Crashes on Deploy - arm64 - Need architecture detection
Rojikku opened this issue · 2 comments
Describe the bug
Upon deployment, either manually or by helmchart, the pod crashes repeatedly.
I have diagnosed this to be because I am deploying to a raspberry Pi.
Normally deployments to things like dockerhub detect the architecture and pull the appropriate version of the container.
It appears your setup has a separate repository for the arm64 version (Though I definitely appreciate that it exists at all).
To Reproduce
helm repo add mittwald https://helm.mittwald.de
helm repo update
helm install replicator mittwald/kubernetes-replicator --namespace cert-manager
To Fix (Poorly, if you want this to work and it hasn't been patched yet)
helm pull mittwald/kubernetes-replicator
tar xvf kuber*
edit the values.yaml and append -arm64 at the end of the repository line (No spaces)
Expected behavior
Automatically detect architecture and use the appropriate docker container (Please?).
Environment:
- Kubernetes version: v1.21.5+k3s2
- kubernetes-replicator version: v2.6.2
- RPi4
Additional context
This might be a feature request depending on your point of view, but I spent like half an hour trying to figure it out, so I'll go with bug.
If there was a note in the readme.md about it I'd say it was a feature request.
No need to manually patch the Helm chart -- you can simply override the image.repository value when installing the Helm chart:
$ helm install --set image.repository=quay.io/mittwald/kubernetes-replicator-arm64 $release_name mittwald/kubernetes-replicator
Although I agree that a multi-architecture image (we're already providing one for mittwald/kube-httpcache) would provide a more seamless experience for non-x86 users. 👍
There has not been any activity to this issue in the last 14 days. It will automatically be closed after 7 more days. Remove the stale label to prevent this.