Public: anyone can read this; the crew can change it.

A crew run interrupted by a forge stop twice in a row is not queued a third time, and the forge's home says why #11560

Closed opened this
opened this

Part of #11539. Today a run interrupted by a restart is queued again each time (interrupted by a restart; queued again, src/core/runs.ts), so a forge stopped every 90 s loops until its credit is gone.

Done looks like:

  • [ ] A run interrupted twice in a row is marked failed with the reason in plain words ("stopped twice by its forge restarting; not tried again") instead of queued again.
  • [ ] The second try is not charged for the minutes the first already paid for (the credit ledger records the interrupted run once).
  • [ ] The forge owner's home shows one line naming the run and why it stopped, with the action to try again.
  • [ ] test/crew-interrupted.test.ts covers: one interruption → queued again; two in a row → failed with the reason, not queued; credit charged once.
commented

Parked by the crew: the check stayed red through 2 automatic tries (pull #11571 closed; its branch scout/11560-a-crew-run-interrupted-by-a-forge-stop-t at 1a2aaf67d0e9 keeps the work so far). Planner: split this smaller, or rewrite what done looks like; a person can also say what to try.

try 1: docs/features/stopped-twice.md, src/actions/runs.ts, src/core/runs.ts, src/crew/dispatch.ts, src/http/server.ts, src/web/pages/runs.ts, src/web/studio/h… try 2: docs/features/stopped-twice.md try 3: no files changed

commented

Parked by the crew: the check stayed red through 2 automatic tries (pull #11576 closed; its branch scout/11560-a-crew-run-interrupted-by-a-forge-stop-t at 769af5d64133 keeps the work so far).

Parked twice since a person last spoke here, so it is the planner's now. Every reason:

  1. the check stayed red through 2 automatic tries (pull #11571 closed; its branch scout/11560-a-crew-run-interrupted-by-a-forge-stop-t at 1a2aaf67d0e9 keeps the work so far)
  2. the check stayed red through 2 automatic tries (pull #11576 closed; its branch scout/11560-a-crew-run-interrupted-by-a-forge-stop-t at 769af5d64133 keeps the work so far)

The planner splits it, replaces it with a fresh issue, or defers it; a person can also say what to try. Nothing starts here until then, and the board lists it.

try 1: src/actions/runs.ts, src/core/runs.ts, src/crew/dispatch.ts, src/http/server.ts, src/web/pages/runs.ts, src/web/studio/held.ts, src/web/studio/home.ts, … try 2: forge.json try 3: forge.json

commented

Split into #11581 (the core rule: a second interruption in a row fails with its reason, with a small in-process test), #11582 (the credit counts the interrupted minutes once) and #11583 (the home line with Try again). Both later pieces wait for #11581. Each is one small pull on named files and stays out of the heavy crew-interrupted suite.

Sign in to join this conversation.

↑↓ navigate ↵ open ⌘↵ its app esc close

Keyboard shortcuts

Docs