-
Notifications
You must be signed in to change notification settings - Fork 953
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
Forbidden Runner version v2.317.0 is deprecated and cannot receive messages #3381
Comments
hi @jonapich Do you have your runner's session id? Also, it seems like those logs you posted are from yesterday? |
@luketomlinson it did happen yesterday; sorry for the mix up. Good catch 😅 I think |
Full log line:
|
Thanks @jonapich. This was rolled back yesterday. I'm going to close this one out. |
@luketomlinson can we have more details about this? We faced the same issue with v2.316.1 and upgraded to v2.317.0 to resolve this problem. What was rolled back? |
👋 @dtakis Currently, if you register your runner with the The bug that was rolled back was incorrectly performing that age check on the latest version. If you register a runner without the |
@luketomlinson We are currently seeing this error again, here's an example session:
Since v2.318.0 was just released one hour ago, perhaps this has broken the logic? |
Taking a look @jacobwgillespie |
Getting this aswell on all our runners using 2.317.0. |
Also getting this on many repos, but not all. We have alarms firing and production deployments that are not happening. We have to run with updates disabled (because we run ephemeral runners in Kubernetes which start, run one job then exit) but we check for newer runner software every day and rebuild our images if a new base image is found. However given that the only newer version is 2 hours old, there is no newer container base image for us to build from. Surely blocking the N-1 release when N is only 2 hours old is an error on Github's part? |
👋 Apologies for this error on our part. It should be rolled back now. Restarting runners should get it back up and rolling again. |
@luketomlinson can confirm all runners up again thanks for sorting. |
Can confirm ours are working as well, thank you for the fast turnaround ❤️ |
@luketomlinson This is happening again with the latest v2.319.1. Did try restaring the runner few times but it stops with-in seconds and throws this message. |
Hi @malwinders-evertz, do you have any runner logs, specifically the |
Yes,
|
@malwinders-evertz Do you happen anything in your infrastructure (proxies/firewalls or what not) that might be throwing a 403 in between the runner and GitHub? I don't see that request make it to our servers, but I'm still digging. It's possible a firewall might be rejecting |
@luketomlinson yes, we do have a VPN setup in our infrastructure, This exact same issue occured earlier as reported, around August 13. Later on August 14 there was a github outage for almost all the services and when that outage got resolved our runner started working as expected. |
Thanks @malwinders-evertz, can you confirm that |
Yes, i will verify that, first thing in the morning as i have to reach out to our IT team regarding this and right now no one from IT is availabe. Will get back to you in the morning. |
@luketomlinson I confirmed with IT, the above address was not allowed over our VPN, once they allowed it and we restarted the runner service things seems working as expected. Thanks for the quick Response. Still wondering why it was throwing |
Thanks @malwinders-evertz. I think that's an error reporting bug on the runner side. It's interpreting any 403 as a deprecated version error, which is not correct. I'm looking into fixing that. |
Count me in as a datapoint. Running |
Had the same issue here. Sep 18 15:34:29 vmss-xxx-xx-xx-eastus000001 runsvc.sh[358641]: Started running service I updated '2.319.0' ...... still same issue |
I'm using Helm's latest version 0.9.3 and get this issue. Here are some logs, sigh.
|
Snippet from the Support Policy notes the following: We support (at maximum) 2 GA images and 1 beta image at a time. We begin the deprecation process of the oldest image label once the newest OS image label has been released to GA So the documentation here https://docs.github.com/en/enterprise-cloud@latest/actions/hosting-your-own-runners/managing-self-hosted-runners/autoscaling-with-self-hosted-runners#controlling-runner-software-updates-on-self-hosted-runners seems conflicting this doc https://github.com/actions/runner-images/?tab=readme-ov-file#software-and-image-support? Does these docs need to be in-sync? |
Got it. Turns out we were using a custom runner based upon the runner image. So we just regenerated the runner image and we were good. I initially missed the minor version being off. |
We encountered the same issue sine 2 days ago, the error log is:
Just now we upgaded the runner version from v2.317.0 to v2.319.1, all errors disappeared. |
This happened to us on 2.317.0, which is the latest version at time of writing:
It happened twice. Timestamps in UTC.
past occurence: #3366 (comment)
The text was updated successfully, but these errors were encountered: