-
Notifications
You must be signed in to change notification settings - Fork 171
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
zarf init fails on multi-node minikube cluster #1998
Comments
This go around of tearing down minikube and bringing it up, it did successfully deploy the registry pod, but the rest of the zarf deploy fails with this error:
|
Following: https://minikube.sigs.k8s.io/docs/tutorials/volume_snapshots_and_csi/ w/ a fix as noted in kubernetes/minikube#12360 This worked for me: minikube start --kubernetes-version 1.24.7 --nodes 3
minikube addons enable volumesnapshots
minikube addons enable csi-hostpath-driver
kubectl patch storageclass csi-hostpath-sc -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'
zarf init --storage-class=csi-hostpath-sc --log-level=trace --confirm Let me know if this works for you! |
Thanks! We went in a different direction where we don't need a cluster locally anymore, but if we do, I'll try it out |
Going to close this issue for now - Zarf does require a storage class for its registry by default (though you can turn this off and back it by something like S3 if you prefer) - also worth noting that we are working on a zarf-init-longhorn package that should be ready soon that will include a storage class of its own. |
We can reopen if we need to @vanakema |
In zarf-dev#2124, an emptydir was mounted when `persistence.enabled: false` is set. However, this is not enough to make registry work: it crashloops as in issue zarf-dev#1998. This is fixed by setting the env variable specifying storage config even if persistence is disabled.
In zarf-dev#2124, an emptydir was mounted when `persistence.enabled: false` is set. However, this is not enough to make registry work: it crashloops as in issue zarf-dev#1998. This is fixed by setting the env variable specifying storage config even if persistence is disabled.
In zarf-dev#2124, an emptydir was mounted when `persistence.enabled: false` is set. However, this is not enough to make registry work: it crashloops as in issue zarf-dev#1998. This is fixed by setting the env variable specifying storage config even if persistence is disabled. Signed-off-by: Adam Michalik <[email protected]>
Environment
Device and OS: MacBook Pro M2 Pro, MacOS
App version: v0.29.1
Kubernetes distro being used: MiniKube (Docker driver) v1.31.1, k8s v1.24.7
Other:
Steps to reproduce
minikube start --kubernetes-version 1.24.7 --nodes 3
zarf init --confirm
Expected result
It works
Actual Result
The registry fails to deploy
Pod Logs (zarf-docker-registry):
Visual Proof (screenshots, videos, text, etc)
It just never deploys, and the Helm deploy just keeps trying till it times out, spitting out essentially the same error as above.
Severity/Priority
High: needed for our developer environment for testing HA application deployments. Not affecting prod (yet)
Additional Information
We tried this as well, which resulted in 1 registry pod deploying, but the 2nd pod had the same error as listed above
zarf init --storage-class=standard --set REGISTRY_PVC_ENABLED=false --confirm
The text was updated successfully, but these errors were encountered: