||||
req *inframonitoringtypes.PostablePods,
|
||||
pageGroups []map[string]string,
|
||||
) (map[string]podPhaseCounts, error) {
|
||||
if len(pageGroups) == 0 || len(req.GroupBy) == 0 {
|
||||
return map[string]podPhaseCounts{}, nil
|
||||
}
|
||||
|
||||
// Merged filter expression (user filter + page-groups IN clauses).
|
||||
reqFilterExpr := ""
|
||||
if req.Filter != nil {
|
||||
reqFilterExpr = req.Filter.Expression
|
||||
}
|
||||
pageGroupsFilterExpr := buildPageGroupsFilterExpr(pageGroups)
|
||||
filterExpr := mergeFilterExpressions(reqFilterExpr, pageGroupsFilterExpr)
|
||||
|
||||
// Resolve tables. Same convention as hosts (distributed names from helpers).
|
||||
adjustedStart, adjustedEnd, _, localTimeSeriesTable := telemetrymetrics.WhichTSTableToUse(
|
||||
uint64(req.Start), uint64(req.End), nil,
|
||||
)
|
||||
samplesTable := telemetrymetrics.WhichSamplesTableToUse(
|
||||
uint64(req.Start), uint64(req.End),
|
||||
metrictypes.UnspecifiedType, metrictypes.TimeAggregationUnspecified, nil,
|
||||
)
|
||||
valueCol := telemetrymetrics.ValueColumnForSamplesTable(samplesTable)
|
||||
|
||||
// ----- timeSeriesFPs -----
|
||||
timeSeriesFPs := sqlbuilder.NewSelectBuilder()
|
||||
timeSeriesFPsSelectCols := []string{
|
||||
"fingerprint",
|
||||
fmt.Sprintf("JSONExtractString(labels, %s) AS pod_uid", timeSeriesFPs.Var(podUIDAttrKey)),
|
||||
}
|
||||
for _, key := range req.GroupBy {
|
||||
timeSeriesFPsSelectCols = append(timeSeriesFPsSelectCols,
|
||||
fmt.Sprintf("JSONExtractString(labels, %s) AS %s", timeSeriesFPs.Var(key.Name), quoteIdentifier(key.Name)),
|
||||
)
|
||||
}
|
||||
timeSeriesFPs.Select(timeSeriesFPsSelectCols...)
|
||||
timeSeriesFPs.From(fmt.Sprintf("%s.%s", telemetrymetrics.DBName, localTimeSeriesTable))
|
||||
timeSeriesFPs.Where(
|
||||
timeSeriesFPs.E("metric_name", podPhaseMetricName),
|
||||
timeSeriesFPs.GE("unix_milli", adjustedStart),
|
||||
timeSeriesFPs.L("unix_milli", adjustedEnd),
|
||||
)
|
||||
if filterExpr != "" {
|
||||
filterClause, err := m.buildFilterClause(ctx, &qbtypes.Filter{Expression: filterExpr}, req.Start, req.End)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
if filterClause != nil {
|
||||
timeSeriesFPs.AddWhereClause(filterClause)
|
||||
}
|
||||
}
|
||||
timeSeriesFPsGroupBy := []string{"fingerprint", "pod_uid"}
|
||||
for _, key := range req.GroupBy {
|
||||
timeSeriesFPsGroupBy = append(timeSeriesFPsGroupBy, quoteIdentifier(key.Name))
|
||||
}
|
||||
timeSeriesFPs.GroupBy(timeSeriesFPsGroupBy...)
|
||||
timeSeriesFPsSQL, timeSeriesFPsArgs := timeSeriesFPs.BuildWithFlavor(sqlbuilder.ClickHouse)
|
||||
|
||||
latestPhasePerPod := sqlbuilder.NewSelectBuilder()
|
||||
latestPhasePerPodSelectCols := []string{"tsfp.pod_uid AS pod_uid"}
|
||||
latestPhasePerPodGroupBy := []string{"pod_uid"}
|
||||
for _, key := range req.GroupBy {
|
||||
col := quoteIdentifier(key.Name)
|
||||
latestPhasePerPodSelectCols = append(latestPhasePerPodSelectCols, fmt.Sprintf("tsfp.%s AS %s", col, col))
|
||||
latestPhasePerPodGroupBy = append(latestPhasePerPodGroupBy, col)
|
||||
}
|
||||
latestPhasePerPodSelectCols = append(latestPhasePerPodSelectCols,
|
||||
fmt.Sprintf("argMax(samples.%s, samples.unix_milli) AS phase_value", valueCol),
|
||||
)
|
||||
latestPhasePerPod.Select(latestPhasePerPodSelectCols...)
|
||||
latestPhasePerPod.From(fmt.Sprintf(
|
||||
"%s.%s AS samples INNER JOIN time_series_fps AS tsfp ON samples.fingerprint = tsfp.fingerprint",
|
||||
telemetrymetrics.DBName, samplesTable,
|
||||
))
|
||||
latestPhasePerPod.Where(
|
||||
latestPhasePerPod.E("samples.metric_name", podPhaseMetricName),
|
||||
latestPhasePerPod.GE("samples.unix_milli", req.Start),
|
||||
latestPhasePerPod.L("samples.unix_milli", req.End),
|
||||
"tsfp.pod_uid != ''",
|
||||
)
|
||||
latestPhasePerPod.GroupBy(latestPhasePerPodGroupBy...)
|
||||
latestPhasePerPodSQL, latestPhasePerPodArgs := latestPhasePerPod.BuildWithFlavor(sqlbuilder.ClickHouse)
|
||||
|
||||
// ----- countPodsPerPhase (outer SELECT) -----
|
||||
countPodsPerPhaseSelectCols := make([]string, 0, len(req.GroupBy)+5)
|
||||
countPodsPerPhaseGroupBy := make([]string, 0, len(req.GroupBy))
|
||||
for _, key := range req.GroupBy {
|
||||
col := quoteIdentifier(key.Name)
|
||||
countPodsPerPhaseSelectCols = append(countPodsPerPhaseSelectCols, col)
|
||||
countPodsPerPhaseGroupBy = append(countPodsPerPhaseGroupBy, col)
|
||||
}
|
||||
countPodsPerPhaseSelectCols = append(countPodsPerPhaseSelectCols,
|
||||
fmt.Sprintf("uniqExactIf(pod_uid, phase_value = %d) AS pending_count", inframonitoringtypes.PodPhaseNumPending),
|
||||
fmt.Sprintf("uniqExactIf(pod_uid, phase_value = %d) AS running_count", inframonitoringtypes.PodPhaseNumRunning),
|
||||
fmt.Sprintf("uniqExactIf(pod_uid, phase_value = %d) AS succeeded_count", inframonitoringtypes.PodPhaseNumSucceeded),
|
||||
fmt.Sprintf("uniqExactIf(pod_uid, phase_value = %d) AS failed_count", inframonitoringtypes.PodPhaseNumFailed),
|
||||
fmt.Sprintf("uniqExactIf(pod_uid, phase_value = %d) AS unknown_count", inframonitoringtypes.PodPhaseNumUnknown),
|
||||
)
|
||||
countPodsPerPhaseSQL := fmt.Sprintf(
|
||||
"SELECT %s FROM latest_phase_per_pod GROUP BY %s",
|
||||
strings.Join(countPodsPerPhaseSelectCols, ", "),
|
||||
strings.Join(countPodsPerPhaseGroupBy, ", "),
|
||||
)
|
||||
|
||||
// Combine CTEs + outer.
|
||||
cteFragments := []string{
|
||||
fmt.Sprintf("time_series_fps AS (%s)", timeSeriesFPsSQL),
|
||||
fmt.Sprintf("latest_phase_per_pod AS (%s)", latestPhasePerPodSQL),
|
||||
}
|
||||
finalSQL := querybuilder.CombineCTEs(cteFragments) + countPodsPerPhaseSQL
|
||||
finalArgs := querybuilder.PrependArgs([][]any{timeSeriesFPsArgs, latestPhasePerPodArgs}, nil)
|
||||
|
||||
rows, err := m.telemetryStore.ClickhouseDB().Query(ctx, finalSQL, finalArgs...)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
defer rows.Close()
|
||||
|
||||
result := make(map[string]podPhaseCounts)
|
||||
for rows.Next() {
|
||||
groupVals := make([]string, len(req.GroupBy))
|
||||
scanPtrs := make([]any, 0, len(req.GroupBy)+5)
|
||||
for i := range groupVals {
|
||||
scanPtrs = append(scanPtrs, &groupVals[i])
|
||||
}
|
||||
var pending, running, succeeded, failed, unknown uint64
|
||||
scanPtrs = append(scanPtrs, &pending, &running, &succeeded, &failed, &unknown)
|
||||
|
||||
if err := rows.Scan(scanPtrs...); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
result[compositeKeyFromList(groupVals)] = podPhaseCounts{
|
||||
Pending: int(pending),
|
||||
Running: int(running),
|
||||
Succeeded: int(succeeded),
|
||||
Failed: int(failed),
|
||||
Unknown: int(unknown),
|
||||
}
|
||||
}
|
||||
if err := rows.Err(); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
return result, nil
|
||||
}
|
||||
@@ -0,0 +1,178 @@
|
||||
package implinframonitoring
|
||||
|
||||
import (
|
||||
"github.com/SigNoz/signoz/pkg/types/inframonitoringtypes"
|
||||
"github.com/SigNoz/signoz/pkg/types/metrictypes"
|
||||
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
|
||||
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
|
||||
)
|
||||
|
||||
const (
|
||||
podUIDAttrKey = "k8s.pod.uid"
|
||||
podStartTimeAttrKey = "k8s.pod.start_time"
|
||||
podPhaseMetricName = "k8s.pod.phase"
|
||||
)
|
||||
|
||||
var podUIDGroupByKey = qbtypes.GroupByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: podUIDAttrKey,
|
||||
FieldContext: telemetrytypes.FieldContextResource,
|
||||
FieldDataType: telemetrytypes.FieldDataTypeString,
|
||||
},
|
||||
}
|
||||
|
||||
var podsTableMetricNamesList = []string{
|
||||
"k8s.pod.cpu.usage",
|
||||
"k8s.pod.cpu_request_utilization",
|
||||
"k8s.pod.cpu_limit_utilization",
|
||||
"k8s.pod.memory.working_set",
|
||||
"k8s.pod.memory_request_utilization",
|
||||
"k8s.pod.memory_limit_utilization",
|
||||
"k8s.pod.phase",
|
||||
}
|
||||
|
||||
var podAttrKeysForMetadata = []string{
|
||||
"k8s.pod.uid",
|
||||
"k8s.pod.name",
|
||||
"k8s.namespace.name",
|
||||
"k8s.node.name",
|
||||
"k8s.deployment.name",
|
||||
"k8s.statefulset.name",
|
||||
"k8s.daemonset.name",
|
||||
"k8s.job.name",
|
||||
"k8s.cronjob.name",
|
||||
"k8s.cluster.name",
|
||||
"k8s.pod.start_time",
|
||||
}
|
||||
|
||||
var orderByToPodsQueryNames = map[string][]string{
|
||||
inframonitoringtypes.PodsOrderByCPU: {"A"},
|
||||
inframonitoringtypes.PodsOrderByCPURequest: {"B"},
|
||||
inframonitoringtypes.PodsOrderByCPULimit: {"C"},
|
||||
inframonitoringtypes.PodsOrderByMemory: {"D"},
|
||||
inframonitoringtypes.PodsOrderByMemoryRequest: {"E"},
|
||||
inframonitoringtypes.PodsOrderByMemoryLimit: {"F"},
|
||||
}
|
||||
|
||||
// newPodsTableListQuery builds the composite QB v5 request for the pods list.
|
||||
// Pod phase is derived separately via getPerGroupPodPhaseCounts (works for both
|
||||
// list and grouped_list modes), so no phase query is included here.
|
||||
func (m *module) newPodsTableListQuery() *qbtypes.QueryRangeRequest {
|
||||
queries := []qbtypes.QueryEnvelope{
|
||||
// Query A: CPU usage
|
||||
{
|
||||
Type: qbtypes.QueryTypeBuilder,
|
||||
Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
|
||||
Name: "A",
|
||||
Signal: telemetrytypes.SignalMetrics,
|
||||
Aggregations: []qbtypes.MetricAggregation{
|
||||
{
|
||||
MetricName: "k8s.pod.cpu.usage",
|
||||
TimeAggregation: metrictypes.TimeAggregationAvg,
|
||||
SpaceAggregation: metrictypes.SpaceAggregationSum,
|
||||
ReduceTo: qbtypes.ReduceToAvg,
|
||||
},
|
||||
},
|
||||
GroupBy: []qbtypes.GroupByKey{podUIDGroupByKey},
|
||||
Disabled: false,
|
||||
},
|
||||
},
|
||||
// Query B: CPU request utilization
|
||||
{
|
||||
Type: qbtypes.QueryTypeBuilder,
|
||||
Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
|
||||
Name: "B",
|
||||
Signal: telemetrytypes.SignalMetrics,
|
||||
Aggregations: []qbtypes.MetricAggregation{
|
||||
{
|
||||
MetricName: "k8s.pod.cpu_request_utilization",
|
||||
TimeAggregation: metrictypes.TimeAggregationAvg,
|
||||
SpaceAggregation: metrictypes.SpaceAggregationAvg,
|
||||
ReduceTo: qbtypes.ReduceToAvg,
|
||||
},
|
||||
},
|
||||
GroupBy: []qbtypes.GroupByKey{podUIDGroupByKey},
|
||||
Disabled: false,
|
||||
},
|
||||
},
|
||||
// Query C: CPU limit utilization
|
||||
{
|
||||
Type: qbtypes.QueryTypeBuilder,
|
||||
Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
|
||||
Name: "C",
|
||||
Signal: telemetrytypes.SignalMetrics,
|
||||
Aggregations: []qbtypes.MetricAggregation{
|
||||
{
|
||||
MetricName: "k8s.pod.cpu_limit_utilization",
|
||||
TimeAggregation: metrictypes.TimeAggregationAvg,
|
||||
SpaceAggregation: metrictypes.SpaceAggregationAvg,
|
||||
ReduceTo: qbtypes.ReduceToAvg,
|
||||
},
|
||||
},
|
||||
GroupBy: []qbtypes.GroupByKey{podUIDGroupByKey},
|
||||
Disabled: false,
|
||||
},
|
||||
},
|
||||
// Query D: Memory working set
|
||||
{
|
||||
Type: qbtypes.QueryTypeBuilder,
|
||||
Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
|
||||
Name: "D",
|
||||
Signal: telemetrytypes.SignalMetrics,
|
||||
Aggregations: []qbtypes.MetricAggregation{
|
||||
{
|
||||
MetricName: "k8s.pod.memory.working_set",
|
||||
TimeAggregation: metrictypes.TimeAggregationAvg,
|
||||
SpaceAggregation: metrictypes.SpaceAggregationSum,
|
||||
ReduceTo: qbtypes.ReduceToAvg,
|
||||
},
|
||||
},
|
||||
GroupBy: []qbtypes.GroupByKey{podUIDGroupByKey},
|
||||
Disabled: false,
|
||||
},
|
||||
},
|
||||
// Query E: Memory request utilization
|
||||
{
|
||||
Type: qbtypes.QueryTypeBuilder,
|
||||
Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
|
||||
Name: "E",
|
||||
Signal: telemetrytypes.SignalMetrics,
|
||||
Aggregations: []qbtypes.MetricAggregation{
|
||||
{
|
||||
MetricName: "k8s.pod.memory_request_utilization",
|
||||
TimeAggregation: metrictypes.TimeAggregationAvg,
|
||||
SpaceAggregation: metrictypes.SpaceAggregationAvg,
|
||||
ReduceTo: qbtypes.ReduceToAvg,
|
||||
},
|
||||
},
|
||||
GroupBy: []qbtypes.GroupByKey{podUIDGroupByKey},
|
||||
Disabled: false,
|
||||
},
|
||||
},
|
||||
// Query F: Memory limit utilization
|
||||
{
|
||||
Type: qbtypes.QueryTypeBuilder,
|
||||
Spec: qbtypes.QueryBuilderQuery[qbtypes.MetricAggregation]{
|
||||
Name: "F",
|
||||
Signal: telemetrytypes.SignalMetrics,
|
||||
Aggregations: []qbtypes.MetricAggregation{
|
||||
{
|
||||
MetricName: "k8s.pod.memory_limit_utilization",
|
||||
TimeAggregation: metrictypes.TimeAggregationAvg,
|
||||
SpaceAggregation: metrictypes.SpaceAggregationAvg,
|
||||
ReduceTo: qbtypes.ReduceToAvg,
|
||||
},
|
||||
},
|
||||
GroupBy: []qbtypes.GroupByKey{podUIDGroupByKey},
|
||||
Disabled: false,
|
||||
},
|
||||
},
|
||||
}
|
||||
|
||||
return &qbtypes.QueryRangeRequest{
|
||||
RequestType: qbtypes.RequestTypeScalar,
|
||||
CompositeQuery: qbtypes.CompositeQuery{
|
||||
Queries: queries,
|
||||
},
|
||||
}
|
||||
}
|
||||
@@ -10,8 +10,12 @@ import (
|
||||
|
||||
type Handler interface {
|
||||
ListHosts(http.ResponseWriter, *http.Request)
|
||||
ListPods(http.ResponseWriter, *http.Request)
|
||||
ListNodes(http.ResponseWriter, *http.Request)
|
||||
}
|
||||
|
||||
type Module interface {
|
||||
ListHosts(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableHosts) (*inframonitoringtypes.Hosts, error)
|
||||
ListPods(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostablePods) (*inframonitoringtypes.Pods, error)
|
||||
ListNodes(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableNodes) (*inframonitoringtypes.Nodes, error)
|
||||
}
|
||||
|
||||
@@ -124,6 +124,17 @@ func CountExpressionForSamplesTable(tableName string) string {
|
||||
return "sum(count)"
|
||||
}
|
||||
|
||||
// ValueColumnForSamplesTable returns the column name holding the sample value:
|
||||
// "last" for the 5m/30m aggregated tables, "value" otherwise.
|
||||
// note all the other columns in the aggregated samples tables are nothing but aggregations.
|
||||
// and so "last" is the value column for these tables.
|
||||
func ValueColumnForSamplesTable(tableName string) string {
|
||||
if tableName == SamplesV4Agg5mTableName || tableName == SamplesV4Agg30mTableName {
|
||||
return "last"
|
||||
}
|
||||
return "value"
|
||||
}
|
||||
|
||||
// start and end are in milliseconds
|
||||
// we have three tables for samples
|
||||
// 1. distributed_samples_v4
|
||||
|
||||
@@ -49,7 +49,7 @@ type HostFilter struct {
|
||||
FilterByStatus HostStatus `json:"filterByStatus"`
|
||||
}
|
||||
|
||||
// Validate ensures HostsListRequest contains acceptable values.
|
||||
// Validate ensures PostableHosts contains acceptable values.
|
||||
func (req *PostableHosts) Validate() error {
|
||||
if req == nil {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "request is nil")
|
||||
|
||||
103
pkg/types/inframonitoringtypes/nodes.go
Normal file
103
pkg/types/inframonitoringtypes/nodes.go
Normal file
@@ -0,0 +1,103 @@
|
||||
package inframonitoringtypes
|
||||
|
||||
import (
|
||||
"encoding/json"
|
||||
"slices"
|
||||
|
||||
"github.com/SigNoz/signoz/pkg/errors"
|
||||
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
|
||||
)
|
||||
|
||||
type Nodes struct {
|
||||
Type ResponseType `json:"type" required:"true"`
|
||||
Records []NodeRecord `json:"records" required:"true"`
|
||||
Total int `json:"total" required:"true"`
|
||||
RequiredMetricsCheck RequiredMetricsCheck `json:"requiredMetricsCheck" required:"true"`
|
||||
EndTimeBeforeRetention bool `json:"endTimeBeforeRetention" required:"true"`
|
||||
Warning *qbtypes.QueryWarnData `json:"warning,omitempty"`
|
||||
}
|
||||
|
||||
type NodeRecord struct {
|
||||
NodeName string `json:"nodeName" required:"true"`
|
||||
Condition NodeCondition `json:"condition" required:"true"`
|
||||
ReadyNodesCount int `json:"readyNodesCount" required:"true"`
|
||||
NotReadyNodesCount int `json:"notReadyNodesCount" required:"true"`
|
||||
NodeCPU float64 `json:"nodeCPU" required:"true"`
|
||||
NodeCPUAllocatable float64 `json:"nodeCPUAllocatable" required:"true"`
|
||||
NodeMemory float64 `json:"nodeMemory" required:"true"`
|
||||
NodeMemoryAllocatable float64 `json:"nodeMemoryAllocatable" required:"true"`
|
||||
Meta map[string]interface{} `json:"meta" required:"true"`
|
||||
}
|
||||
|
||||
// PostableNodes is the request body for the v2 nodes list API.
|
||||
type PostableNodes struct {
|
||||
Start int64 `json:"start" required:"true"`
|
||||
End int64 `json:"end" required:"true"`
|
||||
Filter *qbtypes.Filter `json:"filter"`
|
||||
GroupBy []qbtypes.GroupByKey `json:"groupBy"`
|
||||
OrderBy *qbtypes.OrderBy `json:"orderBy"`
|
||||
Offset int `json:"offset"`
|
||||
Limit int `json:"limit" required:"true"`
|
||||
}
|
||||
|
||||
// Validate ensures PostableNodes contains acceptable values.
|
||||
func (req *PostableNodes) Validate() error {
|
||||
if req == nil {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "request is nil")
|
||||
}
|
||||
|
||||
if req.Start <= 0 {
|
||||
return errors.NewInvalidInputf(
|
||||
errors.CodeInvalidInput,
|
||||
"invalid start time %d: start must be greater than 0",
|
||||
req.Start,
|
||||
)
|
||||
}
|
||||
|
||||
if req.End <= 0 {
|
||||
return errors.NewInvalidInputf(
|
||||
errors.CodeInvalidInput,
|
||||
"invalid end time %d: end must be greater than 0",
|
||||
req.End,
|
||||
)
|
||||
}
|
||||
|
||||
if req.Start >= req.End {
|
||||
return errors.NewInvalidInputf(
|
||||
errors.CodeInvalidInput,
|
||||
"invalid time range: start (%d) must be less than end (%d)",
|
||||
req.Start,
|
||||
req.End,
|
||||
)
|
||||
}
|
||||
|
||||
if req.Limit < 1 || req.Limit > 5000 {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "limit must be between 1 and 5000")
|
||||
}
|
||||
|
||||
if req.Offset < 0 {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")
|
||||
}
|
||||
|
||||
if req.OrderBy != nil {
|
||||
if !slices.Contains(NodesValidOrderByKeys, req.OrderBy.Key.Name) {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)
|
||||
}
|
||||
if req.OrderBy.Direction != qbtypes.OrderDirectionAsc && req.OrderBy.Direction != qbtypes.OrderDirectionDesc {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by direction: %s", req.OrderBy.Direction)
|
||||
}
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
// UnmarshalJSON validates input immediately after decoding.
|
||||
func (req *PostableNodes) UnmarshalJSON(data []byte) error {
|
||||
type raw PostableNodes
|
||||
var decoded raw
|
||||
if err := json.Unmarshal(data, &decoded); err != nil {
|
||||
return err
|
||||
}
|
||||
*req = PostableNodes(decoded)
|
||||
return req.Validate()
|
||||
}
|
||||
42
pkg/types/inframonitoringtypes/nodes_constants.go
Normal file
42
pkg/types/inframonitoringtypes/nodes_constants.go
Normal file
@@ -0,0 +1,42 @@
|
||||
package inframonitoringtypes
|
||||
|
||||
import "github.com/SigNoz/signoz/pkg/valuer"
|
||||
|
||||
type NodeCondition struct {
|
||||
valuer.String
|
||||
}
|
||||
|
||||
var (
|
||||
NodeConditionReady = NodeCondition{valuer.NewString("ready")}
|
||||
NodeConditionNotReady = NodeCondition{valuer.NewString("not_ready")}
|
||||
NodeConditionNone = NodeCondition{valuer.NewString("")}
|
||||
)
|
||||
|
||||
func (NodeCondition) Enum() []any {
|
||||
return []any{
|
||||
NodeConditionReady,
|
||||
NodeConditionNotReady,
|
||||
NodeConditionNone,
|
||||
}
|
||||
}
|
||||
|
||||
// Numeric values emitted by the k8s.node.condition_ready metric
|
||||
// (source: OTel kubeletstats receiver).
|
||||
const (
|
||||
NodeConditionNumReady = 1
|
||||
NodeConditionNumNotReady = 0
|
||||
)
|
||||
|
||||
const (
|
||||
NodesOrderByCPU = "cpu"
|
||||
NodesOrderByCPUAllocatable = "cpu_allocatable"
|
||||
NodesOrderByMemory = "memory"
|
||||
NodesOrderByMemoryAllocatable = "memory_allocatable"
|
||||
)
|
||||
|
||||
var NodesValidOrderByKeys = []string{
|
||||
NodesOrderByCPU,
|
||||
NodesOrderByCPUAllocatable,
|
||||
NodesOrderByMemory,
|
||||
NodesOrderByMemoryAllocatable,
|
||||
}
|
||||
255
pkg/types/inframonitoringtypes/nodes_test.go
Normal file
255
pkg/types/inframonitoringtypes/nodes_test.go
Normal file
@@ -0,0 +1,255 @@
|
||||
package inframonitoringtypes
|
||||
|
||||
import (
|
||||
"testing"
|
||||
|
||||
"github.com/SigNoz/signoz/pkg/errors"
|
||||
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
|
||||
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
|
||||
"github.com/SigNoz/signoz/pkg/valuer"
|
||||
"github.com/stretchr/testify/require"
|
||||
)
|
||||
|
||||
func TestPostableNodes_Validate(t *testing.T) {
|
||||
tests := []struct {
|
||||
name string
|
||||
req *PostableNodes
|
||||
wantErr bool
|
||||
}{
|
||||
{
|
||||
name: "valid request",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: false,
|
||||
},
|
||||
{
|
||||
name: "nil request",
|
||||
req: nil,
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "start time zero",
|
||||
req: &PostableNodes{
|
||||
Start: 0,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "start time negative",
|
||||
req: &PostableNodes{
|
||||
Start: -1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "end time zero",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 0,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "start time greater than end time",
|
||||
req: &PostableNodes{
|
||||
Start: 2000,
|
||||
End: 1000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "start time equal to end time",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 1000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "limit zero",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 0,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "limit negative",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: -10,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "limit exceeds max",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 5001,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "offset negative",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: -5,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "orderBy nil is valid",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: false,
|
||||
},
|
||||
{
|
||||
name: "orderBy with valid key cpu and direction asc",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
OrderBy: &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: NodesOrderByCPU,
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirectionAsc,
|
||||
},
|
||||
},
|
||||
wantErr: false,
|
||||
},
|
||||
{
|
||||
name: "orderBy with valid key cpu_allocatable and direction desc",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
OrderBy: &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: NodesOrderByCPUAllocatable,
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirectionDesc,
|
||||
},
|
||||
},
|
||||
wantErr: false,
|
||||
},
|
||||
{
|
||||
name: "orderBy with valid key memory_allocatable and direction asc",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
OrderBy: &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: NodesOrderByMemoryAllocatable,
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirectionAsc,
|
||||
},
|
||||
},
|
||||
wantErr: false,
|
||||
},
|
||||
{
|
||||
name: "orderBy with condition key is rejected",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
OrderBy: &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: "condition",
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirectionDesc,
|
||||
},
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "orderBy with invalid key",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
OrderBy: &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: "unknown",
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirectionDesc,
|
||||
},
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "orderBy with valid key but invalid direction",
|
||||
req: &PostableNodes{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
OrderBy: &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: NodesOrderByMemory,
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirection{String: valuer.NewString("invalid")},
|
||||
},
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
}
|
||||
|
||||
for _, tt := range tests {
|
||||
t.Run(tt.name, func(t *testing.T) {
|
||||
err := tt.req.Validate()
|
||||
if tt.wantErr {
|
||||
require.Error(t, err)
|
||||
require.True(t, errors.Ast(err, errors.TypeInvalidInput), "expected error to be of type InvalidInput")
|
||||
} else {
|
||||
require.NoError(t, err)
|
||||
}
|
||||
})
|
||||
}
|
||||
}
|
||||
109
pkg/types/inframonitoringtypes/pods.go
Normal file
109
pkg/types/inframonitoringtypes/pods.go
Normal file
@@ -0,0 +1,109 @@
|
||||
package inframonitoringtypes
|
||||
|
||||
import (
|
||||
"encoding/json"
|
||||
"slices"
|
||||
|
||||
"github.com/SigNoz/signoz/pkg/errors"
|
||||
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
|
||||
)
|
||||
|
||||
type Pods struct {
|
||||
Type ResponseType `json:"type" required:"true"`
|
||||
Records []PodRecord `json:"records" required:"true"`
|
||||
Total int `json:"total" required:"true"`
|
||||
RequiredMetricsCheck RequiredMetricsCheck `json:"requiredMetricsCheck" required:"true"`
|
||||
EndTimeBeforeRetention bool `json:"endTimeBeforeRetention" required:"true"`
|
||||
Warning *qbtypes.QueryWarnData `json:"warning,omitempty"`
|
||||
}
|
||||
|
||||
type PodRecord struct {
|
||||
PodUID string `json:"podUID" required:"true"`
|
||||
PodCPU float64 `json:"podCPU" required:"true"`
|
||||
PodCPURequest float64 `json:"podCPURequest" required:"true"`
|
||||
PodCPULimit float64 `json:"podCPULimit" required:"true"`
|
||||
PodMemory float64 `json:"podMemory" required:"true"`
|
||||
PodMemoryRequest float64 `json:"podMemoryRequest" required:"true"`
|
||||
PodMemoryLimit float64 `json:"podMemoryLimit" required:"true"`
|
||||
PodPhase PodPhase `json:"podPhase" required:"true"`
|
||||
PendingPodCount int `json:"pendingPodCount" required:"true"`
|
||||
RunningPodCount int `json:"runningPodCount" required:"true"`
|
||||
SucceededPodCount int `json:"succeededPodCount" required:"true"`
|
||||
FailedPodCount int `json:"failedPodCount" required:"true"`
|
||||
UnknownPodCount int `json:"unknownPodCount" required:"true"`
|
||||
PodAge int64 `json:"podAge" required:"true"`
|
||||
Meta map[string]interface{} `json:"meta" required:"true"`
|
||||
}
|
||||
|
||||
// PostablePods is the request body for the v2 pods list API.
|
||||
type PostablePods struct {
|
||||
Start int64 `json:"start" required:"true"`
|
||||
End int64 `json:"end" required:"true"`
|
||||
Filter *qbtypes.Filter `json:"filter"`
|
||||
GroupBy []qbtypes.GroupByKey `json:"groupBy"`
|
||||
OrderBy *qbtypes.OrderBy `json:"orderBy"`
|
||||
Offset int `json:"offset"`
|
||||
Limit int `json:"limit" required:"true"`
|
||||
}
|
||||
|
||||
// Validate ensures PostablePods contains acceptable values.
|
||||
func (req *PostablePods) Validate() error {
|
||||
if req == nil {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "request is nil")
|
||||
}
|
||||
|
||||
if req.Start <= 0 {
|
||||
return errors.NewInvalidInputf(
|
||||
errors.CodeInvalidInput,
|
||||
"invalid start time %d: start must be greater than 0",
|
||||
req.Start,
|
||||
)
|
||||
}
|
||||
|
||||
if req.End <= 0 {
|
||||
return errors.NewInvalidInputf(
|
||||
errors.CodeInvalidInput,
|
||||
"invalid end time %d: end must be greater than 0",
|
||||
req.End,
|
||||
)
|
||||
}
|
||||
|
||||
if req.Start >= req.End {
|
||||
return errors.NewInvalidInputf(
|
||||
errors.CodeInvalidInput,
|
||||
"invalid time range: start (%d) must be less than end (%d)",
|
||||
req.Start,
|
||||
req.End,
|
||||
)
|
||||
}
|
||||
|
||||
if req.Limit < 1 || req.Limit > 5000 {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "limit must be between 1 and 5000")
|
||||
}
|
||||
|
||||
if req.Offset < 0 {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "offset cannot be negative")
|
||||
}
|
||||
|
||||
if req.OrderBy != nil {
|
||||
if !slices.Contains(PodsValidOrderByKeys, req.OrderBy.Key.Name) {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by key: %s", req.OrderBy.Key.Name)
|
||||
}
|
||||
if req.OrderBy.Direction != qbtypes.OrderDirectionAsc && req.OrderBy.Direction != qbtypes.OrderDirectionDesc {
|
||||
return errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid order by direction: %s", req.OrderBy.Direction)
|
||||
}
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
// UnmarshalJSON validates input immediately after decoding.
|
||||
func (req *PostablePods) UnmarshalJSON(data []byte) error {
|
||||
type raw PostablePods
|
||||
var decoded raw
|
||||
if err := json.Unmarshal(data, &decoded); err != nil {
|
||||
return err
|
||||
}
|
||||
*req = PostablePods(decoded)
|
||||
return req.Validate()
|
||||
}
|
||||
55
pkg/types/inframonitoringtypes/pods_constants.go
Normal file
55
pkg/types/inframonitoringtypes/pods_constants.go
Normal file
@@ -0,0 +1,55 @@
|
||||
package inframonitoringtypes
|
||||
|
||||
import "github.com/SigNoz/signoz/pkg/valuer"
|
||||
|
||||
type PodPhase struct {
|
||||
valuer.String
|
||||
}
|
||||
|
||||
var (
|
||||
PodPhasePending = PodPhase{valuer.NewString("pending")}
|
||||
PodPhaseRunning = PodPhase{valuer.NewString("running")}
|
||||
PodPhaseSucceeded = PodPhase{valuer.NewString("succeeded")}
|
||||
PodPhaseFailed = PodPhase{valuer.NewString("failed")}
|
||||
PodPhaseUnknown = PodPhase{valuer.NewString("unknown")}
|
||||
PodPhaseNone = PodPhase{valuer.NewString("")}
|
||||
)
|
||||
|
||||
func (PodPhase) Enum() []any {
|
||||
return []any{
|
||||
PodPhasePending,
|
||||
PodPhaseRunning,
|
||||
PodPhaseSucceeded,
|
||||
PodPhaseFailed,
|
||||
PodPhaseUnknown,
|
||||
PodPhaseNone,
|
||||
}
|
||||
}
|
||||
|
||||
// Numeric pod phase values emitted by the k8s.pod.phase metric
|
||||
// (source: OTel kubeletstats receiver).
|
||||
const (
|
||||
PodPhaseNumPending = 1
|
||||
PodPhaseNumRunning = 2
|
||||
PodPhaseNumSucceeded = 3
|
||||
PodPhaseNumFailed = 4
|
||||
PodPhaseNumUnknown = 5
|
||||
)
|
||||
|
||||
const (
|
||||
PodsOrderByCPU = "cpu"
|
||||
PodsOrderByCPURequest = "cpu_request"
|
||||
PodsOrderByCPULimit = "cpu_limit"
|
||||
PodsOrderByMemory = "memory"
|
||||
PodsOrderByMemoryRequest = "memory_request"
|
||||
PodsOrderByMemoryLimit = "memory_limit"
|
||||
)
|
||||
|
||||
var PodsValidOrderByKeys = []string{
|
||||
PodsOrderByCPU,
|
||||
PodsOrderByCPURequest,
|
||||
PodsOrderByCPULimit,
|
||||
PodsOrderByMemory,
|
||||
PodsOrderByMemoryRequest,
|
||||
PodsOrderByMemoryLimit,
|
||||
}
|
||||
219
pkg/types/inframonitoringtypes/pods_test.go
Normal file
219
pkg/types/inframonitoringtypes/pods_test.go
Normal file
@@ -0,0 +1,219 @@
|
||||
package inframonitoringtypes
|
||||
|
||||
import (
|
||||
"testing"
|
||||
|
||||
"github.com/SigNoz/signoz/pkg/errors"
|
||||
qbtypes "github.com/SigNoz/signoz/pkg/types/querybuildertypes/querybuildertypesv5"
|
||||
"github.com/SigNoz/signoz/pkg/types/telemetrytypes"
|
||||
"github.com/SigNoz/signoz/pkg/valuer"
|
||||
"github.com/stretchr/testify/require"
|
||||
)
|
||||
|
||||
func TestPostablePods_Validate(t *testing.T) {
|
||||
tests := []struct {
|
||||
name string
|
||||
req *PostablePods
|
||||
wantErr bool
|
||||
}{
|
||||
{
|
||||
name: "valid request",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: false,
|
||||
},
|
||||
{
|
||||
name: "nil request",
|
||||
req: nil,
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "start time zero",
|
||||
req: &PostablePods{
|
||||
Start: 0,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "start time negative",
|
||||
req: &PostablePods{
|
||||
Start: -1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "end time zero",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 0,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "start time greater than end time",
|
||||
req: &PostablePods{
|
||||
Start: 2000,
|
||||
End: 1000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "start time equal to end time",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 1000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "limit zero",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 0,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "limit negative",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: -10,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "limit exceeds max",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 5001,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "offset negative",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: -5,
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "orderBy nil is valid",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
},
|
||||
wantErr: false,
|
||||
},
|
||||
{
|
||||
name: "orderBy with valid key cpu and direction asc",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
OrderBy: &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: PodsOrderByCPU,
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirectionAsc,
|
||||
},
|
||||
},
|
||||
wantErr: false,
|
||||
},
|
||||
{
|
||||
name: "orderBy with phase key is rejected",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
OrderBy: &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: "phase",
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirectionDesc,
|
||||
},
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "orderBy with invalid key",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
OrderBy: &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: "unknown",
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirectionDesc,
|
||||
},
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
{
|
||||
name: "orderBy with valid key but invalid direction",
|
||||
req: &PostablePods{
|
||||
Start: 1000,
|
||||
End: 2000,
|
||||
Limit: 100,
|
||||
Offset: 0,
|
||||
OrderBy: &qbtypes.OrderBy{
|
||||
Key: qbtypes.OrderByKey{
|
||||
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
|
||||
Name: PodsOrderByMemory,
|
||||
},
|
||||
},
|
||||
Direction: qbtypes.OrderDirection{String: valuer.NewString("invalid")},
|
||||
},
|
||||
},
|
||||
wantErr: true,
|
||||
},
|
||||
}
|
||||
|
||||
for _, tt := range tests {
|
||||
t.Run(tt.name, func(t *testing.T) {
|
||||
err := tt.req.Validate()
|
||||
if tt.wantErr {
|
||||
require.Error(t, err)
|
||||
require.True(t, errors.Ast(err, errors.TypeInvalidInput), "expected error to be of type InvalidInput")
|
||||
} else {
|
||||
require.NoError(t, err)
|
||||
}
|
||||
})
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user