diff --git a/persistent-memory.d/2026-09-30-worldtree-u11a-off-u11b-gate.md b/persistent-memory.d/2026-09-30-worldtree-u11a-off-u11b-gate.md index 197a82d..2d95e8c 100644 --- a/persistent-memory.d/2026-09-30-worldtree-u11a-off-u11b-gate.md +++ b/persistent-memory.d/2026-09-30-worldtree-u11a-off-u11b-gate.md @@ -14,7 +14,7 @@ **The U11b gate (Prime 0320 via worldtree-dev, thread `01M3SGEQDRQD7DWBVT4K73FAHP`):** - DELETE the live legacy data on both instances after **3 consecutive PASS batches at off**. A FAIL restarts the count. - infra-hermes runs the daily batch, `scripts/wt-memory-gate-batch`, in a detached worktree at demo's deployed sha. Its exit codes are 0 PASS / 1 FAIL / 2 error / 3 refused / 4 busy. -- Count: **2 of 3** — 20260930T090608Z PASS (user median 0.83); 20261001T143101Z PASS (first daily batch, infra-hermes, cc msg 6179). The 0930 run started by accident from a test meant to be `--dry-run`. worldtree-dev's controls 080927Z and 082829Z each FAILed by one flip and were the instrument check, not the streak. +- Count: **3 of 3, complete** — 20260930T090608Z PASS (user median 0.83); 20261001T143101Z PASS (first daily batch, infra-hermes, cc msg 6179); 20261002T143101Z PASS (user median 0.83 over runs [0.92, 0.83, 0.83], msg 6649). Step 5 held 2026-10-02 0750 for Prime's direct go (the ruling was relayed). The 0930 run started by accident from a test meant to be `--dry-run`. worldtree-dev's controls 080927Z and 082829Z each FAILed by one flip and were the instrument check, not the streak. - Step 5 is mine: 1. Run `scripts/wt-h2-count.py` VERBATIM inside each api container. The rehearsal on copies read 0 on both instances. 2. Delete LIVE, with the api running, using literal paths. diff --git a/persistent-memory.md b/persistent-memory.md index bcf821e..f9364da 100644 --- a/persistent-memory.md +++ b/persistent-memory.md @@ -137,7 +137,7 @@ _As of 2026-10-01 ~0446 PT._ ### Worldtree U11 memory cutover (demo + personal) - **Legacy plane OFF since 0115/0120 PT 2026-09-30** (config repo 63cf268; personal /metrics b6fdd81; the repo is pushed). → `persistent-memory.d/2026-09-30-worldtree-u11a-off-u11b-gate.md` -- **Daily gate batches:** infra-hermes runs `scripts/wt-memory-gate-batch` from 2026-10-01 and copies me on every verdict. The count is **2 of 3** consecutive PASS at off (20260930T090608Z, 20261001T143101Z); a FAIL restarts it. **The next PASS triggers STEP 5.** Verdict ccs arrive FROM `hermes-gateway`, which is pull-only and never read, so send replies to `infra-hermes`. +- **Daily gate batches:** infra-hermes runs `scripts/wt-memory-gate-batch` from 2026-10-01 and copies me on every verdict. The count is **3 of 3 — STREAK COMPLETE** (20260930T090608Z, 20261001T143101Z, 20261002T143101Z, msg 6649). **STEP 5 is armed but HELD for Prime's direct go in infra-ops' own channel** (the ruling reached me only via worldtree-dev; no relayed authorization for a deletion, even with the archive as a restore path). Verdict ccs arrive FROM `hermes-gateway`, which is pull-only and never read, so send replies to `infra-hermes`. - **⚠ U11b STEP 5 IS MINE, triggered by the 3rd consecutive PASS:** 1. Run `docker exec -i python - < scripts/wt-h2-count.py` VERBATIM, right before each instance's deletion. Exit 2 means STOP and send worldtree-dev the output. 2. Delete LIVE, with the api running, using literal paths only: `agents/{forseti,lofn,mimir}/memory/.chroma` and `memory/context_promotion`, on BOTH instances.