feat(workflow-controller): make log level configurable#475
feat(workflow-controller): make log level configurable#475pcandido wants to merge 5 commits intolitmuschaos:masterfrom
Conversation
…s.loglevel Signed-off-by: Paulo Cândido <paulo.candido@ifood.com.br>
…gs.gloglevel Signed-off-by: Paulo Cândido <paulo.candido@ifood.com.br>
|
Hello, great job. You don't need to edit the README.md since this one is generated by the CI. Test 1: Default Values (Backward Compatibility)Command: helm template charts/litmus-agent charts/litmus-agentExpected: - --loglevel
- debug
- --gloglevel
- "1"Result: - --loglevel
- debug
- --gloglevel
- "1"✅ PASS - Default values render correctly Test 2: Custom Values (info, "0")Command: helm template charts/litmus-agent charts/litmus-agent \
--set workflow-controller.appSettings.loglevel=info \
--set workflow-controller.appSettings.gloglevel="0"Expected: - --loglevel
- info
- --gloglevel
- "0"Result: - --loglevel
- info
- --gloglevel
- "0"✅ PASS - Custom values override correctly Test 3: Different Values (error, "2")Command: helm template charts/litmus-agent charts/litmus-agent \
--set workflow-controller.appSettings.loglevel=error \
--set workflow-controller.appSettings.gloglevel="2"Expected: - --loglevel
- error
- --gloglevel
- "2"Result: - --loglevel
- error
- --gloglevel
- "2"✅ PASS - Different values work correctly SummaryAll helm template tests pass successfully:
The PR changes are working as expected! |
Signed-off-by: Paulo Cândido <paulo.candido@ifood.com.br>
|
Thanks for the tip @Vr00mm. The CI actually asked me to run helm-docs locally from the litmus-agent root, so I had to add an extra commit. |
Signed-off-by: Paulo Candido <pcandido.m@gmail.com>
|
Hi everyone, Just a gentle bump on this PR since it has been open for a while. Could someone please take a look when you have a chance? Thanks! |
Signed-off-by: Paulo Candido <pcandido.m@gmail.com>
|
Hi maintainers! 👋 I apologize for the recent force push on this PR. When I tried to resolve the merge conflicts using GitHub's web interface, I made a mistake during the conflict resolution. To fix this properly, I reverted the changes (via force push) and redid the merge carefully to ensure everything is correct now. The PR is now updated and ready for review. Sorry for any inconvenience this may have caused, and I appreciate your patience! 🙏 Looking forward to your feedback when you have a chance to review. Thanks! |
What this PR does / why we need it:
Adds the ability to override the workflow-controller log level and glog verbosity via Helm values. Today
--loglevelis hardcoded todebugand--gloglevelto"1"in the Deployment, so operators cannot reduce log volume without forking or post-rendering.In multi-cluster setups (e.g. 20+ clusters), the litmus-agent-workflow-controller at debug generates a large amount of logs (e.g. 30k+ records per 15 minutes), increasing ingestion and storage cost. This change adds:
appSettings.loglevel(default:debug, for compatibility) — controls the application log level (debug, info, warn, error). Installers can set e.g.workflow-controller.appSettings.loglevel: "info"to cut debug noise.appSettings.gloglevel(default:"1", for compatibility) — controls glog verbosity (numeric; higher = more verbose). Installers can set e.g.workflow-controller.appSettings.gloglevel: "0"to reduce glog output.Both keep existing behavior by default so the chart remains backward compatible.
Which issue this PR fixes
fixes #474
Special notes for your reviewer:
loglevel: debugandgloglevel: "1"for backward compatibility.litmus-agent/Chart.yamldependency on workflow-controller updated to 0.2.3.Checklist
@Vr00mm @ispeakc0de @Jonsy13