LiveHard

  • 홈
  • 태그
  • 방명록

servicemonitor 2

grafana metric 수집이슈

grafana exporter 및 servicemonitor 셋업을 하고 prometheus에 잘수집되는지 확인해보았더니 다음과같은 에러가 났다. x509: certificate signed by unknown authority for prometheus Hi, 사실 이 에러자체는 prometheus에 tls insecure설정을 함으로써 회피가 가능하나. 나는 그것외에도 특이한점을 발견했다. endpoint는 pod ip:pod port/metrics 로 되어있는데 ERROR로그에는 Get :/metrics로 redirect되어있는것이다. 이로인해 발생한 에러였고, 이 원인을 먼저확인해야했다. k8s서버에서 curl pod ip:pod port를 날리면 아래와 같이 redirect 경로를 return해..

k8s 및 인프라/prometheus 2024.03.15

prometheus operator

prometheus의 단점은 polling방식의 수집으로 인해서, 새로운 exporter가 추가될 경우, prometheus서버의 config 또는 configmap의 해당 exporter수집경로를 등록하고, prometheus를 재시작해야한다는점이다. prometheus operator를 통해서 해결할수있다. 기존에는 app -> exporter -> prometehus서버 순으로 수집된다면, app->exporter->servicemonitor->prometheus 서버로 수집된다. 조금더 보면 servicemonitor는 k8s CRD 리소스로 샘플 매니페스트는 다음과같다. apiVersion: monitoring.coreos.com/v1 kind: ServiceMonitor metadata: na..

k8s 및 인프라/prometheus 2024.03.15
이전
1
다음
더보기
프로필사진

LiveHard

250x250
  • 분류 전체보기 (29)
    • network (0)
    • python (0)
      • fastapi (0)
    • k8s 및 인프라 (13)
      • k8s (4)
      • istio (1)
      • certmanager (0)
      • keycloak (0)
      • prometheus (4)
      • rook-ceph (4)
    • CICD (0)
    • MLOps (6)
      • GPU (5)
      • Langserve (1)
    • airflow (5)
    • kubeflow (1)
    • minIO (3)
    • redis(keydb) (1)
    • 일상 (0)

Tag

S3, Prometheus, GPU Operator, rook-ceph, istio, object storage, vGPU, MinIO, ceph, gfd, servicemonitor, k8s, airflow, prometheus operator, Metric, TaskFlow, helm, nvidia, rdma, gpu-operator,

최근글과 인기글

  • 최근글
  • 인기글

최근댓글

공지사항

페이스북 트위터 플러그인

  • Facebook
  • Twitter

Archives

Calendar

«   2025/07   »
일 월 화 수 목 금 토
1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31

방문자수Total

  • Today :
  • Yesterday :

Copyright © Kakao Corp. All rights reserved.

티스토리툴바