-
Notifications
You must be signed in to change notification settings - Fork 993
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
Error messages when removing a self-hosted runner #971
Comments
@wyphan did you click the You should either click |
No, I didn't click "Force remove this runner". |
@wyphan do you mind sharing a link to the repository or organization that you have this runner configured? and also the runner's name if you still remember? |
The two instances of the error message were for two different repositories:
|
@wyphan @TingluoHuang I'm also experiencing the same error message. |
Does anyone have the runner diag log available for me to check? |
I think I know what happened. When the interactive runner auto-upgrade to a newer version, it got partially detached from the terminal. So, after the upgrade, the runner is still running in the background with If you run |
There are two issues to fix with this:
|
Note that this issue is benign and the runner was still removed. |
@TingluoHuang That is correct. When I was still using them, usually I SSH into the machine, start GNU screen, then start the runner interactively, and detach from the GNU screen session. Edit: typo |
I've started getting this error when an ephemeral runner on Windows finishes. |
Also of note is that it reports that the job result was Canceled, but the job was not canceled.
The ephemeral runner is removed from the org, but the |
Facing the same issue via clicking the Force removal of self-runner |
Is there any plan to fix this issue? We're facing several of these errors daily as we're relying on ephemeral self-hosted runners.. |
Any update on this issue? I'm facing this issue daily with enterprise-level ephemeral self-hosted runners (containerized)... It is blocking us to implement proper runner autoscaling |
I am seeing runners not picking up jobs, staying idle, then exiting with this
|
I am experiencing same issue with actions-runner-controller on AWS EKS after trying force remove the runner. All of my runner pods are keep created and terminated itself within 2 minutes. Any updates on this or workaroud to avoid terminating the runner? |
we are seeing this issue too |
Any update on this? |
+1 |
This is still relevant.
|
Why is this not fixable for three years now? is it possible that you only kill the run.sh process which does not affect the other two? when i check the processes, i can see 3 runner processes.
Try 1 - using ./config.sh remove only
Try 2 killing two processes before ./config.sh remove
Try 3 killing all three processes before ./config.sh remove - works
|
* When running many GitHub Runners, I sometimes see them exit uncleanly because GitHub's config.sh races with shutting down the Listener which can causes Pods to error * Help GitHub along by killing the Runner.Listener which might reduce the frequency here, though not a real solution ``` An error occurred: Access denied. System:ServiceIdentity;DDDDDDDD-DDDD-DDDD-DDDD-DDDDDDDDDDDD needs View permissions to perform the action. ``` Rel: actions/runner#971
* When running many GitHub Runners, I sometimes see them exit uncleanly because GitHub's config.sh races with shutting down the Listener which can causes Pods to error * Help GitHub along by killing the Runner.Listener which might reduce the frequency here, though not a real solution ``` An error occurred: Access denied. System:ServiceIdentity;DDDDDDDD-DDDD-DDDD-DDDD-DDDDDDDDDDDD needs View permissions to perform the action. ``` Rel: actions/runner#971
* When running many GitHub Runners, I sometimes see them exit uncleanly because GitHub's config.sh races with shutting down the Listener which can causes Pods to error * Help GitHub along by killing the Runner.Listener which might reduce the frequency here, though not a real solution ``` An error occurred: Access denied. System:ServiceIdentity;DDDDDDDD-DDDD-DDDD-DDDD-DDDDDDDDDDDD needs View permissions to perform the action. ``` Rel: actions/runner#971
I'm still seeing this error, in this case when an idle runner is terminated because the Node it is on is being deleted as part of autoscaling (down) the Kubernetes cluster: Logs look approximately like this (ANSI color codes, timestamps, and some other stuff removed)
|
Describe the bug
When removing a self-hosted runner, I get the following error messages:
The removal seems to succeed though, as refreshing the page removes the self-hosted runner from the list.
To Reproduce
Steps to reproduce the behavior:
./config.sh remove --token XXX
Expected behavior
A clear and concise description of what you expected to happen.
Runner Version and Platform
Version of your runner?
Sorry I forgot to check, but in both cases they were downloaded from the official download links as given out in "Add runner"
OS of the machine running the runner? Linux
x86_64
. This has happened twice: with Ubuntu Linux 20.04 LTS and CentOS 8.The text was updated successfully, but these errors were encountered: