magic-nix-cache-action

magic-nix-cache-action copied to clipboard

I am using the magic-nix-cache-action project in https://github.com/ereslibre/homelab. Thanks for this project!

During the post job cleanup, the output states:

...
2023-08-15T18:59:06.1176097Z ##[error]The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
...

And the job is considered failed. An example of a failed run in the mentioned repo can be found here.

Do you have any idea on how to workaround this? I think the post job cleanup might have taken longer than what GitHub allows for post jobs and at some point it's terminated.

These are probably related to https://github.com/actions/runner-images/issues/6680 and https://github.com/actions/runner-images/discussions/7188 . Underlying the cryptic error message is probably one similar to https://github.com/actions/runner-images/issues/6680#issuecomment-1529147603 .

Since this may be due to an out of memory condition, modulating the swap size as recommended in https://github.com/actions/runner-images/discussions/7188#discussioncomment-6750749 or, for example using easimon/maximize-build-space,

      - name: maximize build space
        uses: easimon/maximize-build-space@fc881a613ad2a34aca9c9624518214ebc21dfc0c # ratchet:easimon/maximize-build-space@v10
        with:
          root-reserve-mb: 38912
          swap-size-mb: 4096
          remove-dotnet: "true"
          remove-android: "true"

may help to reduce or eliminate such errors. Obviously the relevant value of swap-size-mb would vary depending on the details of your workload.