Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Custom Image Registry #354

Open
geoff-carr-bzy opened this issue Jul 6, 2021 · 3 comments
Open

Custom Image Registry #354

geoff-carr-bzy opened this issue Jul 6, 2021 · 3 comments

Comments

@geoff-carr-bzy
Copy link

geoff-carr-bzy commented Jul 6, 2021

How can the default pause container image used for the portworx-api and portworx-kvdb pods be overridden? I have specified a custom image registry in the install spec which works for just about everything else but for the pause container it appends the original registry and image path as follows: -

Back-off pulling image "docker.repo/k8s.gcr.io/pause:3.1"

I need it to be: -

docker.repo/pause:3.1

I have tried to edit and update the portworx-api DaemonSet but it then reverts back to the original so I am guess the operator is resetting it.

Found this which suggests adding // to the custom registry path which fixes the pause container pull but then breaks everything else: -

#319

@geoff-carr-bzy
Copy link
Author

geoff-carr-bzy commented Jul 6, 2021

Same issue for the stork-scheduler pods: -

Back-off pulling image "docker.repo/gcr.io/google_containers/kube-scheduler-amd64:v1.18.6"

however the stork pods work: -

Containers:
  stork:
    Container ID:  docker://c13d2b36af8fa46a5fb6e6fd3272610590bab374862ddead9da7d4ac3cea268e
    Image:         docker.repo/openstorage/stork:2.4.4
    Image ID:      docker-pullable://openstorage/stork@sha256:76d88da63e48198e91fe424b9956d6376a61a620d31f0045ab454e1d03de5a66

@benhwebster
Copy link

Also having this issue, we're looking to migrate from the Pure Storage Orchestrator CSI driver and this is blocking us from deploying this operator, @piyush-nimbalkar apologies for the ping, this issue has been open for a while.

@piyush-nimbalkar
Copy link
Contributor

Sorry guys for the late response. This just slipped my radar.
Due to an old bug which was not considering gcr.io and k8s.gcr.io as image repositories, we had this misconstructed path. To avoid breaking existing deployments, we came up with an annotation to fix these repositories.

You can add operator.libopenstorage.org/common-image-registries: "gcr.io,k8s.gcr.io" annotation to the StorageCluster object. The operator will then consider them as normal registries like docker.io or quay.io and should construct the correct path for the above images that are causing issues.

apiVersion: core.libopenstorage.org/v1
kind: StorageCluster
metadata:
  name: px-cluster
  namespace: kube-system
  annotations:
    operator.libopenstorage.org/common-image-registries: "gcr.io,k8s.gcr.io"
spec:
  ...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants