1apiVersion: opentelemetry.io/v1beta1
2kind: OpenTelemetryCollector
3metadata:
4 name: collector-with-ta-prometheus-cr
5spec:
6 serviceAccount: opentelemetry-targetallocator-sa
7 priorityClassName: edge-p4-operability-services
8 image: us-east1-docker.pkg.dev/ret-edge-pltf-infra/thirdparty/otel/opentelemetry-collector-contrib@sha256:ff3b385fc8a4bc7709e2bd8b4024259b649f78732aecc874a963250b5a5c3dd4 #TBD: Fix bzl path automation
9 resources:
10 requests:
11 cpu: 25m
12 memory: 128Mi
13 config:
14 service:
15 pipelines:
16 metrics:
17 exporters:
18 - googlemanagedprometheus
19 processors:
20 - attributes
21 - resource
22 - groupbyattrs
23 - resourcedetection
24 - memory_limiter
25 - batch
26 receivers:
27 - prometheus
28 telemetry:
29 logs:
30 encoding: "json"
31 level: "WARN"
32 exporters:
33 googlemanagedprometheus:
34 metric:
35 prefix: prometheus.googleapis.com
36 compression: gzip
37 processors:
38 attributes:
39 actions:
40 - value: ${cluster_hash}
41 action: upsert
42 key: cluster_hash
43 - value: ${cluster_uuid}
44 action: upsert
45 key: cluster_name
46 - value: ${cluster_provider}
47 action: upsert
48 key: cluster_type
49 batch:
50 # batch metrics before sending to reduce API usage
51 send_batch_max_size: 0
52 send_batch_size: 10000
53 timeout: 10s
54 groupbyattrs:
55 keys:
56 - cluster
57 - location
58 - project_id
59 - namespace
60 memory_limiter:
61 # drop metrics if memory usage gets too high
62 check_interval: 5s
63 limit_mib: 4096
64 resource:
65 attributes:
66 - value: ${cluster_uuid}
67 action: upsert
68 key: cluster
69 - value: ${gcp_region}-${gcp_zone}
70 action: upsert
71 key: location
72 - value: ${gcp_project_id}
73 action: upsert
74 key: project_id
75 resourcedetection:
76 detectors:
77 - gcp
78 override: false
79 timeout: 5s
80 receivers:
81 prometheus:
82 config:
83 scrape_configs:
84 - job_name: otel-collector
85 scrape_interval: 60s
86 static_configs:
87 - targets:
88 - 0.0.0.0:8888
89 report_extra_scrape_metrics: true
90 target_allocator:
91 collector_id: "$${POD_NAME}"
92 endpoint: http://collector-with-ta-prometheus-cr-targetallocator
93 interval: 60s
94 managementState: managed
95 mode: statefulset
96 podDisruptionBudget:
97 minAvailable: 1
98 targetAllocator:
99 serviceAccount: opentelemetry-targetallocator-sa
100 image: us-east1-docker.pkg.dev/ret-edge-pltf-infra/thirdparty/ghcr.io/open-telemetry/opentelemetry-operator/target-allocator@sha256:c490ab98f73f683d11af86ac2b0c69be89bc787abfcb38525f870a7d377173da #TBD: Fix bzl path automation
101 resources:
102 requests:
103 cpu: 10m
104 memory: 10Mi
105 enabled: true
106 observability:
107 metrics:
108 enableMetrics: true
109 prometheusCR:
110 enabled: true
111 podMonitorSelector: {}
112 scrapeInterval: 1m0s
113 serviceMonitorSelector: {}
114 upgradeStrategy: automatic
View as plain text