Adjust dropped metrics from cAdvisor

This change drops pod-centric metrics without a non-empty 'container' label.

Previously we dropped pod-centric metrics without a (pod, namespace) label set
however these can be critical for debugging.
This commit is contained in:
Philip Gough 2021-09-24 16:22:48 +01:00
parent 7f1092cdde
commit 56f96e6389
2 changed files with 15 additions and 3 deletions

View File

@ -127,9 +127,7 @@ function(params) {
action: 'drop', action: 'drop',
regex: '(' + std.join('|', regex: '(' + std.join('|',
[ [
'container_fs_.*', // add filesystem read/write data (nodes*disks*services*4)
'container_spec_.*', // everything related to cgroup specification and thus static data (nodes*services*5) 'container_spec_.*', // everything related to cgroup specification and thus static data (nodes*services*5)
'container_blkio_device_usage_total', // useful for containers, but not for system services (nodes*disks*services*operations*2)
'container_file_descriptors', // file descriptors limits and global numbers are exposed via (nodes*services) 'container_file_descriptors', // file descriptors limits and global numbers are exposed via (nodes*services)
'container_sockets', // used sockets in cgroup. Usually not important for system services (nodes*services) 'container_sockets', // used sockets in cgroup. Usually not important for system services (nodes*services)
'container_threads_max', // max number of threads in cgroup. Usually for system services it is not limited (nodes*services) 'container_threads_max', // max number of threads in cgroup. Usually for system services it is not limited (nodes*services)
@ -138,6 +136,15 @@ function(params) {
'container_last_seen', // not needed as system services are always running (nodes*services) 'container_last_seen', // not needed as system services are always running (nodes*services)
]) + ');;', ]) + ');;',
}, },
{
sourceLabels: ['__name__', 'container'],
action: 'drop',
regex: '(' + std.join('|',
[ // metrics are available at slice level
'container_fs_.*',
'container_blkio_device_usage_total',
]) + ');.+',
},
], ],
}, },
{ {

View File

@ -61,11 +61,16 @@ spec:
sourceLabels: sourceLabels:
- __name__ - __name__
- action: drop - action: drop
regex: (container_fs_.*|container_spec_.*|container_blkio_device_usage_total|container_file_descriptors|container_sockets|container_threads_max|container_threads|container_start_time_seconds|container_last_seen);; regex: (container_spec_.*|container_file_descriptors|container_sockets|container_threads_max|container_threads|container_start_time_seconds|container_last_seen);;
sourceLabels: sourceLabels:
- __name__ - __name__
- pod - pod
- namespace - namespace
- action: drop
regex: (container_fs_.*|container_blkio_device_usage_total);.+
sourceLabels:
- __name__
- container
path: /metrics/cadvisor path: /metrics/cadvisor
port: https-metrics port: https-metrics
relabelings: relabelings: