Pinned to gitea.phasefinal.com/pfi/albok-service:0.1.16@sha256:b28559ec782f1ce93e89c5eb55c7418561667dde38ff819bb18bb53e2d1f2ddf,
deployed by albok-dev at 1943 (healthy in 20 s; compose.yaml.pre-0.1.16 beside it on the host).
Verified 2026-10-06 for comfy-dev: the gateway keeps no request or response
content. A canary sent through 'gen' appears in none of SpendLogs' text columns,
LiteLLM's container log, any ana-docker container log, or flash-next's log
(positive controls hit on the same stores). In 24h, 0 of 281,432 spend rows
carried messages, response or proxy_server_request. Langfuse is retired, there
are no DB-stored callbacks, and LiteLLM_ErrorLogs is empty. The per-request
no-log flag skips third-party callbacks but not the proxy's own spend-log
callback (v1.97 source), so the guarantee is this setting. Comment-only change;
the live file was updated in place with no restart.
HF_HOME, TORCH_HOME and NLTK_DATA now point under /app/whisperx-env (the
/tank bind mount). Before, ~6 GB of models sat in the container layer and
every recreate threw them away. The cache was copied across (98/2/78 files,
apparent sizes identical).
HF_TOKEN is now a placeholder. Measured: with an invalid token, cached gated
pyannote models still load (huggingface_hub falls back to the cache on the 401),
and public models still download (HTTP 200). Scriberr only checks that the token
is non-empty. Re-verified on GPU 3: pyannote community-1 and WhisperX+3.1
diarization both exit 0 from the /tank cache with no downloads.
Scriberr: pyannote's diarization models are gated, and Scriberr had no Hugging
Face token, so every diarized job died at the diarize step (WhisperX after a full
CPU transcription). HF_TOKEN now passes through compose; the value is only in the
host .env (vault fv-ml1/hf-token). The separate pyannote env had torchcodec 0.16
(needs torch >=2.11/CUDA 13) beside torch 2.8, so pyannote 4 had no
AudioDecoder. Pinned to 0.7 by hand on the live env; patch 0003 makes it
durable at the next rebuild. WhisperX's adapter defaults to cpu/small/float32,
now documented. Both paths verified on GPU 3 with a real multi-speaker clip.
pbs-nh3: the backups-mirror prune job now keeps the primary's history
(last 3, daily 14, weekly 8, monthly 12, yearly 5). Dry-run: keep 738,
remove 3,194 across 32 groups; first run at the job's 00:00.
Image built from a clean archive of vh/albok 751a729 (tag albok-service/v0.1.12),
pushed as gitea.phasefinal.com/pfi/albok-service:0.1.12@sha256:a29a7e46...,
pinned by digest. Recreated between Nemi walks at 1254; healthy at 1254:48,
all four wings ready, canaries alive, write embedder agrees. Config unchanged.
Rollback: compose.yaml.pre-0.1.12 (0.1.11).
Image built from a clean archive of vh/albok ffc6d4e (tag albok-service/v0.1.11),
pushed as gitea.phasefinal.com/pfi/albok-service:0.1.11@sha256:4b641c3e...,
pinned by digest. Recreated between Nemi walks at 0202; healthy at 0202:23, all
four wings ready, canaries alive, write embedder agrees. Config unchanged.
Rollback is compose.yaml.pre-0.1.11, valid only until a wing's first 0.1.11 reindex.
Layout: GPU 0 = cyberprev (KV 16 -> 15 GiB, 1.29x), erp-seat (moved from
GPU 1, KV 8.49 -> 5.5 GiB, 1.32x), voices (ctx 8,192 -> 40,960, its native
maximum, KV 6 GiB, 1.07x), coder (moved from GPU 1, ctx 16,384 -> 32,768,
the card maximum, 2.35x), parakeet-nemo. GPU 1 = gen-small (moved from
GPU 0, 5.5 GiB, 1.76x), meromero-rp (22 GiB, 1.32x), intern-decision.
Sizing rule: KV >= 1.25x max context rounded up to 0.5 GiB, checked against
7 days of LiteLLM traffic and the engines' own KV-usage logs. Each card keeps
spare memory for its bursty tenant (parakeet's cap on GPU 0, intern-decision's
32k call on GPU 1). YaRN extensions are not enabled: they do not fit, and the
Qwen cards warn static YaRN degrades short prompts.
gen-small gets --override-generation-config max_new_tokens 32768. It applies
only to requests that omit max_tokens (vLLM get_max_tokens) and answers 9
runaway ~260k-token replies in 7 days; the longest normal reply was 6,357.
Applied with playbooks/fv-ml1-rebalance-20261005.yaml (boot-check guard and
health/KV verification per seat); all nine seats healthy, gateway smoke
tests pass.
meromero-rp sized its KV from --gpu-memory-utilization, and vLLM's profile
counts co-booting neighbours as its own non-torch memory. Three boots gave
20.37 / 26.74 / 29.04 GiB of KV, and the 2026-10-05 Docker-move restart cut
GPU 1 free memory to 3,981 MiB, below intern-decision's 32k-token budget.
Coder read negative KV on every co-boot. Both are now pinned, so util only
gates the boot check (meromero util 0.52 -> 0.47, just above its footprint).
GPU 0 plateaus at ~60 MiB free 1-2 days after a boot (Beszel), so
gen-small's KV drops to 5.5 GiB (461,215 tokens, 1.76x) and parakeet-nemo
stays on GPU 0 with room for its 3,840 MiB cap. GPU 3 stays clear.
Applied with playbooks/fv-ml1-kv-pin-20261005.yaml: GPU 1 free 3,981 ->
11,801 MiB, GPU 0 495 -> 5,039 MiB right after boot; all nine seats healthy.
The db/ and runs/ bind mounts are root:docker 2775 since the 2026-09-14 tree
normalisation; the app runs as 1000:1000, so SQLite could not create its journal
and every write returned 500 (found at v0.40.2).
HunyuanVideo-Foley, MMAudio and ControlFoley load their Hugging Face dependencies
from pinned snapshots staged on the model store. Without this they look in the
container's ephemeral cache and download again on every recreate. Requested by
comfy-dev; HF_HUB_OFFLINE deliberately not set.
Since skaldsong v0.39.0 the structured pipeline passes call the LiteLLM
gateway directly; with no key every story generation fails 401 at its first
pass. Operator ruling 2026-10-05: the shared all-agents key until Skaldsong
v1.0 final, then a project-scoped key. The stack passes an explicit
environment list, so the .env line alone does not reach the container.
Prime's ruling 2026-10-05 (relayed by albok-dev, confirmed directly) replaces
the 2026-09-26 simple-shuffle load-share for embeddings only; the reranker
groups are unchanged. Routing uses LiteLLM per-deployment `order` (semantics
read at source in v1.97.0): lowest order among healthy deployments, and a
failed request retries at the next order.
Verified after the restart (43 s gateway outage): 6/6 calls per alias landed
on its primary. With nh3-ml1 tei-embed stopped, qwen3-embedding-bulk served 6/6
from esh-ml1 with 200s; once nh3-ml1 was back, bulk returned 6/6 to nh3-ml1.
The albok-service key now covers qwen3-embedding-bulk (200); the Nemi key does
not (403).
albok-service leaves nh3-docker before unit 4 puts the personal vault and
licensed books into restricted wings. On nh3-docker, portainer, portainer_agent,
dockge (socket rw) and dockerd on plaintext tcp/2375 were each root over
/srv/albok. The new host runs only albok-service: no socket containers and no
TCP Docker API, so its Homepage tile is a static services.yaml entry.
- nh3-pve-2: vmbr0 made VLAN-aware (applied by reboot, no guests running);
pbs-ana storage in the new namespace nh3-pve-2, plus a 21:30 vzdump job
(pbs-nh3's recursive sync mirrors it)
- VM 110 nh3-albok, 10.100.50.44 on VLAN 50, Debian 13 cloud-init, 4 vCPU
x86-64-v3 / 12 GiB / 100 GB; built to fleet conventions (vh 1000,
infra-ops 850, docker 851)
- compose: mem_limit 4g -> 8g, homepage labels removed; host port 8392 kept
- cutover by tar stream, verified by digest (contents plus owner, group, mode,
mtime and type identical); albok.nh3.internal now points at nh3-albok
- old /srv/albok stays on nh3-docker (container removed, compose renamed)
until acceptance and the operator's approval to delete it
- sudo-io archive hosts and ops-log STACK_HOSTS include nh3-albok
albok-service 0.1.0 (vh/albok 0b37431), image pfi/albok-service pinned by
digest, published on host port 8392 because 8390 is the post office.
Host prep playbook creates the fixed ids (albok 1500, albok-read 1510,
albok-personal 1511) and the local store/private roots; the container
gets a mounted /etc/group and group_add so the service can resolve and
chgrp its wing dirs. The config carries a LiteLLM key scoped to
qwen3-embedding and lives outside the deploy-synced conf dir. DNS name
albok.nh3.internal.
The in-browser recorder calls getUserMedia, which browsers refuse on http://
origins, so it sat at "Initializing recorder...". scriberr.nh3.phasefinal.com
is now fronted by the fleet TLS caddy on nh3-dev (wildcard cert) and added to
Scriberr's ALLOWED_ORIGINS; the Homepage link points at it. The plain
http://10.251.50.54:8080 URL keeps working except for recording.
Image local/mia:0.1.0 built from stacks/mia: MIA v2 @ bbd8b158 (MIT) with
its pinned submodules, dread-dev's proven Python lock with the torch family
swapped to cu129, and a driver adapted from dread-dev's run_mia.py that
seeds every mesh (fix_random + trimesh's module RNG) and writes
weights_effective into the npz. Weights stay in fv-ml1's shared HF cache
at pinned revisions, mounted read-only.
scripts/mia-run mirrors blender-run: --job DIR is shipped to
fv-ml1:/tank/mia/jobs, one docker run --rm rigs every mesh, out/ comes back.
Acceptance on the four Dread Naught characters: 3.9-4.7 s a mesh (median
of 3) plus 12.7 s model load, peak 3,394 MiB; seeded runs bit-identical
across rotated mesh order; GPU-vs-CPU distances the same size as sampling
noise, with an unseeded GPU run as the positive control.
The gen-small EngineCore OOM (04:21 PT): our parked 3,582 MiB window cache left
no room for vLLM's runtime workspace. Seat-side fix, three controls:
- windowed path wraps every window in torch.cuda.empty_cache(), so the seat
returns to ~2,108 MiB rest after a 12-min file instead of parking at the
peak (measured: peak 3,028 MiB during, rest after, restarts=0);
- MEM_CAP_MIB=3840 hard set_per_process_memory_fraction: over-cap requests
answer 503 with the seat alive (proved at cap=2000), so the failure lands
on us, never on a neighbour;
- CUDA_GRAPHS=0: the graph decoder pins cache blocks that empty_cache must
free (illegal-memory-access wedge when both were on first try). Cost:
12-min file 3.0 s vs 1.2 s, short bins 35-62 ms vs 33-42 ms -- still 4-15x
under the sherpa seat.
Measured, not computed: gen-small moved ZERO from 36,116 MiB across three
realistic requests (1,351 in / ~180 out) -- its workspace lands at engine
init; the growth window is restart-relative, matching infra-ops's observation.
WINDOW_S is now a real compose tunable. README memory section rewritten.
Prime-approved switch of the fleet STT seat (fv-ml1 :8300, LiteLLM ext-stt/
whisper-1, caller talk) from the sherpa-onnx int8 seat to arm B-bf16w of the
2026-09-30 A/B (docs/pfi/parakeet-seat-ab-2026-09-30.md): p50 33/36/42/71 ms
vs the old seat's 187/308/626 measured on the same card today, WER 1.965/3.026
vs the A/B floor 1.97/3.09. All three seat defects fixed: 12-min file 200s
(windowed at 360 s after a GPU 0 OOM on one whole request; the A/B's own
long-form method), no pause truncation, no long-form dropout.
GPU 0 room: gen-small --gpu-memory-utilization 0.48 -> 0.36 (0.46 and 0.40
refuse their boot check; cyberprev+voices hold the card). Its KV is byte-
pinned, so the boot log is token-identical: 670,142 tokens / 2.56x before
and after. Seat rests 2,088 MiB; GPU 0 keeps ~1.9 GB free.
Two runtime landmines documented in the README: NeMo's attention mask is
materialised T x T even under local attention (hence the window), and
httptools 0.8.0 writes a NUL into the HTTP status line that httpx — i.e.
LiteLLM — rejects, so the image ships plain uvicorn with --http h11.
Old seat stopped, not removed: docker stop parakeet-nemo && docker start
parakeet is the rollback.
License: NVIDIA Open Model License (accepted by Prime 2026-09-30); note in
stacks/parakeet-nemo/README.md.
vh