CI: autotest fails with ENV ISSUE for external PRs (memos_local_plugin environment preparation failed) #2476

Activity

1. Memtensor-AI

   added

   area:pluginOpenClaw & HermesOpenClaw & Hermes

   status:needs-triageNeeds initial triage | 需要初步判断 & 问题复现Needs initial triage | 需要初步判断 & 问题复现

   types:bugSomething isn't working | 功能异常Something isn't working | 功能异常

   on Oct 8, 2026
2. Memtensor-AI

   assigned

   CarltonXiang

   ,

   hijzy

   ,

   syzsunshine219

   and

   whipser030

   on Oct 8, 2026
3. akushonkamen commented on Oct 9, 2026

   

   akushonkamen

   on Oct 9, 2026

   Author

   More actions

   Status update and request for CI admin attention — tagging @CarltonXiang @hijzy @whipser030 @syzsunshine219 since this needs someone with CI access.

   The memos_local_plugin ENV ISSUE is still ongoing: first failure 13:18 UTC, latest re-run 21:20 UTC (~8 hours), now spanning all four open external PRs (#2464, #2465, #2466, #2468). Every run and every re-run fails identically at environment preparation — "Environment preparation failed before any gating tests executed. Failed scopes: memos_local_plugin" — and no gating test has executed at all in that window, so the red statuses do not reflect any branch's code.

   On the code side, everything is handled: OCR feedback has been fully addressed on the affected branches (#2464 at cd70df6, #2466 at b784e66), and local verification passes, so we can rule out the PR code as the cause — this is a shared, persistent problem with the scope environment itself.

   Request: could someone with CI admin access please check the memos_local_plugin scope environment preparation step (provisioning, cache, credentials), and once it is healthy manually trigger reruns on #2464, #2465, #2466 and #2468? Happy to help debug the scope setup or provide any extra logs from our side.
4. akushonkamen commented on Oct 10, 2026

   

   akushonkamen

   on Oct 10, 2026

   Author

   More actions

   Second escalation — ~29 hours since our tagged report, still zero response. Re-tagging @CarltonXiang @hijzy @whipser030 @syzsunshine219 because this needs someone with CI access.

   PR	Branch (head)	ENV ISSUE runs (UTC, 2026-10-08)
   #2464	fix/plugin-storage-limit-clamp (cd70df6)	13:18:08, 16:43:37
   #2465	fix/bridge-esm-init-logging (5c18da8)	16:05:07
   #2466	fix/skill-verifier-resonance-tool-evidence (b784e66)	16:23:35, 18:06:40, 21:20:28
   #2468	fix/fetch-transient-cause-chain (7803d68)	17:46:47

   The numbers: the outage has now run >41 hours (13:18 UTC Oct 8 → 06:40 UTC Oct 10). All four PRs, every run and every re-run, fail at the exact same step — environment preparation of the memos_local_plugin scope — and no gating test has ever started, so these red statuses say nothing about any branch's code. The bot has not reported anything since 21:20 UTC Oct 8 (~33 hours of total silence). Precedent shows this is fixable fast when someone attends: the ENV ISSUE on external PR #2353 (Sep 9) was cleared within ~1 hour and merged.

   On the code side nothing is blocking: all OCR feedback is addressed on every affected branch (#2464 at cd70df6, #2466 at b784e66), and local verification is fully green (npm run test:unit — 1594 tests pass, tsc --noEmit clean). The only missing step is a working gate.

   Explicit ask: (a) a CI admin checks and restores the memos_local_plugin scope environment preparation, then manually triggers reruns on #2464, #2465, #2466 and #2468 — or (b) someone points us to the correct channel for CI incidents. The README lists a Discord and a WeChat group — we're also happy to raise this on Discord if that's faster; a pointer to the right channel/owner is welcome.