starlingx/metrics-server-armada-app
Code Issues Proposed changes

Increase livenessProbe and readinessProbe timeout

Browse Source 
Performing stress test on mainline for 1 pcore support, it was
noticed that the metrics-server pod is restarting due to low
default timeout (1 second).

Increasing this value to 10 seconds will have no impact on app
performance, but will delay restart in case of probe failure.

Test Plan:
  PASS: build metrics-server packages
  PASS: install metrics-server updated and check the
        livenessProbe and readinessProve timeout.

Closes-bug: 2091006

Change-Id: I11915b24a7189bf1b733c11a037cfe766efcf748
Signed-off-by: Hediberto C Silva <hediberto.cavalcantedasilva@windriver.com>
This commit is contained in:
Hediberto C Silva 2024-12-04 09:58:58 -03:00
parent 8bf3798c1b
commit e3b7632f49
1 changed files with 8 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
stx-metrics-server-helm/stx-metrics-server-helm/fluxcd-manifests/metrics-server/metrics-server-static-overrides.yaml
View File

@ -20,6 +20,14 @@ tolerations:
nodeSelector:
node-role.kubernetes.io/control-plane: ""
livenessProbe:
periodSeconds: 20
timeoutSeconds: 10
readinessProbe:
periodSeconds: 20
timeoutSeconds: 10
resources: null
addonResizer: