Remove Deployment Trigger Type: ImageChange

bcgov / digital-journeys

PSA Forms System

https://bcgov.github.io/digital-journeys/

Apache License 2.0

8 stars 7 forks source link

Remove Deployment Trigger Type: ImageChange #390

Closed warrenchristian1telus closed 2 years ago

warrenchristian1telus commented 2 years ago

Please remove the Type: ImageChange from the DeploymentConfigs, so that changes deployed to one namespace (/tools) do not trigger updates in the others.

This is to temporarily resolve the deployment issues while we work toward the permanent tagged image fix in #282.

Note that the deployed changes will still be susceptible to cluster-wide upgrades and pod cycling, as they will still share base images in tools namespace until #282 is implemented.

kkapoor-fresh commented 2 years ago

Pull request https://github.com/bcgov/digital-journeys/pull/391 created for this fix.

kkapoor-fresh commented 2 years ago

The fix has been deployed to DEV, TEST and PROD. @warrenchristian1telus I saw a few workflow actions running after this fix, so I am assuming this looks good but please confirm at your end once. @freshjwarren

kkapoor-fresh commented 2 years ago

@warrenchristian1telus @Abuchana I guess this ticket is good to close.

Abuchana commented 2 years ago

@kkapoor-fresh @warrenchristian1telus Thank you, Kunal! Warren, I'll let you official close it...

warrenchristian1telus commented 2 years ago

Hi @kkapoor-fresh I found an ImageChange trigger on the first DC I checked in prod. Can you please review the deployments and make sure they have all been removed?

kkapoor-fresh commented 2 years ago

@warrenchristian1telus The image triggers have been removed from the web, api and camunda services as these are the only services actively build and deployed by the pipeline currently. These automatic deployments are running fine after removing the image triggers as the refresh of environments done via code updates is not refreshing all the environments but only one which is relevant. To tackle deployment refresh issue during cluster upgrades, #282 should be good enough.