unable to upgrade calimari on a an apple arm 64 worker

[Rose-Northey]

Severity

Blocks any customer running a Kubernetes Agent as a Worker on an ARM64 cluster

Version

Introduced by commit 07fa8c5cfb0. Present after 2026.2.9720

Latest Version

I could reproduce the problem in the latest build

What happened?

When the k8 node is created on arm64, its pods can only be aRM64. The upgrade is hard coded to be on a aMD64 pod. This means the upgrade cannot be completed because it is trying to complete the task on a pod which cannoy exist.
Expected: The upgrade should be successful on arm-64 architecture as well as amd-64.

Reproduction

Reproduction for users on an apple silicon mac:

1. run docker desktop
2. run octopus server
3. Start minikube
4. in Octopus UI, go to Infrastructure → Worker Pools → Add Worker → add Kubernetes Agent
   Note: when running the helm command, replace localhost with host.docker.internal in both the serverUrl and serverCommsAddresses values before running in the terminal.
5. Edit the worker details → Upgrade the version of calimari
6. The error will appear in the task

Error and Stacktrace

[POD EVENT] FailedScheduling | 0/1 nodes are available: 1 node(s) didn't match Pod's node affinity/selector. preemption: 0/1 nodes are available: 1 Preemption is not helpful for scheduling. (Count: 1)

More Information

Workaround

No response