You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The /system.slice/var-lib-docker-containers-<container_id>-shm.mount version either is always 0 or exactly the same as the other version of the same stat with id "/docker/<container_id>"
These ones always return 0 for the shm.mount version:
container_cpu_*
container_memory_*
Can we either automatically ignore the redundant shm.mount version and add some way to filter these out so I'm not storing unneeded timeseries data in prometheus. cadvisor.txt
The text was updated successfully, but these errors were encountered:
See attached file. Basically cadvisor is gathering stats about the same container twice:
container_cpu_system_seconds_total{id="/docker/dfae05cccdaa4c7b38c0f19cbe8b13674474f81bc2454956dfa08b04464a2115",image="google/cadvisor:v0.24.0",name="cadvisor"} 243.51
container_cpu_system_seconds_total{id="/system.slice/var-lib-docker-containers-dfae05cccdaa4c7b38c0f19cbe8b13674474f81bc2454956dfa08b04464a2115-shm.mount",image="google/cadvisor:v0.24.0",name="cadvisor"} 0
The /system.slice/var-lib-docker-containers-<container_id>-shm.mount version either is always 0 or exactly the same as the other version of the same stat with id "/docker/<container_id>"
The examples where it's exactly the same:
container_fs_usage_bytes
container_last_seen
container_network_*
container_spec_cpu_*
container_start_time_seconds
These ones always return 0 for the shm.mount version:
container_cpu_*
container_memory_*
Can we either automatically ignore the redundant shm.mount version and add some way to filter these out so I'm not storing unneeded timeseries data in prometheus.
cadvisor.txt
The text was updated successfully, but these errors were encountered: