fix(examples): remove dead code comment #12194
Conversation
|
This stack of pull requests is managed by Graphite. Learn more about stacking. |
| @@ -242,8 +242,7 @@ resource "kubernetes_deployment" "main" { | |||
| } | |||
|
|
|||
| spec { | |||
| # replicas = data.coder_workspace.me.start_count | |||
| replicas = 1 | |||
| replicas = data.coder_workspace.me.start_count | |||
| selector { | |||
| match_labels = { | |||
| "app.kubernetes.io/name" = "coder-workspace" | |||
There was a problem hiding this comment.
Also, FWIW, I was using this on a GKE autopilot cluster which doesn't support overprovisioning and my pods only ever got `250m` CPU. I think this should probably be configurable, rather than the only option.
See this comment inline on Graphite.
There was a problem hiding this comment.
If we are using the count on deployment and the deployment gets scaled to 0. Does the count value on the pod matter?
There was a problem hiding this comment.
Oh I see, I'm dumb. Though thinking about it now, I wonder if it would be better to just scale up and down the replicas rather than delete and recreate the deployment each time. I'll just remove the dead code comment for now.
There was a problem hiding this comment.
My issue still stands about the overprovisioning issue, though. Would like your thoughts
| @@ -242,8 +242,7 @@ resource "kubernetes_deployment" "main" { | |||
| } | |||
|
|
|||
| spec { | |||
| # replicas = data.coder_workspace.me.start_count | |||
| replicas = 1 | |||
| replicas = data.coder_workspace.me.start_count | |||
| selector { | |||
| match_labels = { | |||
| "app.kubernetes.io/name" = "coder-workspace" | |||
There was a problem hiding this comment.
If we are using the count on deployment and the deployment gets scaled to 0. Does the count value on the pod matter?
5526932 to
78e1c15
Compare
No description provided.