Job gets stuck in "Waiting for Resource"
**UPDATE[2023-04-24]: Please follow this troubleshooting guide https://docs.gitlab.com/ee/ci/resource_groups/#jobs-get-stuck-in-waiting-for-resource **
NOTE: Please visit Make Resource Groups more reliable and flexible (&6774 - closed) for more detailed and actionable information. This issue contains various aspects including expected behavior and unexpected behavior, thus it's not likely that can be resolved by a single solution. We're planning to introduce UX and reliability improvements in the epic.
Summary
Intermittently multiple customers have reported jobs becoming stuck "waiting for resource" This can persist for many hours or even indefinitely.
Internal ZD Links:
340665
338834
340954
341546
341515
Steps to reproduce
- Use
resource_group - run pipeline until stuck
Example Project
What is the current bug behavior?
Job becomes stuck in "waiting for resource" job cannot be cancelled and pipeline is stuck even when resources are verifiably available.
What is the expected correct behavior?
job runs normally
Relevant logs and/or screenshots
Output of checks
This bug happens on GitLab.com
Possible fixes
After adding more information about resource group usage, we need to determine where the resource acquisition is getting stuck.