Incidentsby GlyphElevated Linux worker queue times
Sep 2, 21:44 UTCSep 2, 22:05 UTC
Duration
21m
Impact
Minor
Root cause
Capacity
Expo, 90 days
24 incidents
Affected
EAS BuildEAS WorkflowsExpo Application Services - EAS BuildExpo Application Services - EAS Workflows
Final update
A fix was deployed at 2:50pm PT/5:50pm ET. The job queue has been running as expected. The root cause of this incident is that a new version of a cache needed more vCPUs than it was given, and we have increased the CPU count as the immediate fix along with additional monitoring for CPU usage on these cache instances.
Timeline
- Resolved · Sep 2, 22:05 UTC
A fix was deployed at 2:50pm PT/5:50pm ET. The job queue has been running as expected. The root cause of this incident is that a new version of a cache needed more vCPUs than it was given, and we have increased the CPU count as the immediate fix along with additional monitoring for CPU usage on these cache instances.
- Identified · Sep 2, 21:52 UTC
We have determined that some of our package caches are under high load, and this is causing elevated build times and in turn causing higher Linux worker queue times. We are currently working on a fix. In-progress builds are continuing to complete and enqueued builds are continuing to be processed.
- Investigating · Sep 2, 21:44 UTC
Queue times are elevated for Linux workers for EAS Build and Workflows. Android build jobs may be delayed. We are currently investigating the issue.
More from Expo
Full historyProject dashboards fail to loadlasted 11mElevated submission error rateslasted 52mElevated iOS submission error rateslasted 2h 37mElevated API server error rates affecting submissions and buildslasted 1h 7mWebsite and services partially unavailable due to high database loadlasted 2h 13mWebsite and services partially unavailable due to high database loadlasted 0m