Bug #51868
closed
mgr/prometheus: Prometheus manager module metrics having the wrong type
Added by Patrick Seidensal almost 3 years ago.
Updated over 2 years ago.
Backport:
pacific, octopus
Description
Some of the metrics should be a `counter` instead of `gauge`. If the values of those metrics are being reset, Prometheus would just show them the way the are returned and not increasing the what has already been collected. Metrics of type counter should never decrease, just increase.
The affected metrics are:
- ceph_pool_rd
- ceph_pool_rd_bytes
- ceph_pool_rw
- ceph_pool_rw_bytes
- Subject changed from mgr/dashboard: Prometheus manager module metrics having the wrong type to mgr/prometheus: Prometheus manager module metrics having the wrong type
- Pull request ID set to 42506
- Status changed from In Progress to Fix Under Review
- Status changed from Fix Under Review to Pending Backport
- Copied to Backport #51979: pacific: mgr/prometheus: Prometheus manager module metrics having the wrong type added
- Copied to Backport #51980: octopus: mgr/prometheus: Prometheus manager module metrics having the wrong type added
- Copied to Backport #51981: nautilus: mgr/prometheus: Prometheus manager module metrics having the wrong type added
- Backport changed from pacific, octopus, nautilus to pacific, octopus
- Status changed from Pending Backport to Resolved
Also available in: Atom
PDF