joshtronic/igor
joshtronic/igor
Watch
2
Code Issues 3 Pull requests 2 Activity Actions

claude exits rc=1 at finalization on igor#334 — work abandoned on two consecutive ticks #342

New issue
Closed
opened 2026-07-04 06:13:24 +00:00 by igor · 1 comment
igor commented 2026-07-04 06:13:24 +00:00
Collaborator
Copy link

Diagnosis

The claude subprocess crashed with rc=1 on two back-to-back ticks working issue joshtronic/igor#334, each after ~10 minutes of apparently-complete work (edits done, make test/check-sync green, lint skipped, /security-review run). Both crashes land at the finalization phase (just before/at PR-body write), so ship-safety correctly refuses to commit the dirty worktree and re-queues — but the net effect is repeated wasted full-length runs with no PR shipped. A third attempt on the same issue (MVP variant) was still in flight at window close (00:59:44→), so this is not self-healed within the hour.

This is a real crash, not a timeout (elapsed 598s/607s « 30m cap) and not an auth/usage backoff. The fix(observability): preserve claude's stream on a nonzero exit commit explains the stream preserved line but does not address the rc=1 exit itself; crash-logs at .../crash-logs/20260704T002440-rc1-tier-1-issue and ...T003940-rc1-tier-1-issue are the post-mortem artifacts.

Evidence

Jul 04 00:24:40 igor tick.sh[1033813]: [agent] crash: claude exited rc=1 (tier-1-issue); stream preserved to /home/igor/.local/state/agent/crash-logs/20260704T002440-rc1-tier-1-issue for post-mortem
Jul 04 00:24:40 igor tick.sh[1033813]: [agent] claude exited 1 (elapsed 598s)
Jul 04 00:24:40 igor tick.sh[1033813]: [agent] ship-safety: claude exited 1 mid-run -- NOT committing partial worktree (4 dirty file(s)); re-queuing
Jul 04 00:24:41 igor tick.sh[1033813]: [agent] outcome: no work produced (elapsed 598s)
...
Jul 04 00:39:40 igor tick.sh[1051950]: [agent] crash: claude exited rc=1 (tier-1-issue); stream preserved to /home/igor/.local/state/agent/crash-logs/20260704T003940-rc1-tier-1-issue for post-mortem
Jul 04 00:39:40 igor tick.sh[1051950]: [agent] claude exited 1 (elapsed 607s)
Jul 04 00:39:40 igor tick.sh[1051950]: [agent] ship-safety: claude exited 1 mid-run -- NOT committing partial worktree (4 dirty file(s)); re-queuing
Jul 04 00:39:41 igor tick.sh[1051950]: [agent] outcome: blocked (repeated noop, prior count: 1)

Fixed looks like

Inspect the two preserved crash streams to identify what fails at finalization (PR-body write / security-review skill teardown). Either the underlying rc=1 stops recurring, or #334 ships a PR on retry instead of repeatedly re-queuing. A blocked (repeated noop) on the same issue across consecutive ticks should not silently keep burning full-length runs.

service: agent.service
severity: medium
window: 2026-07-04 00:00-01:00 (filed by the hourly logwatch pass)

## Diagnosis The claude subprocess crashed with `rc=1` on two back-to-back ticks working issue `joshtronic/igor#334`, each after ~10 minutes of apparently-complete work (edits done, `make test`/`check-sync` green, lint skipped, `/security-review` run). Both crashes land at the finalization phase (just before/at PR-body write), so ship-safety correctly refuses to commit the dirty worktree and re-queues — but the net effect is repeated wasted full-length runs with no PR shipped. A third attempt on the same issue (MVP variant) was still in flight at window close (00:59:44→), so this is not self-healed within the hour. This is a real crash, not a timeout (elapsed 598s/607s « 30m cap) and not an auth/usage backoff. The `fix(observability): preserve claude's stream on a nonzero exit` commit explains the `stream preserved` line but does not address the rc=1 exit itself; crash-logs at `.../crash-logs/20260704T002440-rc1-tier-1-issue` and `...T003940-rc1-tier-1-issue` are the post-mortem artifacts. ## Evidence ``` Jul 04 00:24:40 igor tick.sh[1033813]: [agent] crash: claude exited rc=1 (tier-1-issue); stream preserved to /home/igor/.local/state/agent/crash-logs/20260704T002440-rc1-tier-1-issue for post-mortem Jul 04 00:24:40 igor tick.sh[1033813]: [agent] claude exited 1 (elapsed 598s) Jul 04 00:24:40 igor tick.sh[1033813]: [agent] ship-safety: claude exited 1 mid-run -- NOT committing partial worktree (4 dirty file(s)); re-queuing Jul 04 00:24:41 igor tick.sh[1033813]: [agent] outcome: no work produced (elapsed 598s) ... Jul 04 00:39:40 igor tick.sh[1051950]: [agent] crash: claude exited rc=1 (tier-1-issue); stream preserved to /home/igor/.local/state/agent/crash-logs/20260704T003940-rc1-tier-1-issue for post-mortem Jul 04 00:39:40 igor tick.sh[1051950]: [agent] claude exited 1 (elapsed 607s) Jul 04 00:39:40 igor tick.sh[1051950]: [agent] ship-safety: claude exited 1 mid-run -- NOT committing partial worktree (4 dirty file(s)); re-queuing Jul 04 00:39:41 igor tick.sh[1051950]: [agent] outcome: blocked (repeated noop, prior count: 1) ``` ## Fixed looks like Inspect the two preserved crash streams to identify what fails at finalization (PR-body write / security-review skill teardown). Either the underlying rc=1 stops recurring, or #334 ships a PR on retry instead of repeatedly re-queuing. A `blocked (repeated noop)` on the same issue across consecutive ticks should not silently keep burning full-length runs. --- service: agent.service severity: medium window: 2026-07-04 00:00-01:00 (filed by the hourly logwatch pass) <!-- agent:logwatch -->
igor added spent time 2026-07-04 06:13:26 +00:00
58 seconds
igor commented 2026-07-04 14:23:52 +00:00
Author
Collaborator
Copy link

Closing as superseded noise (CoS cleanup). This auto-filed report describes the 50-turn cap discarding completed work — the exact failure now fixed by the hardcoded --max-turns 100 in #344, and #334 itself shipped as MVP #341 (merged). Root cause owned by #344; no separate action here.

Closing as superseded noise (CoS cleanup). This auto-filed report describes the 50-turn cap discarding completed work — the exact failure now fixed by the hardcoded `--max-turns 100` in #344, and #334 itself shipped as MVP #341 (merged). Root cause owned by #344; no separate action here.
igor closed this issue 2026-07-04 14:23:52 +00:00
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
master
agent/502-chore-reduce-agents-md-to-the-check-sync-sentinel
agent/489-chore-delete-unconsumed-in-repo-prompt-copies
agent/439-event-driven-needs-you-notification-so-the
No results found.
Labels
Clear labels   
Agent

   
Compat/Breaking
Breaking change that won't be backward compatible

  
Kind/Bug
Something is not working

  
Kind/Documentation
Documentation changes

  
Kind/Enhancement
Improve existing functionality

  
Kind/Feature
New functionality

  
Kind/Security
This is security issue

  
Kind/Testing
Issue or pull request related to testing

  
Priority
Critical
 The priority is critical

  
Priority
High
 The priority is high

  
Priority
Low
 The priority is low

  
Priority
Medium
 The priority is medium

  
Reviewed
Confirmed
 Issue has been confirmed

  
Reviewed
Duplicate
 This issue or pull request already exists

  
Reviewed
Invalid
 Invalid issue

  
Reviewed
Won't Fix
 This issue won't be fixed

  
Status
Abandoned
 Somebody has started to work on this but abandoned work

  
Status
Blocked
 Something is blocking this issue or pull request

  
Status
Need More Info
 Feedback is required to reproduce issue or to continue work

No labels
Agent
Compat/Breaking
Kind/Bug
Kind/Documentation
Kind/Enhancement
Kind/Feature
Kind/Security
Kind/Testing
Priority
Critical
Priority
High
Priority
Low
Priority
Medium
Reviewed
Confirmed
Reviewed
Duplicate
Reviewed
Invalid
Reviewed
Won't Fix
Status
Abandoned
Status
Blocked
Status
Need More Info
Milestone
Clear milestone
No items
No milestone
Assignees
Clear assignees
No assignees
joshtronic
1 participant
Notifications
Total time spent: 58 seconds
igor
58 seconds
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
joshtronic/igor#342
No description provided.