Cortex is not getting deployed on in OKE Cluster(in Oracle Cloud Infra).

[kadursandy]

Cortex is not getting deployed on in OKE Cluster(in Oracle Cloud Infra). We use 3 worker nodes with ARM based architecture.

K8S Cluster Version "v1.22.5"
Helm Chart "1.3.0"
Cortex Version: v1.11.0

Pre-requisite: OKE cluster with 3 worker nodes (ARM architecture/processor).OCI CLI and Kubectl configured for cluster access.

To Reproduce
Steps to reproduce the behavior:

1. Helm Install - helm install cortex --namespace cortex -f cortex-values.yaml cortex-helm/cortex --version 1.3.0
2. Kubectl get all -n cortex

Expected behavior
All the Cortex micro services to be up and running without any issues.

Environment:

• Infrastructure: OKE(K8S v1.22.5)
• Deployment tool: Helm chart v1.3.0

Storage Engine

• Blocks

Additional Context
All the pods are gone to "crashloopbackoff" state.
Question:

1. Is Cortex architecture dependant, Does it support x86 arch only on worker nodes?
2. Does is support arm64 arch/processor on worker nodes?
3. if support for arm64 is not there, then when will the support come in .

Please let me know. OCI OKE clusters supports both arm64 and x86_64 based processors on worker(compute) nodes.

[alanprot]

What is the error?
What docker image are you using? I don't think cortex build arm images yet.

[kadursandy]

Image: quay.io/cortexproject/cortex:v1.11.0
pod status: CrashLoopBackOff
Error: There is no specific error in logs/describe/events, message Back-off restarting failed container is there.

[chillaxd]

Yes, by default there is no cortex arm docker image. But you can build your own.
You can check chiranjitdatta/cortex-arm64:v1.11.0 for cortex arm binary version of v1.11.0 which I have built.

[stale]

This issue has been automatically marked as stale because it has not had any activity in the past 60 days. It will be closed in 15 days if no further activity occurs. Thank you for your contributions.