Failure signature
The M1 - Mac Ventura (13) macOS tests job stalls in the pre-test Show Bot Info task on VSM-XAM-70.Ventura.arm64 until Azure DevOps cancels the job:
The job running on agent VSM-XAM-70.Ventura.arm64 ran longer than the maximum time of 120 minutes.
The Operation will be canceled.
All five occurrences use job e59aa738-7d46-5a5e-8f43-c6b31f5d2976 and task fdc1737c-f64b-5ec2-e381-d3e861748b97, both at attempt 1. No tests start.
Confirmed occurrences
Total: 5 occurrences across 2 distinct monitored PRs and 5 distinct heads. First observed at 2026-08-18T04:42:57Z; latest observed at 2026-08-23T04:41:17Z.
Classification and impact
CI infrastructure failure. The same pre-test task, worker, record IDs, and hard timeout recur on unrelated dependency-update PRs. The job never reaches build or test execution, so this is not evidence of a source or deterministic test regression. It blocks the complete Ventura stage and the overall simulator check.
Investigation performed
- Inspected authenticated Azure DevOps build metadata, trigger fields, timelines, task logs, attempts, and retained run history.
- Verified each occurrence was an automatic initial PR run on a distinct head, with attempt 1 and no previous attempts.
- Correlated 5 exact-signature occurrences among 15 retained simulator runs for these PRs.
- Searched open and closed
dotnet/macios issues separately by exact timeout text, task name, worker, and Ventura component terms. Existing timeout issues concern test execution or provisioning and do not match this pre-test task stall.
The lower-level reason Show Bot Info stops making progress remains unresolved. Azure task logs are retention-limited.
Workaround
Rerun failed jobs after refreshing the PR head, merge-conflict state, retry budgets, and equivalent active attempts. This may unblock the PR but does not fix the affected worker/task.
Failure signature
The M1 - Mac Ventura (13) macOS tests job stalls in the pre-test
Show Bot Infotask onVSM-XAM-70.Ventura.arm64until Azure DevOps cancels the job:All five occurrences use job
e59aa738-7d46-5a5e-8f43-c6b31f5d2976and taskfdc1737c-f64b-5ec2-e381-d3e861748b97, both at attempt 1. No tests start.Confirmed occurrences
f0243b8d39cefffaf2c857b7bf0713331c111b0020260820.1)3fb3386245d841672f793bead4bf8c56f3d7014920260819.2)67e16a518ef18a296c6c9040021675b35ead479e20260823.3)7e32119e9ac03e47fbdb101036d93d683ff67a5720260821.6)e28113097531b53792a97564a04a949b74658aa620260818.3)Total: 5 occurrences across 2 distinct monitored PRs and 5 distinct heads. First observed at
2026-08-18T04:42:57Z; latest observed at2026-08-23T04:41:17Z.Classification and impact
CI infrastructure failure. The same pre-test task, worker, record IDs, and hard timeout recur on unrelated dependency-update PRs. The job never reaches build or test execution, so this is not evidence of a source or deterministic test regression. It blocks the complete Ventura stage and the overall simulator check.
Investigation performed
dotnet/maciosissues separately by exact timeout text, task name, worker, and Ventura component terms. Existing timeout issues concern test execution or provisioning and do not match this pre-test task stall.The lower-level reason
Show Bot Infostops making progress remains unresolved. Azure task logs are retention-limited.Workaround
Rerun failed jobs after refreshing the PR head, merge-conflict state, retry budgets, and equivalent active attempts. This may unblock the PR but does not fix the affected worker/task.