Bitbucket pipeline steps being queued indefinitely with runners
プラットフォームについて: Cloud のみ - この記事は、 クラウド プラットフォームのアトラシアン製品にのみ適用されます。
要約
Bitbucket Pipelines builds that run on self-hosted runners can sometimes remain in a queued state indefinitely. This issue can occur even when there are other runners available to accept new connections.
環境
Bitbucket Cloud Pipelines
原因
This issue may arise due to a communication failure between the Runners and the backend of Bitbucket Pipelines. In some rare instances, the Runner will fail to reach Bitbucket to confirm the current Runner status, causing a desync between the two.
回避策
Delete the existing runner.
- Go to the repository preferences
- Navigate to Runners to access the runner settings page.
- Select the runner that needs to be deleted and click on the three dots.
- Choose Delete.
- Create a new runner
ソリューション
If prefer to maintain your existing runner, raise a ticket with Bitbucket Cloud Support and provide us with the "RUNNER_UUID" or a HAR file when loading the runner settings page.
Our engineering team can then update the backend for this specific runner to make it available again.
How to get runner UUID
- Identify the runner that is available but not being assigned any pipeline steps.
From the "docker container run" command which pipelines provides when you create your runner there would be a field showing something as below :
RUNNER_UUID={9221374-acbd-574d-be9b-7c112gt526c1b}