search

openshift-pipelines / pipeline-service

SaaS for Tekton Pipelines
Apache License 2.0
23 stars 44 forks source link

Add Deadlocked ResolutionRequests panel and update Pipelines controller restart panel #1037

Closed divyansh42 closed 1 month ago

divyansh42 commented 1 month ago

Add Deadlocked ResolutionRequests panel same as App SRE dashboard and make changes in Pipelines controller restart panel as per App SRE dashboard

divyansh42 commented 1 month ago

image @gabemontero This is how the present panel looks like

gabemontero commented 1 month ago

looks good @divyansh42

and I think none of the rows in the json after your insertion changed because everything fit on one row

So at the moment I believe this would be "mergeable"

I'll check out your PR locally and confirm it looks the same for me as well once I have a cluster up.

divyansh42 commented 1 month ago

and I think none of the rows in the json after your insertion changed because everything fit on one row

@gabemontero Yes, that was the reason

gabemontero commented 1 month ago

I see the same panel you showed below @divyansh42 and the panels below look OK when I expand the sections.

Per our chat in scrum let's move onto the resolution request controller restart panel. I'll add some comments to your education Jira to reiterate what we discussed.

thanks

divyansh42 commented 1 month ago

@gabemontero this is the modified dashboard. Please take a look. I have added the panel in the second row. Please let me know if that doesn't look good.

image
divyansh42 commented 1 month ago

I see one of the tests is failing. However, I am not able to get to the logs. It seems I don't have access.

gabemontero commented 1 month ago

I see one of the tests is failing. However, I am not able to get to the logs. It seems I don't have access.

yeah getting logs for us requires a backdoor of logging into the Konflux prod-rh01 cluster; I have not bothered getting into that with you because this repo is getting deprecated, and soon you won't need to worry about these tests.

also, the "upgrade" test is known to be notoriously flakey.

The deployment-ocp-414 test is a good indication things are OK.

I'm going to merge this, and then let you bring up a cluster, and run dev_setup.sh without the --use-current-branch so that it will run against main branch.

If the panels still look as you expect, you'll go onto the new "update infra-deployments" process I've alluded to previously that will start discussing in detail in today's stand up / scrum