Pick an option on each card and press Done; your answer goes straight
back to the session. The raiser's pick is marked. "I'm not sure" is a real answer.
1
In plain terms. A nightly test that simulates a big move-out job (painting, flooring, electrical, and cleaning work) has failed every night since Monday because 4 leftover test work tickets are stuck in the property software and can't be cancelled — should someone open the property software and cancel or check those 4 tickets by hand, or should Smith delete PropFlow's own copies of them without confirming the property software's side first?
The full question, as raised
turnover_abebb8b4-47c6-4d97-bc17-64baee54c66d (bench property appfolio-45 / The Willows) has 4 child WOs (WILL-003467/68/69/70, AppFolio SR#s 3016-3019) whose cancelWorkOrderL4 call returns UPSTREAM_NOT_FOUND (AppFolio's /api/work_orders?filter[service_request][id]= returns 0 rows) — permanent/non-retryable per isPermanentL4Failure. This has blocked turnover-eval-daily's turnover-multi-trade scenario every night since 2026-09-29 (2 failed nightly runs so far) and will keep failing until resolved. Should a human check/cancel SR 3016-3019 directly on jpco.appfolio.com, or authorize Smith to delete the local propflow-prod rows (the turnover + its 4 WO rows) without an AppFolio-side confirmation?
Why, and what was tried
The session raising this recommends: Check AppFolio and cancel/verify the 4 tickets by hand Deleting PropFlow's local records first would destroy the only lead back to a real AppFolio work order if one still exists there — reversible order is AppFolio first, local cleanup second. Tried first, unsuccessfully: Confirmed real (not noise): read the eval code, the nightly run's own log, and the propflow-prod rows directly — all agree the same 4 test work orders have been stuck since 2026-09-29 and every nightly retry hits the identical permanent AppFolio error. Tried to check AppFolio myself (read-only work-order list, property 45) to settle whether the 4 tickets truly don't exist there or the code is using the wrong id — the browser-login flow in my environment failed (LOGIN_NOT_LANDED) before it could answer that.
7d491ccd is parked on this.
2
In plain terms. A nightly test that simulates a big move-out job (painting, flooring, electrical, and cleaning work) has failed every night since Monday because 4 leftover test work tickets are stuck in the property software and can't be cancelled — should someone open the property software and cancel or check those 4 tickets by hand, or should Smith delete PropFlow's own copies of them without confirming the property software's side first?
The full question, as raised
turnover_abebb8b4-47c6-4d97-bc17-64baee54c66d (bench property appfolio-45 / The Willows) has 4 child WOs (WILL-003467/68/69/70, AppFolio SR#s 3016-3019) whose cancelWorkOrderL4 call returns UPSTREAM_NOT_FOUND (AppFolio's /api/work_orders?filter[service_request][id]= returns 0 rows) — permanent/non-retryable per isPermanentL4Failure. This has blocked turnover-eval-daily's turnover-multi-trade scenario every night since 2026-09-29 (2 failed nightly runs so far) and will keep failing until resolved. Should a human check/cancel SR 3016-3019 directly on jpco.appfolio.com, or authorize Smith to delete the local propflow-prod rows (the turnover + its 4 WO rows) without an AppFolio-side confirmation?
Why, and what was tried
The session raising this recommends: Check AppFolio and cancel/verify the 4 tickets by hand Deleting PropFlow's local records first would destroy the only lead back to a real AppFolio work order if one still exists there — reversible order is AppFolio first, local cleanup second. Tried first, unsuccessfully: Confirmed real (not noise): read the eval code, the nightly run's own log, and the propflow-prod rows directly — all agree the same 4 test work orders have been stuck since 2026-09-29 and every nightly retry hits the identical permanent AppFolio error. Tried to check AppFolio myself (read-only work-order list, property 45) to settle whether the 4 tickets truly don't exist there or the code is using the wrong id — the browser-login flow in my environment failed (LOGIN_NOT_LANDED) before it could answer that.