Runner Docker image 2.235.0 fails: "v2.325.0 is deprecated" — bug or deprecation?

[sabrinamorales-hotmart]

Describe the bug
We were running a self-hosted runner using the official Docker image 2.235.0 (pinned in our runner Dockerfile). It was working fine until it suddenly stopped receiving messages and fails with an access/deprecation error. The latest available image we see is 2.238.0, but the pinned 2.235.0 stopped working without any prior warning.

To Reproduce
Steps to reproduce the behavior:

1. Build / run a runner based on the official Docker image tagged 2.235.0 (example Dockerfile line: FROM ghcr.io/actions/runner:2.235.0).
2. Register the runner and make it available to receive jobs.
3. Observe the runner logs — the listener fails with an AccessDenied / deprecation error.

Expected behavior
A runner image pinned to 2.235.0 should continue to function (or there should be a clear, prior deprecation/retirement notice and migration instructions). If images are being force-deprecated, a documented deprecation schedule or clear communication would be expected so we can plan upgrades.

Runner Version and Platform

2.235.0

OS of the machine running the runner? OSX/Windows/Linux/...

• Docker container: Linux (built from official runner Dockerfile, multi-arch build with platforms linux/amd64 and linux/arm64)

What's not working?

The runner stops receiving messages and the listener logs the following error:

[RUNNER 2025-09-05 23:14:16Z ERR  Listener] GitHub.DistributedTask.WebApi.AccessDeniedException: Runner version v2.325.0 is deprecated and cannot receive messages.

Job Log Output

If applicable, include the relevant part of the job / step log output here. All sensitive information should already be masked out, but please double-check before pasting here.

Relevant log snippet we observed:

[RUNNER 2025-09-05 23:14:16Z ERR  Listener] GitHub.DistributedTask.WebApi.AccessDeniedException: Runner version v2.325.0 is deprecated and cannot receive messages.

Runner and Worker's Diagnostic Logs

If applicable, add relevant diagnostic log information. Logs are located in the runner's _diag folder. The runner logs are prefixed with Runner_ and the worker logs are prefixed with Worker_. Each job run correlates to a worker log. All sensitive information should already be masked out, but please double-check before pasting here.

Questions / clarification requested from maintainers

• Was the 2.235.0 image (or the runner version referenced in the error) intentionally deprecated/retired by the project? If so, is there a deprecation/retirement calendar or policy (how long images remain supported after release)?
• Is this behavior expected (i.e., forced deprecation) or is this a bug / accidental configuration on the server-side that caused older tags to be refused?
• Is there a recommended mitigation or workaround for pinned images (e.g., must we always track the latest tag, is there a support window, or must we rebuild from source)?
  Thanks — this broke our production CI unexpectedly, so any guidance or confirmation would be appreciated.

[pkking]

Same issue here, some of the ARC autoscaling listener report error "An error occured: Runner version v2.325.0 is deprecated and cannot receive messages."
But some are running normally :(

[namhyuck-kim-otsk]

Any update on this?

[AlbertoOS]

I don't think there will be any updates from the team, the best course of action would be to rebuild images using the latest available version (v2.328.0 - https://github.com/actions/runner/releases) and later setup some form of auto-update or weekly build/check to avoid discovering this way.

I've also found this issue so it has happened before and will probably happen in the same way again.