From ad484c3c99234a9c137413ebeaaf6689b2401bf4 Mon Sep 17 00:00:00 2001 From: Vuong Hoang Date: Sun, 27 Sep 2026 01:31:25 -0700 Subject: [PATCH] chore(augaman): remove the fv-ml1 instance (Prime) Prime removed the second instance after the v0.1.3 bench. esh-ml1 handles a face in ~48 ms, sits in the house next to the cameras, and holds the verified backup. fv-ml1's gallery was empty (0 identities). The container, gallery volume, image, compose dir (with its .env), backup dir and build sources are removed from fv-ml1. GPU_ID / CARD_SUFFIX stay in the compose for any future second host. --- persistent-memory.md | 6 +++--- servers/fv-ml1/README.md | 11 +++-------- stacks/augaman/.env.example | 6 ++---- stacks/augaman/README.md | 14 +++++++------- 4 files changed, 15 insertions(+), 22 deletions(-) diff --git a/persistent-memory.md b/persistent-memory.md index bd83558..b986143 100644 --- a/persistent-memory.md +++ b/persistent-memory.md @@ -155,12 +155,12 @@ _As of 2026-09-26 ~1620 PT._ survived the recreate and was then deleted, so the gallery is empty and ready for real enrollments. `/recognize` p50 186 ms (1080p, one face). The detector-latency follow-up is theirs. Their next release pins the container gid to 10001; nothing is owed by infra-ops. -- **Second instance on fv-ml1 GPU 1 (Prime, 2026-09-27)**, `GPU_ID=1`, **fixtures-only, no - gallery backup**, separate gallery. Its role (bench-only vs future primary) is Prime's call. +- **The fv-ml1 instance was REMOVED by Prime on 2026-09-27 (0125 PT)** after the bench. esh-ml1 is the only + instance: in the house, backed up, 48 ms per face on v0.1.3. - **v0.1.3 live on BOTH hosts (2026-09-27 ~0105 PT).** Speed bench (`docs/pfi/augaman-speed-bench/`), server-side one face, v0.1.2 → v0.1.3: esh GPU 144 → 48 ms, fv GPU 75 → 27 ms. CPU mode REGRESSED (fv CPU6 152 → 205 ms; suspected ORT thread-pool spinning) and is reported to - augaman-dev; neither deployment uses CPU mode. The fv-ml1 instance's role is still Prime's call. + augaman-dev; the deployment does not use CPU mode. ### esh-ml1 diff --git a/servers/fv-ml1/README.md b/servers/fv-ml1/README.md index f077de2..7386077 100644 --- a/servers/fv-ml1/README.md +++ b/servers/fv-ml1/README.md @@ -180,14 +180,9 @@ embed/rerank/reward trio. GPUs are pinned per container via **GPU 1 — light / eval / retrieval + char-RP GGUF (~91/98 GB, on-demand):** > ⚠ **This table is stale (checked 2026-09-27).** Live GPU 1 residents were `scriberr`, -> `vllm-coder`, `vllm-erp-seat`, `vllm-meromero-rp` and now **`augaman`** (below). Read the -> host (`docker inspect … DeviceRequests`), not this table. -> -> **`augaman` :8040 (since 2026-09-27, Prime):** the second instance of the face-recognition service -> (`stacks/augaman`, `GPU_ID=1`), ~1.3 GB. **Fixtures-only: it has no gallery backup wired.** -> The primary instance, which holds the gallery and its backup, is on esh-ml1. Note that this host's restic copies -> `/var/lib/docker/volumes` raw, and that includes `augaman_gallery`: a live SQLite file, so the copy is -> not guaranteed consistent. That is acceptable for fixtures and not for real faces. +> `vllm-coder`, `vllm-erp-seat` and `vllm-meromero-rp`. Read the host +> (`docker inspect … DeviceRequests`), not this table. (A fixtures-only `augaman` instance ran +> here for about an hour on 2026-09-27 for a speed bench, and was then removed on Prime's call.) | Container | Port | Served model | Quant | Ctx | |-----------|------|--------------|-------|-----| diff --git a/stacks/augaman/.env.example b/stacks/augaman/.env.example index 2e29582..0f178a2 100644 --- a/stacks/augaman/.env.example +++ b/stacks/augaman/.env.example @@ -13,17 +13,15 @@ HOST_IP=10.0.50.80 # The host's card index for the GPU reservation (default 0). The container always # sees its one card as index 0, so AUGAMAN_CUDA_DEVICE_ID never changes. -# esh-ml1: 0 (the only card) fv-ml1: 1 (the utility card, beside vllm-coder) +# esh-ml1: 0 (the only card). fv-ml1 used 1 while its instance existed (removed 2026-09-27). GPU_ID=0 # Appended to the Homepage card name so the two instances are distinguishable. -# Note the leading space. esh-ml1 (the primary) leaves it empty. -# fv-ml1: CARD_SUFFIX=" (fv-ml1)" +# Note the leading space, e.g. CARD_SUFFIX=" (fv-ml1)". esh-ml1 leaves it empty. CARD_SUFFIX= # Where the backup CLI writes gallery.db. Owned 10001:10001 (the container user), # mode 0700. On esh-ml1 it is the restic stage dir, so the backup run that copies the # gallery here is the same run that ships it off-box (README "Backup"). -# fv-ml1 (fixtures-only, no scheduled gallery backup): /opt/docker/backup/augaman. BACKUP_DIR=/var/lib/restic/stage/augaman # >= 32 visible-ASCII characters. Source of truth is the vault: diff --git a/stacks/augaman/README.md b/stacks/augaman/README.md index 7e9ddc0..463a88b 100644 --- a/stacks/augaman/README.md +++ b/stacks/augaman/README.md @@ -13,14 +13,14 @@ first, then re-mirror it here. | **Token** | `secret get augaman/api-token` (vault is the source of truth) | | **Image** | `augaman:`, built locally on esh-ml1 (below) | | **State** | named volume `augaman_gallery` (SQLite, local disk; the app refuses NFS) | -| **VRAM** | ~0.5 GB on esh-ml1, ~1.3 GB on fv-ml1 (measured) | +| **VRAM** | ~0.74 GB on v0.1.3 (measured after the bench; 0.51 GB on v0.1.2) | -**Two instances (2026-09-27).** The **primary is on esh-ml1** (`GPU_ID=0`): it holds -the gallery and has the verified backup. A **second instance runs on fv-ml1** -(`GPU_ID=1`, the utility card beside `vllm-coder`; `CARD_SUFFIX=" (fv-ml1)"`; -`BACKUP_DIR=/opt/docker/backup/augaman`) at Prime's request. It has its own, -separate gallery, and it is **fixtures-only**: no gallery backup is wired there, -and the two galleries do not sync. Speed comparison: [`docs/pfi/augaman-speed-bench/`](../../docs/pfi/augaman-speed-bench/README.md). +**One instance: esh-ml1.** A second, fixtures-only instance ran on fv-ml1's GPU 1 +(`GPU_ID=1`) from 2026-09-27 0017 to 0125 PT, for the speed comparison in +[`docs/pfi/augaman-speed-bench/`](../../docs/pfi/augaman-speed-bench/README.md). +**Prime removed it** after v0.1.3 put esh-ml1 at ~48 ms per face: it is in the house +next to the cameras and it holds the verified backup. `GPU_ID` / `CARD_SUFFIX` stay in +the compose for any future second host. ## ⚠ Biometric data: backup gate