You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Gate retries: rerun after main changes and treat dependency-download failures as retryable before escalating to ready-for-human #397
(a) Re-queue after a fix lands. When main advances past a fix for a known-failing test, re-queue parked items whose only failure was that test.
(b) Infra failures don't consume the cap. Classify network and download failures (cargo registry, HTTP/2 framing errors, and similar) as infra, and retry them without consuming the retry cap.
(c) Crashed reviews aren't feedback. Don't count a crashed reviewer run (review_exit non-zero with no findings) as unresolved feedback. Retry the review instead.
(d) Mechanical fixups before escalating. Before escalating to ready-for-human, run mechanical fixups automatically: cargo xtask manifest (or the repo's equivalent) and adding a sign-off to agent-authored merge commits.
Acceptance
An item reaches ready-for-human only when the remaining failure is attributable to the PR's code or needs a real decision.
The escalation comment says which failure class consumed each retry.
Related: #383 (control-band alarms on gate/flaky failures and ready-for-human rate), #395 (quarantine PRs for cross-PR flaky tests).
Triage: This is a process/behavior change to the gate's retry and escalation logic containing four loosely-coupled proposals (re-queue after a fix lands, classify infra failures as non-cap-consuming, treat crashed reviews as retryable, and run mechanical fixups before escalation). Each requires design judgment about what counts as infra vs. code failure, how to detect a fix landing on main, and how to classify crashed reviewer runs, rather than a single fully-specified actionable item. It also touches signing policy (adding a Signed-off-by to agent-authored merge commits) and has blast radius beyond this repository, since it changes escalation behavior that affects downstream repos like rocm-cli. This is exactly the kind of design/policy decision reserved for a human.
Owner decision (Michael, 2026-10-31): all four proposals are approved as engine work. I split them into child tickets, and this issue stays open as the parent tracker.
Infra and network failures are matched against a configured list of error patterns. Matching failures don't use up the retry cap.
"A fix landed" means the failing test's file changed on main. Parked items are requeued at that point.
A crashed review is a nonzero exit with no verdict parsed. It's retried without charging a round, and it doesn't count as feedback.
Mechanical fixups are commands set per repo in config, such as cargo xtask manifest, and they run before escalating.
Signed-off-by is added only to the agent's own merge commits.
Each child ticket has its own acceptance criteria and regression tests. This issue's acceptance still applies across all four: an item reaches ready-for-human only when the remaining failure comes from the PR's code or needs a real decision, and the escalation comment names the failure class behind each retry.
Activity
mikeroysoft-agent commentedon Oct 31, 2026
Triage: This is a process/behavior change to the gate's retry and escalation logic containing four loosely-coupled proposals (re-queue after a fix lands, classify infra failures as non-cap-consuming, treat crashed reviews as retryable, and run mechanical fixups before escalation). Each requires design judgment about what counts as infra vs. code failure, how to detect a fix landing on main, and how to classify crashed reviewer runs, rather than a single fully-specified actionable item. It also touches signing policy (adding a Signed-off-by to agent-authored merge commits) and has blast radius beyond this repository, since it changes escalation behavior that affects downstream repos like rocm-cli. This is exactly the kind of design/policy decision reserved for a human.
mikeroySoft commentedon Oct 31, 2026
Owner decision (Michael, 2026-10-31): all four proposals are approved as engine work. I split them into child tickets, and this issue stays open as the parent tracker.
mainchanges the file of the known failing test.The design calls:
main. Parked items are requeued at that point.cargo xtask manifest, and they run before escalating.Signed-off-byis added only to the agent's own merge commits.Each child ticket has its own acceptance criteria and regression tests. This issue's acceptance still applies across all four: an item reaches
ready-for-humanonly when the remaining failure comes from the PR's code or needs a real decision, and the escalation comment names the failure class behind each retry.