My GKE clusters are not monitored by Stackdriver, despite having Cloud Monitoring enabled. In particular https://app.google.stackdriver.com/gke simply reports "You do not have any resources of this type being monitored by Stackdriver".
- Clusters running Kubernetes 1.2.4, created in mid-May 2016.
- Some had Cloud Monitoring enabled on creation, some had it enabled via
gcloud container clusters update serve --monitoring-service=monitoring.googleapis.com. - All clusters are running the
heapsterpod with 2/2 ready, status of Running, no restarts e.g. asheapster-v1.0.2-594732231-76sdj. - All clusters are running the
heapsterservice with a cluster IP, no external IP and port 80/TCP. - Logs of
heapsterpod,heapstercontainer look like:
I0609 09:15:05.000189 1 manager.go:79] Scraping metrics start: 2016-06-09 09:14:00 +0000 UTC, end: 2016-06-09 09:15:00 +0000 UTC
I0609 09:15:05.023729 1 manager.go:152] ScrapeMetrics: time: 23.435264ms size: 25
- Logs of
heapsterpod,heapster-nannycontainer look like:
I0609 09:18:03.968201 1 nanny_lib.go:90] The number of nodes is 1
I0609 09:18:03.971459 1 nanny_lib.go:98] The container resources are &{map[memory:{213909504.000000000 BinarySI} cpu:{0.100000000 DecimalSI}] map[cpu:{0.100000000 DecimalSI} memory:{213909504.000000000 BinarySI}]}
I0609 09:18:03.971522 1 nanny_lib.go:102] The expected resources are &{map[cpu:{0.100000000 DecimalSI} memory:{218103808.000000000 BinarySI}] map[cpu:{0.100000000 DecimalSI} memory:{218103808.000000000 BinarySI}]}
I0609 09:18:03.971634 1 nanny_lib.go:106] Resources are within the expected limits.
- https://status.cloud.google.com shows all services green.
