fleet/fleetd
3
0
Fork 0
Code Issues 73 Pull Requests Actions Packages Projects Releases 1 Wiki Activity

CB-581: fail-safe release() so a throw never orphans the pane #59

Closed
agent wants to merge 0 commits from worker/cb581-d24826-5 into main
pull from: worker/cb581-d24826-5
merge into: fleet:main
Conversation 0 Commits - Files Changed -
agent commented 2026-08-15 10:09:52 +02:00
Member
Copy Link
Copy Source

Fixes gitea issue #57.

release() called worktrees.hasUncommitted() (shells out to git status) with no protection. Any throw there skipped notifyReleased and launcher.stop, leaving an orphaned pane still burning a fleet slot and any blocked bridge_send caller unresolved. reapIdle() called release() bare inside a loop, so one throwing session aborted the rest of the reaping pass.

Changes (SessionManager.java):

  • release(): wraps the dirty-check/memberLifecycle block in try/catch/finally. On any throw, the worktree is preserved (fail toward not destroying uncommitted work, per the CB-576 precedent) and a WARN is logged naming the pane, terminal, and worktree. notifyReleased runs in finally; launcher.stop(paneId) runs unconditionally afterward so the pane always stops.
  • reapIdle(): wraps each per-session release() in try/catch (matching drainAll's existing shape), logging a WARN naming pane/terminal/worktree, so one bad session can't skip the rest of the pass.

Tests (SessionManagerTest.java, package-local RecordingWorktrees double since FakeWorktrees/WorktreeSessionManagerTest were out of scope for this ticket):

  • releasePreservesWorktreeWhenDirtyCheckThrows
  • releaseStillStopsThePaneWhenDirtyCheckThrows
  • releaseStillNotifiesTheListenerWhenDirtyCheckThrows
  • reapIdleSurvivesOneSessionThatFailsToRelease (3 idle sessions, middle one's worktree removal throws — first and third still released, all three panes stopped)
  • unchangedRegressionCleanCompletedReleaseStillRemovesTheWorktree / ...DirtyCompletedReleaseStillPreservesTheWorktree / ...ShutdownDrainStillPreservesTheWorktree

mvn clean install: BUILD SUCCESS, Tests run: 717, Failures: 0, Errors: 0, Skipped: 0 (SessionManagerTest alone: 32 tests, 0 failures).

Fixes gitea issue #57. release() called worktrees.hasUncommitted() (shells out to git status) with no protection. Any throw there skipped notifyReleased and launcher.stop, leaving an orphaned pane still burning a fleet slot and any blocked bridge_send caller unresolved. reapIdle() called release() bare inside a loop, so one throwing session aborted the rest of the reaping pass. Changes (SessionManager.java): - release(): wraps the dirty-check/memberLifecycle block in try/catch/finally. On any throw, the worktree is preserved (fail toward not destroying uncommitted work, per the CB-576 precedent) and a WARN is logged naming the pane, terminal, and worktree. notifyReleased runs in finally; launcher.stop(paneId) runs unconditionally afterward so the pane always stops. - reapIdle(): wraps each per-session release() in try/catch (matching drainAll's existing shape), logging a WARN naming pane/terminal/worktree, so one bad session can't skip the rest of the pass. Tests (SessionManagerTest.java, package-local RecordingWorktrees double since FakeWorktrees/WorktreeSessionManagerTest were out of scope for this ticket): - releasePreservesWorktreeWhenDirtyCheckThrows - releaseStillStopsThePaneWhenDirtyCheckThrows - releaseStillNotifiesTheListenerWhenDirtyCheckThrows - reapIdleSurvivesOneSessionThatFailsToRelease (3 idle sessions, middle one's worktree removal throws — first and third still released, all three panes stopped) - unchangedRegressionCleanCompletedReleaseStillRemovesTheWorktree / ...DirtyCompletedReleaseStillPreservesTheWorktree / ...ShutdownDrainStillPreservesTheWorktree mvn clean install: BUILD SUCCESS, Tests run: 717, Failures: 0, Errors: 0, Skipped: 0 (SessionManagerTest alone: 32 tests, 0 failures).
agent added 1 commit 2026-08-15 10:09:52 +02:00
CB-581: fail-safe release() so a throw never orphans the pane or aborts reapIdle
CI / build (pull_request) Successful in 52s
Details
CI / contract (pull_request) Successful in 1m1s
Details
8fd2d7e5e7 
hasUncommitted shells out to git and can throw; release() now catches that
inside a try/finally so notifyReleased and launcher.stop always run, and
defaults to preserving the worktree on a throw (can't tell dirty vs clean,
so don't risk deleting unrecoverable work). reapIdle wraps each per-session
release in try/catch, matching drainAll, so one bad session no longer
skips the rest of the reaping pass.
ltms closed this pull request 2026-08-15 12:47:36 +02:00
Some checks are pending
CI / build (pull_request) Successful in 52s
Details
CI / contract (pull_request) Successful in 1m1s
Details

Pull request closed

Please reopen this pull request to perform a merge.
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
Clear labels
blocked
Cannot start until something else lands. The body says what.
 needs-live-proof
Merged and green, but never shown working on the running daemon. Not the same as done.
 ready-to-delegate
Scope, files and acceptance criteria are written. A worker can be briefed from the body alone.
 silent-default
A feature that compiles, passes tests, and ships turned off. Nine recurrences and counting.
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
agent kevin ltms
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: fleet/fleetd#59
Delete Branch "worker/cb581-d24826-5"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?