Prometheus triggering kernel panic
Summary
As reported by the customer:
One of our GitLab rails hosts crashed and rebooted. It appears that prometheus caused a kernel panic. We don't currently have prometheus configured in our setup. Do you have any idea why this would happen?
Steps to reproduce
This has only occurred once and we are unable to reproduce the issue.
What is the current bug behavior?
- Kernel panic triggered
- Server restarts
- Comes back online working again
What is the expected correct behavior?
Kernel calls should not cause a kernel panic
Relevant logs and/or screenshots
For privacy reasons please see internal Zendesk support issue for log output: https://gitlab.zendesk.com/agent/tickets/86291
Output of checks
Results of GitLab environment info
Customer is on 10.0.4. Have requested this information.
Results of GitLab application Check
Possible fixes
This is the only kernel call that I can find that is made by Prometheus: https://gitlab.com/gitlab-org/gitlab-ce/blob/master/app/models/project_services/prometheus_service.rb#L82
I think we should review what is happening here.