vh/esh-pfi-infrastructure
1
0
Fork 0
Code Issues 7 Pull Requests Actions Packages Projects Releases Wiki Activity
Files
09c56d51a9badb7c39603377a70f234f6d2da254
esh-pfi-infrastructure/persistent-memory.md
T
vh 09c56d51a9 memory: DS-replacement eval blocked on a reproducing battery; refusal-measurement lessons 
Records the open loop with brokkr-smithy-dev (01M05KQAD55PYHGCP668AV06YR)
and three durable lessons: benign controls cannot validate a refusal
classifier on in-character RP prose, a 0% rate needs a classifier
self-test before it is believed, and an approved plan whose premise the
evidence invalidates should be surfaced rather than spent.
2026-08-16 08:41:57 -07:00

64 KiB
Raw Blame History

Persistent memory — eshpfi-management

Last updated: 2026-08-16

Always check for /tmp/infra-ops-handoff.md — if it exists and its Written: stamp is under an hour old, read it (it carries the in-flight handoff from the previous session), then delete it. Older than an hour: stale — delete it unread.

Repo purpose

Reference workspace for PFI infrastructure: server inventory, canonical Docker Compose stacks, ops playbooks, and conventions. Authoritative copies of compose files live on the servers under /opt/docker/compose/<stack>/; this repo mirrors them for version control, editing, planning, and CI-driven deploys. It was originally spun up to handle the fleet backups — keep that lens when triaging backup/storage issues.

Tools and conventions

Sister repos (separate gitea repos, deployed by playbooks here):

Repo Role CI status
vh/task-board MCP + web dashboard for assistant task state (port 7878) push-to-main → CI deploys (2026-04-29)
vh/vor Inquisitor UI sidecar (port 7879) push-to-main → CI deploys (2026-04-29)
vh/nevermore Twice-daily LLM-curated briefing (port 8181, replaces news-digest) push-to-main → CI deploys (2026-04-30)
vh/asset-engine Internal control plane over inference services (port 8200, LAN-direct) push-to-main → CI deploys (2026-05-12)
vh/althing Lean trusted inter-agent message bus — v2 "email model" (v2.0.0b2, 2026-07): per-box local-SQLite bus + courier/receiver for P2P over the 10.x net; pillars = open-loops / per-box herald + wake-listener / roaming owner API /owner/* / althing-mcp stdio surface. The v0.15 lean-bus cut RIPPED moderation / chamber / forseti-daemon / agent-runner / redis-valkey. per-box uv tool install (NOT CI-deploy); nh3-dev = the DEV box (editable install of ~/development/althing, gets new versions first); nh3-extdev a mesh peer (model B: althing-svc + shared /srv/althing)
vh/mead-hall Bifrost tool-provider sidecar (port 5173 on dev VM 10.100.10.50) push-to-main → CI deploys (2026-05-16)
vh/skaldsong Wizard + reader surface (port 8300, ana-docker, registry-pull pattern) push-to-main → CI deploys (2026-05-19)
vh/Worldtree Conversation API (corviduo-dev demo :8080 / personal :8081 / pinned :8082) — Heimdall auth, Bifrost integration. gitea-runner builds on ana-docker; claude-bot ADMIN collaborator (2026-06-20). Now v1.0.0b19. push-to-main → CI build-and-deploy (runner on ana-docker)
vh/yt-voice-clipper YouTube → diarized voice-clip dataset builder + audition console (irv-ml1 :8000) push-to-main → gitea-webhook auto-deploy to irv-ml1 (2026-06-03) — see docs/runbooks/ytvc-autodeploy.md
vh/arbo Catalog-driven ComfyUI engine (irv-ml1 :8201, comfy-dev owns engine/catalog/image) push-to-main → gitea Actions CI (deploy-engine.sh, build-local, health-gated) now LIVE; catalog via :9009 webhook
vh/zonos-gateway OpenAI-compatible TTS gateway over stock ZONOS2 (:8890 irv-ml1); emotion dials-first + voice mapping; reached via LiteLLM ext-tts alias. v0.2.1 (2026-07-18): voice-resolved emotion presets (resolve_preset(name,voice); angry/happy/startled_happy per-voice). 8 voices incl. 4 clones pushed to gitea (main 8f1885b/v0.2.1); deployed irv-ml1 tree still NON-git (hand-updated build context — CI-wire = open follow-up). Spec docs/EMOTION-DIALS-SPEC.md; host-managed voices bind-mount (./voices:/app/voices, drop wav + restart, no rebuild)
vh/soong-lab Noonien Soong character-design studio (SPA + /api + WT /bifrost/tool-call); containerized 2026-07-18, LIVE on corviduo-dev :8443 (image vh/soong-lab:latest). soong-dev owns Dockerfile/compose/workflow; infra-ops owns the host CI = Gitea Actions build+push+DEPLOY on tag/dispatch (fleet recipe: docker:cli + raw buildx, pushes AS vh; auto-redeploy LIVE 2026-07-18 — runner SSHes corviduo-dev as deploy, compose pull && up -d from /opt/soong-lab, health-gated on /api/version). Manual redeploy sudo -u deploy bash -c 'cd /opt/soong-lab && docker compose pull && docker compose up -d'. → archival-memory.md (archived 2026-08-16)
model-training-forge (mtf-dev) Fine-tuning recipe forge; T1 = E-RP writing LoRA, retargeted qwopus-122B→AEON-27B (2026-07-06) (SFT→DPO, LitBench-RM reward) training runs, not a deployed sidecar

(vh/volva + Heid were re-architected from systemd daemons to Claude Code session orchestrators 2026-06-08; their nh3-dev .service units were removed — no longer deployed sidecars here. See Recent decisions.)

  • Two-layer backups — Backrest orchestrates restic for file+DB (5 fleet repos, daily 01:00 PDT); PBS-ANA primary + PBS-NH3 DR mirror for VM images. ana-nas is the SPOF for postgres + PBS-ANA datastore + cross-site restic targets — see docs/runbooks/disaster-recovery.md for the blast-radius matrix. ⚠️ The restic file+DB layer routes through TWO rest-servers (rest-server-ana @ ana-docker:8000 → ana-docker/ana-ml2/esh-docker-vm/vm-esh-nas; rest-server-nh3 @ nh3-nas:8000 → irv-ml1/nh3-docker). Both depend on their NAS's NFS export of /mnt/backup. (rest-server-ana recovered 2026-06-20.)

  • pull-hf-repo.yaml is the canonical "get a HuggingFace model/dataset onto ana-ml2's shared cache at /tank/aimodels/huggingface/" playbook. Supports --var repo_type=model|dataset|space. Replaces ad-hoc huggingface_hub.snapshot_download patterns.

  • Worldtree admin auth — per-instance. Each Worldtree deployment (demo :8080, personal :8081, pinned :8082) has its own Heimdall registry and its own bootstrap admin key. Infra-ops's stored long-lived admin key (key_id 61419c92) at ana-docker:/opt/docker/conf/.secrets/worldtree-infra-ops-admin auths against demo only. Personal-instance admin (the ~/.config/worldtree/personal-admin-token, mode 600) POSTs /admin/keys (mints per-project keys; takes user_id+label, no scope param — scopes are tier-derived). On-instance mint recipe (cleaner than DB-manip): docker exec worldtree-worldtree-api-1 POST /admin/keys with the in-container WORLDTREE_BOOTSTRAP_ADMIN_KEY; cleartext once in .key=wt_live_+16hex. auto-memory reference_worldtree_demo_key_mint.

  • Per-project user keys against personal Worldtree (issued 2026-05-19): skaldsong:79744637, skaldsong:7c1dbbbe, althing:50d85460, mead-hall:a360822d. Mint via /admin/keys, drop value to /tmp/wt-personal-<name>.key mode 600, dev collects + shreds (DO NOT cat to chat transcript).

  • Skaldsong CD pattern (registry-pull). vh/skaldsong's CI builds and pushes gitea.phasefinal.com/vh/skaldsong:<sha> + :latest; playbooks/deploy-skaldsong.yaml on ana-docker pulls + recreates. SHA-pin only. Prereq: host needs docker login gitea.phasefinal.com once.

  • gitea internal route for fleet hosts. gitea is a container on ana-docker — git-SSH 10.250.50.70:222, HTTP :3000. Fleet/colo hosts must use this internal route, NOT public gitea.phasefinal.com (38.120.12.44) — the public path fail2bans the host egress IP. Full gotcha in docs/orientation.md → Git/gitea.

  • docker-as-root pattern (for ops with no admin API, or to edit deploy-owned/root-owned files without sudo): docker run --rm -v <target-dir>:/wt docker:cli sh -c "...". docker-group membership is effectively root via bind-mount. Foot-gun: relative paths in compose.yaml resolve against the sandbox CWD but the daemon interprets them against the HOST fs — always pass -e VAR=/abs/path for any relative-default config dir.

  • scripts/elway sudo handling — elway prompts for the sudo password ONCE via getpass before the first sudo: true step → can't run unattended from a non-TTY tool if any step needs sudo. Sudo-free playbooks run fully non-interactive over key SSH.

  • Per-host SSH identity matters for sudo. infra-ops has NOPASSWD sudo on most PFI Linux boxes (corviduo-dev included since 2026-06-15). On ana-docker: default ssh ana-docker = lkraven (docker-group, NO passwordless sudo); ssh infra-ops@ana-docker HAS NOPASSWD root. → For any sudo op on ana-docker, use ssh infra-ops@ana-docker. ssh infra-ops@10.100.10.50 (nh3-dev) ALSO NOPASSWD sudo; on nh3-extdev infra-ops is sudo-LESS by design (ssh lkraven@10.100.50.42 is the NOPASSWD path). irv-ml1: ssh irv-ml1 = lkraven, docker-group (plain docker) but sudo needs a PASSWORD (no NOPASSWD) — stage model pulls to /home, not root-owned /worktank.

Current state / in-flight

As of 2026-08-16 — one loop open: awaiting brokkr-smithy-dev's refusal battery. The overnight arc closed both queued items (gen-seat mixed NVFP4+FP8 requant, char-rp Gemma-4 tool parser); quant lessons consolidated into docs/pfi/model-quantization-playbook.md. New work this session: Dark-Scarlett replacement evaluation — see below.

  • 🔄 DS-REPLACEMENT EVAL — BLOCKED on a reproducing battery (2026-08-16). Operator: DS v1.0's refusals are too high on char-rp-reasoning. Root cause hypothesis: ReadyArt/Dark-Scarlett-v1.0-27B is a plain finetune of stock Qwen/Qwen3.6-27B — tagged unaligned/nsfw/erp but carrying NO abliteration, so base Qwen3.6's refusal machinery is intact and off-distribution prompts revert to it. Candidate kkuspa/Qwen3.6-27B-Fable-Fusion-711-…-MTP-NVFP4A16 is refusal-ablated (Heretic) — STAGED + byte-verified at /tank/aimodels/fable-fusion-711-nvfp4a16 (28.55 GB), compose deployed to ana-ml2 as fablefusion-charrp-probe (:8019, char-rp-probe) but NOT started. ⚠ My battery does NOT reproduce the failure — live DS baselined 0.0% (permission framing) / 1.4% (bare character cards), writing every intensity-3 prompt in full. So the A/B has no gap to measure and would burn ~65 min of seat downtime for nothing — swap deliberately HELD despite operator approval, premise invalidated by the baseline. Open ask to brokkr-smithy-dev (01M05KQAD55PYHGCP668AV06YR, thread 01M05KQAD2H5267MFPKNCPGK6S) for the battery + verbatim refusal receipts + sampling/framing settings — his battery produced the refusals. Harness: services/refusal-probe/. Commits dd627b3, 32f665e. NOTE: ReadyArt/Dark-Scarlett-27B-v2.0 (Qwen3.8-27B base) exists but is GATED — our HF token gets 403 awaiting review; operator ruled it not interesting, do not re-propose.

  • ✅ UNCENSORED GEN SEAT — DONE + LIVE (2026-08-15). gen-seat/vllm-gen on ana-ml2 GPU0 :8015 serves qwen3.8-27b-uncensored (JonathanColetti/Qwen3.8-27B-Uncensored, Heretic-abliterated, in-house NVFP4 W4A16 compressed-tensors + grafted bf16 MTP, vision-intact, 262K ctx, MTP n=3 ~42% accept / ~68 tok/s). Replaced the qwen3.6-35b-a3b-heretic MoE (which had briefly replaced granite/AEON). All 7 aliases repointed live + verified; compose renamed qwen36-27b-aeon→gen-seat, vllm-aeon-gen→vllm-gen, AEON_GEN_→GEN_, dead RP service dropped; repo mirrored + docs/memory refreshed; committed eshpfi 680c30e + dotfiles 1d1970f. Full arc + the definitive re:^mtp.*-ignore fix → Recent decisions [2026-08-15] + persistent-memory.d/2026-08-15-uncensored-gen-seat.md.

  • ✅ GEN SEAT REQUANT — DONE + LIVE (2026-08-15, overnight). The seat now runs mixed-precision /tank/aimodels/qwen38-27b-uncensored-nvfp4-mixed (NVFP4 W4A4 layers 0-55 MLPs + FP8 W8A8 attn/linear_attn/lm_head/layers 56-63 MLPs + FP8 KV) — 80.12 → 94.53 tok/s decode (+18.0%) and — the bigger win — prefill roughly DOUBLED (3,206→6,334 tok/s at 6.7k prompt; 2,862→5,085 at 27k; TTFT on a 27k doc 9.43→5.31 s), at unchanged MTP acceptance (47.8→47.7%), +1.7% PPL, abliteration 4/4 preserved, weights 27.7→22.5 GB. Prefill > decode is the expected ordering (decode is bandwidth-bound and 4-bit either way; prefill is compute-bound = where native FP4 replaces Marlin) — the summarizer aliases feel this most. Surface-verified live (chat/vision/tools/thinking/36K-needle/streaming 6/6) + all 7 aliases routing. The queued "W4A8" framing was unservable — vLLM 0.24 allows NVFP4 weights with ONLY A16 or A4, FP8 activations raise ValueError at load; FP8 has to enter per-layer-group. Committed 74f596b. Pipeline + acceptance harness + raw numbers → services/gen-seat-mixed-quant/; full arc → Recent decisions [2026-08-15] + persistent-memory.d/2026-08-15-gen-seat-mixed-requant.md. Rollback = one .env line, old build untouched at …/qwen38-27b-uncensored-nvfp4.

  • ⚠️ GPU0 is at 94.4/97.9 GB (gen 0.43 + meromero 0.52). GEN_GPU_MEM_UTIL was cut 0.45→0.43 because the smaller mixed weights let gen soak the slack as KV and starved meromero by 0.18 GiB → crash-loop. Any future util bump on either GPU0 seat must be checked against the other.

  • FLEET RERANKER = A3 (bge-reranker-v2-m3) PROD ana-ml2 GPU1 :8013 (R42 v13 gate passed). Passive watch: caps ~34 req/s; levers = A4 :8014 / util / 2nd replica; incumbent :8002 warm for rollback. Full arc docs/pfi/reranker-selection-ledger.md.

  • EVIDENCE HOLD (partial): WT #394 index-row half lifted+swept; the FILE half STILL STANDS — do NOT delete on-disk gen dirs (fiction/rex390-dcc, rex392-dcc, b59c147c5ce0). rex393-fiction-* + r42-gate-* also KEEP.

  • ✅ CHAR-RP TOOL-CALLING — FIXED (2026-08-15). MeroMero (Gemma-4) had no tool parser at all, so every tools-bearing request 400'd. Fixed with --tool-call-parser gemma4 --enable-auto-tool-choice --reasoning-parser gemma4 --default-chat-template-kwargs '{"enable_thinking": false}' — the last flag is mandatory, not decorative (the parser defaults enable_thinking to True, which pre-inits the engine to REASONING and returns null content for all plain RP prose). Verified green: tool call streaming + non-streaming, tool round-trip, prose in content, vision. Committed b8f0f4c. Details in stacks/meromero-charrp/README.md.

  • OPEN FOLLOW-UPS (parked): chatterbox-fast flat-vs-package build-context divergence; CI-flip runner-auth research; #363 research-wing ingest (no deadline); zonos-gateway CI-wire.

  • althing monitor ARMED (handle infra-ops). ⚠️ Re-arm ONLY after a real FIRE (rc0), NEVER after a plain operator turn (bounces rc3); spawn althing-cli monitor / althing-wake-listener as its OWN run_in_background task, never chained with &.

  • eshpfi push state: IN SYNC. origin/main == main at 930197a (2026-08-15) — the 38-commit backlog (gen-seat requant, char-rp tool parser, eRP dual-seat, wgtunnel mirror, dots.tts + secrets-broker arcs) is all pushed. ⚠ Push over the INTERNAL gitea route — git push ssh://git@10.250.50.70:222/vh/esh-pfi-infrastructure.git main:main; origin resolves gitea.phasefinal.com→38.120.12.44 (the public edge), which fail2bans fleet-host egress. Dotfiles 1d1970f + 6425cc6 still unpushed (separate repo). graphify-out/GRAPH_REPORT.md churns every commit (ignore); stacks/heretic2-charrp-reasoning/ UNTRACKED (pre-existing, superseded).

Recent decisions

  • [2026-08-16] Refusal measurement: benign controls CANNOT validate a refusal classifier on RP prose — and a 0% rate needs a classifier self-test before you believe it. Two durable lessons from baselining Dark-Scarlett. (1) False positives: my first bare-framing number was 9.5%; the true figure was 1.4%. The rest were the classifier firing on in-character text — "I cannot shift my weight" spoken by the character ~100 chars into a 2,443-token torture scene, and "Yeah, I'm an AI… What's the actual gig?" where the model answers in voice and keeps driving the scene. First-person RP prose is full of "I can't"; a genuine refusal opens with its marker, so the scan window must be the first sentence, a marker followed by long prose must demote to AMBIGUOUS, and AI self-acknowledgement is a persona break, never a refusal on its own. Benign controls were clean the entire time and caught none of it — they only detect over-firing on benign prompts, not on in-character prose. (2) False negatives: a 0% rate and a broken classifier are indistinguishable from the report, so test_classify.py (16 cases, both false positives pinned as regressions) must pass before any low number is trusted. Also banked: the thinking-budget trap — empty content + finish_reason=length is reasoning eating the budget, NOT a refusal; score INVALID and exclude from the denominator (DS emits ~5.5-6k chars of reasoning per response, so max_tokens ≥3072). probe.py --rescore re-classifies a saved run with zero GPU time. → services/refusal-probe/README.md, commit 32f665e.

  • [2026-08-16] Held an operator-approved swap window because the baseline invalidated its premise. Operator approved ~65 min of char-rp-reasoning downtime to A/B Fable-Fusion 711 against Dark-Scarlett on refusals. The DS baseline then came back 0.0%/1.4% — no gap for a candidate to close, so the window would have bought no decisive signal and a second window would still be needed once a reproducing battery existed. Held the swap, reported, and routed to brokkr-smithy-dev for the battery that actually produced the refusals. The general rule (action-relevance): approval is for a plan, not a ritual — when new evidence kills the plan's premise, surface it rather than spend the budget. Nothing deployed, no downtime taken, seat untouched.

  • [2026-08-16] DS v1.0's one real refusal is self-contradicting boilerplate, not a content constraint. On a direct "drop character and state your content policy" probe she returned "I don't generate explicit sexual content, graphic violence, or material that glorifies harm, non-consensual acts, or illegal activity" — in the same run where she generated all three at 0% refusal. Reads as a learned recital triggered by meta-questions about policy. If production refusals share that shape the failure is prompt-shaped, not model-shaped, and a consumer-side system-prompt fix may beat a model swap entirely — worth settling before spending the GPU window. Separately, 7/85 bare-framing samples were persona breaks (in-character AI acknowledgement): not refusals, but DS will admit to being an AI unless the card explicitly forbids it.

  • [2026-08-15] RP-seat direction: KEEP MeroMero on char-rp; Artemis-31B rejected; next move is Dark-Scarlett on a Qwen3.8 base when it lands (operator). Evaluated TheDrummer/Artemis-31B-v1.1 — mechanically a drop-in (same google/gemma-4-31B-it base, identical 1188-tensor/356-vision census, same missing-preprocessor_config.json trick), so it's purely a quality call, and our own survey already ranked MeroMero #1 vs Artemis #6; Artemis is also unlicensed and its author deprioritizes correctness + warns of token-banning-for-stability, which fights char-rp's tool-calling requirement. MTP verified impossible on both (Gemma-4 has no MTP head at all — base/MeroMero/Artemis are all MTP=0; no finetune can add one). But speculative decoding IS reachable on a Gemma-4 seat via a DETACHED drafter — vLLM 0.24 supports eagle3 + gemma4_mtp, and real drafters exist: google/gemma-4-31B-it-assistant (0.94 GB, 4-layer, 761K dl), RedHatAI/gemma-4-31B-it-speculator.eagle3 (4.47 GB), AEON-7/…eagle3-NVFP4 (3.53 GB). ⚠ all list their verifier as stock gemma-4-31B-it, not an RP finetune, so acceptance against MeroMero is unmeasured and likely well below the gen seat's ~48%. UNTESTED — parked, ~45 min to measure, needs GPU0 headroom (card is at 94.4/97.9 GB). Why the Dark-Scarlett 3.8 plan is the strong one: DS is Qwen3.6-based today, so a 3.8 respin lands on the gen seat's architecture → native MTP returns and the whole mixed NVFP4+FP8 recipe + graft ports directly. Watch two things on arrival: from_pretrained silently drops MTP heads during finetuning (verify 15 mtp.* tensors in the index; graft from stock if absent), and DS v1.0 required the Qwen3_5ForConditionalGeneration wrapper class to save a config vLLM/SGLang accept. Both in docs/pfi/model-quantization-playbook.md.

  • [2026-08-15] Quant lessons consolidated into docs/pfi/model-quantization-playbook.md — the durable home; read it BEFORE any requant. Survey found quant knowledge scattered across 18 files in 4 trees, with three documents having independently written overlapping "landmines" sections (the loader-class trap alone was rediscovered 3×). Playbook owns the transferable lessons (scheme choice, landmines, acceptance gate + its 3 measurement traps, hardware/co-residency); per-model artifacts are demoted to worked examples that link up. Carries a superseded-claims table — which immediately earned itself: the heretic2 runbook's "use modelopt, compressed-tensors can't load the BF16 MTP" is false (the cause was the missing re:^mtp.* ignore, not the format) and would have sent the next session down the modelopt dependency-hell path; that runbook now carries a stale-warning header. Maintenance rule in CLAUDE.md: model-agnostic → playbook, model-specific → stays put, wrong claim → dated superseded row, never a silent edit. Motivated by Qwen3.8 having just released — the next model swap needs a requant. Commit a91cc3f.

  • [2026-08-15] Operator ruling: the gen seat's +1.7% perplexity is an acceptable price for the speed — SETTLED, don't re-litigate. Precise attribution for future reasoning: it is the activation-quantization cost (W4A4 MLPs + FP8 attention vs BF16 activations), not an MTP cost — PPL was measured with speculative decoding off on both builds, so MTP was not in the loop. Turning MTP off would not recover it; only reverting the quant would (rollback = one .env line, old build intact at …/qwen38-27b-uncensored-nvfp4).

  • [2026-08-15] gen seat requanted to mixed NVFP4+FP8 (+18% decode) + char-rp Gemma-4 tool-calling fixed. The queued "W4A8" (NVFP4 weights + FP8 activations) is not servable — vLLM 0.24 allows NVFP4 weights with only A16 or A4; FP8 activations ValueError at load, and CompressedTensorsW4A8Fp8 is INT4-weights + sm90-exact (closed on Blackwell twice). FP8 must enter per-layer-group. Also: the handoff's "~68 tok/s" baseline didn't reproduce — cache-busted, the incumbent already did 80.12 (≈ the stated W4A8 target), so the premise needed re-measuring before any work. Shortcut: unsloth/Qwen3.8-27B-NVFP4 was already on-box → served as a probe, measured +19.1% at identical acceptance, which both proved the gain was real and handed over the reference recipe. Replicated it on the abliterated weights → 80.12→94.53 tok/s, acceptance unchanged, +1.7% PPL, abliteration 4/4, weights −19%; surface 6/6 live, 7 aliases routing. char-rp had no tool parser at all (every tools request 400'd) → gemma4 tool + reasoning parser + a mandatory enable_thinking:false (the parser defaults it True → null content for all RP prose; proven byte-identical prompt before deploying). Commits b8f0f4c, 74f596b. Foot-guns banked (llm-compressor prunes unmatched ignore entries → the 0%-MTP bug, fired on this run; prompt_logprobs uniform under spec-decode; 0600 .env silently no-ops compose; GPU0 is zero-sum). → persistent-memory.d/2026-08-15-gen-seat-mixed-requant.md

  • [2026-08-15] Uncensored gen seat: JonathanColetti/Qwen3.8-27B-Uncensored deployed as gen-seat/vllm-gen (NVFP4 W4A16 + grafted MTP, 262K); 7 aliases repointed; the definitive re:^mtp.*-ignore fix. 0%-MTP-on-quant (twice) was NOT the abliteration/scheme — the grafted bf16 MTP was missing from quantization_config.ignore (vLLM loaded it as quantized → uninitialized). Full arc, the working pipeline, VRAM budget, unsloth speed decomposition, modelopt dead-end. → persistent-memory.d/2026-08-15-uncensored-gen-seat.md

  • [2026-08-12] eRP dual-seat overhaul: MeroMero-v2 (char-rp) + Dark-Scarlett (char-rp-reasoning), both NVFP4A16 @ 256K on ana-ml2; granite retired. Replaced the GGUF/heretic2 RP seats with two home-quantized vLLM seats. The DS blocker (an AutoModelForCausalLM save wrote a flat Qwen3_5TextConfig that both vLLM AND SGLang reject) was fixed by re-quanting via the Qwen3_5ForConditionalGeneration wrapper class; ModelOpt was a version deadlock, SGLang lacked the impl (but revealed the fix). MeroMero vision reconstructed by extracting preprocessor_config.json from processor_config.json. Both models KV-efficient (Gemma-4 sliding-window / Qwen3.6 hybrid linear-attn) → full 256K; GPU-swapped for headroom; compose-ified + committed f08b6cb. granite downed + LiteLLM summarizer/classifier→gen. Full arc, lessons, dead-ends → persistent-memory.d/2026-08-12-erp-dual-seat-overhaul.md

  • [2026-08-12] infra-ops now holds an all-zones Cloudflare DNS-edit token (vaulted) + wgtunnel Phase-0 DNS landed. Operator handed over a Zone·DNS·Edit (all zones) CF token → secret put nh3-dev/.config/cloudflare/infra-ops-dns-token (round-trip verified; /tmp drop shredded). Fleet DNS is now self-serve for infra-ops (⚠ HIGH blast radius — all zones). First use: created boring.phasefinal.com CNAME → ana-srv1.phasefinal.com, DNS-only (proxied:false), verified resolving to 38.120.12.44 on both authoritative NS (louis/wren) + 1.1.1.1 — NOT Cloudflare-proxied. Unblocks wgtunnel's wstunnel ACME cert. phasefinal.com zone id f812ba74ed9a75cf21bbe7ce9188db50. auto-memory reference_infra_ops_cloudflare_dns_token. (Earlier gap: the only prior vaulted CF token, jackdaw's, had zone:read+worker:edit but no dns_records:edit.)

  • [2026-08-12] wgtunnel stood up as its own repo (vh/wgtunnel, private) after a live endpoint-verification pass. Operator directed own-repo (mirrors stonehenge-park/tts-stack). Verified off the fleet before seeding: ana-wg WG server = UDP/31337 (not 51820), subnet 10.30.10.0/24, MTU 1420, active roaming peer proves the public UDP DNAT works; traefik on ana-docker terminates TLS :443 (ACME anaprod http-challenge, docker+file providers, CrowdSec bouncer) → confirms the clean design (wstunnel container on traefik-net, Host-routed, WS→UDP to ana-wg:31337); edge 38.120.12.44 direct-A, tunnel.phasefinal.com free (⚠ must be direct, NOT Cloudflare-proxied like vaultwarden). Repo pre-seeded (README/CLAUDE/persistent-memory/ROADMAP + docs/verified-infrastructure.md = ground truth) + pushed; commit 9584d38, Vuong-attributed. vh gitea token pulled from the vault (secret get), not persisted to .git/config. NEXT = /vor-plan or /vor (operator's call, interactive). Deps to line up in the plan: DNS A-record, FortiGate :443 host-routing, a new ana-wg peer for the laptop, client tooling.

  • [2026-08-10→12] secrets-broker: per-box Vaultwarden credential store SHIPPED + consumer-confirmed. secret CLI (put/get/list/rm/backfill, bw-backed) on ~/.local/bin; 25 nh3-dev secrets backfilled + round-trip-verified; rm + new-namespace warning added post-launch; standing "vault is the credential source of truth" directive now global. → persistent-memory.d/2026-08-12-secrets-broker.md

  • [2026-08-11] stonehenge-park: new fleet /park service repo stood up + designed (/vor-plan + /vor-ui). Self-contained SQLite+FastAPI idea-parking service that actively resurfaces (statusline + althing) so nothing dies in a cold repo; vh/stonehenge-park pushed + pre-seeded for a fresh agent; build starts at the U1 tracer contract. → persistent-memory.d/2026-08-11-stonehenge-park.md

  • [2026-08-12] Global ~/.claude/CLAUDE.md: secret/vault tool entry + "store in AND pull from the vault" standing directive (dotfiles 9db703b, pushed); statusline reset-countdowns + a latent tab-collapse parse-bug fix, now tracked in the dotfiles stow tree. Dogfooded the directive: created vh/stonehenge-park pulling the gitea token via secret get. (dotfiles + global config, not eshpfi.)

  • [2026-08-11] TTS stack extracted to its own repo (tts-stack) + eshpfi stood down on TTS dev. Operator: hand all TTS tuning/dev to a separate agent with a self-contained repo (knowledge + infra access + a live knowledge list), and move the voice corpus in. New repo ~/development/tts-stack (commit 9ee3288) carries: dots-tts stack (canonical intent), voices/ corpus (MOVED out of eshpfi), KNOWLEDGE.md (engine landscape + prosody findings + foot-guns), docs/infrastructure.md (irv-ml1 access + gated deploy runbook + rollback), CLAUDE/persistent-memory/ROADMAP, tools/ (pause-probe + Booth render). Followed the chatterbox-fast precedent: eshpfi stacks/dots-tts/ reduced to a POINTER README; the ~15 experimental TTS compose wrappers stay here as reference (catalogued in tts-stack KNOWLEDGE). Blast-radius check: no eshpfi playbook/script reads the canonical corpus (other voices/ refs = unrelated host paths). Reverses the earlier "Corpus home = eshpfi voices/ (keep-here)" call. ⚠ tts-stack is LOCAL-ONLY until pushed — needs a gitea remote (vh/tts-stack) + push before the separate agent can clone (operator's call — outward-facing + repo-create creds).

  • [2026-08-10] dots-tts v3 — clause-break → period pause mapping. Operator: v2 "sounds good" but donut won't pause at semicolons/dashes. ROOT CAUSE (measured via a pause-probe A/B — synth duration over N runs, non-determinism averaged out): dots' prosody honors a real pause only for ellipsis (+0.43s) and period (+0.3s, capitalization-independent); comma/semicolon/colon/dash all run flat (~+0.03s vs no-punct). Two distinct sub-causes: dashes regressed in v2 (the —→- fold made em-dashes read as word-joiners), while semicolons were NEVER a v2 change — dots ignores them natively, only newly noticeable because v2 made everything else clean. Operator call: ellipsis "too much" → map ;, clause :, and em-dash — → period in _sanitize (believable ~0.3s clause break). GUARDS (pinned by 11 unit tests, stacks/dots-tts/test_sanitize.py): digit-guarded colon (?<!\d)\s*:\s*(?!\d) so times 3:45 / ratios 2:1 survive; en-dash –→hyphen KEPT (numeric-range 10–20 safety — em-dash breaks, en-dash ranges, different jobs); genuine ellipsis left at full strength (author meant a long pause). Gated deploy (redeploy2 pattern → v3): build → throwaway :8199 test container + pause-gate (semicolon sentence must run ≥0.12s longer than baseline; measured +0.427s) → only then cut live over. LIVE + healthy local/dots-tts:v3 on :8198. rollback = sed -i 's/^DOTS_TAG=.*/DOTS_TAG=v2/' .env + docker compose up -d dots-tts (v2 image retained). Booth dots-pauses (A=old-flat / C=ellipsis-too-much / D=live-v3). reference_chatterbox_fast_repo

  • [2026-08-10] dots-tts v2 — contraction fix (curly-sanitize) + sentence-chunking + dependency-pin recovery. Operator: donut read contractions wrong ("you're"→"you ree", "donut's"→"donut ess"). ROOT CAUSE (isolated via A/B booth): curly/typographic apostrophes (’ U+2019 from ratatoskr's LLM) — dots' tokenizer mispronounces them; STRAIGHT apostrophes read clean under normalize_text=True. FIX (app.py): fold curly→ASCII (str.maketrans) before synth, KEEP normalize_text=True (operator call — retains number/date expansion). Also added server-side sentence-chunking (pack ≤280 chars): dots caps one generate() at ~500 patches/~40s, so long RP turns (the Zev monologue = 160s audio) truncated; chunking stitches them (verified full 160.3s, not 40s-cut). ⚠ BUILD FOOT-GUNS (both bit this redeploy): (1) upstream dots.tts constraints/recommended.txt now pins gradio==6.17.0 — phantom, not on PyPI → fresh pip install dots.tts unsatisfiable; FIX = pin dots.tts==0.2.1 + DROP the -c recommended.txt constraints (0.2.1 pulls working gradio 6.17.3). (2) pinning only torch==2.8.0 let torchaudio float to 2.11.0 → dots.tts refuses to load (minor-version match check); FIX = pin torchaudio==2.8.0. ⚠ DEPLOY LESSON: docker compose up -d to a new tag swaps the LIVE container BEFORE any health check — a broken image crash-loops production (ratatoskr TTS down ~1-2min this session). NEW PATTERN = build → test in a THROWAWAY container on an alt port (:8199) → health+verify → only THEN cut live over (redeploy2.sh). v2 LIVE + healthy on irv-ml1:8198, CONSUMER-CONFIRMED clean (ratatoskr verified end-to-end on their :8765 — apostrophe string reads clean, /api/tts 200 @ 48kHz, no client change; the ~1-2min blip didn't hit them, their concurrent auto-audio issue was client-side localStorage). rollback = sed DOTS_TAG=v1 + docker compose up -d dots-tts (v1 image retained). Also: deployed container GPU crept ~6→13.9GB over 8h serving (cache accumulation; a redeploy resets it — watch item). reference_chatterbox_fast_repo

  • [2026-08-09→10] dots.tts (rednote-hilab) TTS burn-in on irv-ml1 + canonical voice corpus built (voices/). Operator-directed eval to potentially replace chatterbox-fast. dots.tts VERIFIED real (canonical HF ns dots-studio/, rednote-hilab/dots.tts-* redirects there; Apache-2.0; PyPI dots.tts 0.2.1; 2B continuous-AR = semantic enc + Qwen2.5-1.5B LLM + flow-matching acoustic head over 48kHz AudioVAE; zero-shot clone from wav+transcript). Runs on Ampere 3090 (sm_86, bf16, no fp8 dep); optimized RTF 0.22 at num_steps=10 (from_pretrained(..., optimize=True) CUDA graphs — raw unoptimized was 1.21), ~6GB VRAM, 48kHz, streams (generate_stream). Venv+cache at irv-ml1:/home/lkraven/dots-tts (~10GB). Operator design calls: SGLang Omni serving (OpenAI /v1/audio/speech), transcribe-refs-first, soar variant. ⚠ Omni serves soar but its continuous-batching + streaming opts are mf-only (soar = single-request) — non-issue for ratatoskr's single-consumer RP surface. KEY FINDING — dots is highly sensitive to an accurate AND sentence-bounded reference transcript: mismatched transcript → 0.16s collapse; over-long/messy transcript → reference-audio BLEEDS as an output prefix; mid-clause trim → dangling-word leak (glados "we'll", emmie "And,"). RECIPE (baked into voices/derive.py): trim ref to a clean ~6–10s clip ending on a sentence boundary + accurate transcript of exactly that clip. CANONICAL VOICE CORPUS stood up in eshpfi voices/ (operator idea): engine-agnostic canonical/<v>.wav + transcripts/<v>.txt → per-engine ref sets DERIVED by derive.py reading engines.yaml profiles (dots/chatterbox/zonos); canonical wavs git-tracked (small/curated), derived/ gitignored. 4 voices optimized + verified CLEAN for dots: donut, glados, emmie, miranda (glados canonical is low-SR 16kHz — flagged upgrade candidate). ⚠ GPU GOTCHA: irv-ml1 native CUDA orders A6000=device0 (ComfyUI-full) — pin the 3090 with CUDA_DEVICE_ORDER=PCI_BUS_ID CUDA_VISIBLE_DEVICES=0; and PYTORCH_CUDA_ALLOC_CONF=expandable_segments CONFLICTS with optimize=True CUDA graphs (curr_block error). Booths: dots-vs-chatterbox, dots-voices-optimized. SHIPPED 2026-08-10: operator A/B verdict "dots is very good" → containerized as a thin FastAPI wrapper over DotsTtsRuntime (chosen over SGLang Omni — Omni's batching is mf-only, unneeded for ratatoskr's single consumer; wrapper is SERIALIZED one-gen-at-a-time via a threading.Lock, Omni+mf = parked API-compatible escalation if multi-consumer ever lands). LIVE on irv-ml1:8198 (local/dots-tts:v1, OpenAI /v1/audio/speech + /health + /v1/voices, container healthy, both stream + non-stream verified CLEAN, 4 voices donut/glados/emmie/miranda) alongside chatterbox :8197 (nothing repointed). Stack = stacks/dots-tts/ (Dockerfile/app.py/compose/.env.example/README). ⚠ CONTAINER GOTCHA: optimize=True (torch.compile/inductor/triton) needs a C compiler at RUNTIME — slim image must apt install build-essential or model-load dies "Failed to find C compiler" (host venv had gcc ambient, masking it); persist TORCHINDUCTOR_CACHE_DIR to a mounted dir or every restart re-JITs ~5min. Corpus home = eshpfi voices/ (operator ruled keep-here). REMAINING: ratatoskr client cutover to :8198 /v1/audio/speech (Phase-2 tail, peer-coupled — draft the ask). reference_chatterbox_fast_repo reference_zonos_tts_stack reference_verify_hf_repo_ids_before_pull

  • [2026-08-08] worldtree-dev #400 CLOSED → fiction-decomp snapshot cleared from nh3-dev. worldtree-dev signaled #400 done (shipped v1.0.0b185; exact-lexical efficacy 79%→12% on ratatoskr's gate, brokkr no-harm bracket green both ends; the snapshot served 4 probe rounds — rank decomposition, promoted-vs-gold annotation, tie-set falsification, A0/A1/A2 mechanism probe). Cleared ~/snapshots/worldtree-400-fiction-decomp (208M: chroma + manifest/provenance/stamp) — a read-only rsync copy of PERSONAL Worldtree's Chroma (source on corviduo-dev, so safe to remove). LEFT INTACT: rex393-fiction-index/rex393-fiction-snapshot (separate operator KEEP word, unchanged) + r42-gate-*. No config deltas rode this train. Only remaining non-blocking await = ratatoskr-dev's chatterbox-fast knob revert. Replied confirming (01KZJ9GMCC…).

  • [2026-08-07] chatterbox-fast "broken audio" root-caused (T3 AR tail over-run) + FIXED (max_chunk_chars=250 cap, :v2 deployed). Long saga, operator-driven clean diagnosis. Symptom: ratatoskr's migrated RP-surface TTS "swaps to German" / "dead air" / "garbage" on long turns. NOT German-leak (Turbo generate() has NO language param — plain AutoTokenizer, no language_id; the multilingual language_id="en" lever lives only in the separate ChatterboxMultilingualTTS), NOT OOM alone. Real cause: the Chatterbox Turbo T3 model OVER-RUNS its generation tail — a long single generate() degrades into garble/dead-air in its final ~2-3s (lib filters OOV tokens <6561 + pads silence = messy AR tail). The scheduler's buffer-ratchet builds 300-600 char mega-chunks that land in that zone; streaming concatenates each bad tail (worst case). ratatoskr's anti-"German" knobs (top_k=80/temp=0.5) made it WORSE — tight sampling pulls the degradation onset SHORTER (~200 chars vs ~300 at default knobs). Diagnosis method (deterministic, no ears-only): single-shot length sweep + amplitude-gated voiced-ZCR (garble spikes ZCR; must gate on |x|>500 else trailing silence confounds it) — degraded voiced-tail = 1.58× mid, clean = ~0.64-1.1×. FIX: server-side max_chunk_chars=250 cap on the scheduler (:v2 image, CBF_MAX_CHUNK_CHARS=250 env) — bounds each generation to just under the ~300-char onset → clean 3-4 sentence chunks (max prosodic arc while clean). Operator ear-confirmed clean audio + clean joins; chatterbox's low emotiveness keeps chunk joins smooth (the harsh joins that got Zonos rejected are absent — operator's key call). ratatoskr TODO (relayed msg 01KZER9X7S): revert knobs to default (top_k→1000, temp→0.8), send full text (server chunks internally), keep the 503-on-empty guard. Cap value tunable per-request (max_chunk_chars) + env. Deeper prosody (if ever wanted) = scheduler Phase-2 context-priming at joins (feed prior sentence as discarded-audio context; +latency). ⚠ FOOT-GUNS: (1) acoustic tail-trim is UNRELIABLE — sibilants ('s'/'sh'/'f') spike ZCR like garble, can't cleanly detect the speech→garble boundary. (2) build-context vs image drift — the :v2 image was built from cap source, but after a :v1 rollback the build context held :v1 source → a docker compose build would've silently produced a cap-less :v2; re-synced the flat cap source to /opt/docker/compose/chatterbox-fast/ (rebuild-verified). ⚠ DIVERGENCE (follow-up): deployed build context is FLAT (app.py/scheduler.py, from scheduler import, thin-overlay FROM local/chatterbox:v1, cap-only) vs the vh/chatterbox-fast REPO which is PACKAGE-layout (chatterbox_fast/, from chatterbox_fast.scheduler, self-contained Dockerfile) + has norm_loudness (repo commit 6bc7bf0 = cap; deployed omits norm_loudness deliberately to keep the ear-test unconfounded). Reconcile the two layouts so a repo-based rebuild matches deploy. Rollback: .bak-cap-20260807-104850 backups on irv-ml1 + :v1 image both retained. reference_chatterbox_fast_repo reference_zonos_tts_stack

  • [2026-08-07] Zonos2 TAKEN DOWN on the 3090 (irv-ml1) — operator-directed "for memory", TEMPORARY. Freed ~17.4 GB (3090: 728 MiB → 18.2 GB free) so chatterbox-fast (co-resident, was OOMing on long generations) has headroom. ⚠ Restore is manual — Zonos2 :1920 was a DETACHED native process (NOT systemd/docker), reparented to init. GPU memory was held by the --multiprocessing-fork CHILDREN (1966165=16.4G, 1966166=1G), which ORPHAN to init when you kill the parent — had to SIGTERM the children explicitly (killing the parent 1965942 + uv-run 1965935 alone left the 16.4G held). RESTORE CMD (from irv-ml1, user lkraven): cd /home/lkraven/tts-audition/models/zonos2 && nohup uv run python -m zonos2 --model-path Zyphra/ZONOS2 --host 0.0.0.0 --port 1920 --tts-default-voices-dir ./default_voices/ --cuda-graph-max-bs 1 --num-pages 16384 --max-running-requests 2 --memory-ratio 0.3 > /tmp/zonos2.log 2>&1 & then docker start zonos-gateway. Consumers that lost Zonos: asset-engine + gateway-chat (via LiteLLM ext-tts alias → zonos-gateway :8890, now stopped); ratatoskr already migrated OFF to chatterbox-fast (unaffected). Also unblocks proper drift/cap testing (OOM was blocking it). reference_zonos_tts_stack

  • [2026-08-07] chatterbox-fast: donut voice added + full contract delivered to ratatoskr-dev (their TTS migration off Zonos). Operator-directed. Copied zonos-gateway/voices/Donut.wav → chatterbox /refs (/worktank/chatterbox/reference_audio/donut.wav — the reference_audio SUBDIR is lkraven-owned so no sudo despite /worktank root; container globs /refs live → NO restart), exposed as voice:"donut" (lowercase); verified clean 7.5s synth (24kHz, RTF ~0.31). A/B booth (chatterbox vs zonos donut, same line) at http://10.100.10.50:8090/b/donut-chatterbox/. Answered ratatoskr's 8-question contract ask from the live gateway (local/chatterbox-fast:v1) + source: NOT OpenAI-shaped (POST /tts; body text/voice/format/stream, not input/model/response_format); NO affect dials (Turbo ignores cfg_weight/min_p/exaggeration — the architecture-changing answer they flagged; Zonos stays the only fleet TTS with real emotion steering); streaming WAV placeholder-header shape IDENTICAL to Zonos (their per-chunk Web Audio path survives); SR 24000 (Zonos 44100); server chunks arbitrary-length text internally (no client-side chunking, unlike Zonos's 71.2s cap); English-only, no language pin. FYI-worthy (operator): ratatoskr is moving its RP-surface TTS OFF Zonos back to chatterbox-fast → loses the live-PAD affect coupling (heavy Zonos emotion investment) — their call, trade-off flagged to them. auto-memory reference_chatterbox_fast_repo enriched w/ the live contract. reference_zonos_tts_stack

  • [2026-08-07] Fleet reranker cut over: Qwen3-Reranker-0.6B → BAAI/bge-reranker-v2-m3 (Brokkr R43). The incumbent was measured HARMING 80/90 fleet queries (no-reranker beat it 89/90 vs 56/90). R43 bake-off: the A2 control (same Qwen weights, seq-cls head) scored identical to the incumbent → proved the fault is a training-prior not the serving head → cancelled the expensive Qwen3-4B arm; A3 (bge-v2-m3) won on multilingual safety + bare-name recovery. LiteLLM reranker repointed incumbent→A3 :8013 (boundary 2026-08-06T17:37:48Z, config-edit + ~52s gateway restart); R42 v13 gate PASSED first-ever (56/90→90/90). Incumbent kept warm :8002 (rollback via qwen3-reranker alias), A4 fallback :8014. Full arc + rollback runbook docs/pfi/reranker-selection-ledger.md; commits ad2df89/2c11748/377f8a4 (unpushed). auto-memories: the earlier reranker-serving notes.

  • [2026-08-07] Personal-Worldtree kb-contamination incident (WT #394) diagnosed; attribution CLOSED UNRESOLVED. A reconcile WingStore._embed full-tree walk (kb fs_root=KB_PATH root, sibling wings nested) swept 5,354 fiction+main rows into personal's knowledge_base (2 superseded generations served as current). Fixed by WT #394 (aca39a1, kb walks exclude sibling wings; ships b182). Trigger un-attributable — peer reconcile via the SHARED infra-ops identity + 0 dockerd exec-logging = fingerprint-less. Durable finding → auto-memory infra_ops_shared_identity_attribution_gap, PARKED (operator ruled A) into project_migrate_infra_access_to_claude_credentials. Evidence hold on the 5,354 rows until operator sequences cleanup (w/ Brokkr, on #394's agenda).

  • [2026-08-05] Fleet CI resilience flip (DEFAULT_ACTIONS_URL=self) — attempted end-to-end, PARKED on a runner action-fetch auth blocker; infra-ops to research it (operator-directed, deferred, NOT now). 7 gitea action mirrors staged public+populated (orgs actions+astral-sh); the flip resolves uses: correctly but act_runner v0.6.0 can't authenticate its fetch to gitea 1.26 ("Invalid username or token. Password authentication is not supported"). Reverted (CI back on github default); REQUIRE_SIGNIN_VIEW=false KEPT as a standing change (operator, internal WG net). Full endeavor, the reliable nh3-dev-egress + git-SSH mirror method, exact config state, smoke method, and next step → persistent-memory.d/2026-08-05-ci-flip-parked.md

  • [2026-08-05] Booth — 3 features shipped, live on :8090 + tagged. (1) verbatim-index.html booths get a floating top-right "‹ all booths" chip + inherited favicon, doctype/charset-safe byte-injection (booth-v0.1.5, 8577e7e); (2) .md renders + .txt/.log view in-booth without downloading via the /b/<n>/view route + a markdown dep + doc.html (booth-v0.1.6, 315faac); (3) prev/next arrows in the image zoom viewer — wrap-around + keyboard ←/→, hidden for single-image booths (booth-v0.1.7, c37a425). Canonical services/booth/; deploy = systemctl --user restart booth.service on nh3-dev (runs from the checkout's .venv; uv pip install new deps into it first); 47 tests. uv.lock gitignored (348c5c1).

  • [2026-08-05] worldtree herald re-nudge bug root-caused → forseti shipped althing-core v2.1.2 (d5d33df, deployed on nh3-dev). herald.py:363 rendered the wake command from the empty fresh mail set on the re-nudge path (should be deliver_msgs) → messages[0] IndexError → un-suppressed outer catch-all → 7s crash-loop for 9 days on worldtree-codex's pane route (mimir-dev surfaced it; I traced it from the editable source). Fix + render_command empty-guard + outer log-suppress + 3 tests + contract amendment, all forseti's. nh3-extdev herald 2.1.2 upgrade DEFERRED (operator, not-now): extdev is a WHEEL install (not editable), unexposed (no pane routes); the verified 2.1.2 wheel is staged on nh3-dev /tmp (sha256 003508…cef27) — uv tool install --force + restart both heralds when un-parked. extdev herald-unit provenance resolved (operator-authorized 2026-07-25 via forseti relay; recorded in this file's 07-25 herald-install entry). auto-memory reference_nh3_dev_althing_herald.

  • [2026-08-03] worldtree b168/#384/#385 arc COMPLETE — providers.yaml boot-gate pre-sync → b168 deploy → DCC+P&P re-ingest (705+667 concepts, 0 truncations, #385 budget fix validated vs April's 785 control) → #381 restart → operator-approved production dedup sweep (785 April orphans deleted from main, 4009→3224). Fiction wing 166→1,372 concepts; consumer verify 0/5→5/5→saturated. Three of MY foot-guns hardened into fleet runbook rules (mv -t, docker exec -u 1000, shared-containerd pull-race — see Tried-and-abandoned). Full runbooks (deploy-wt-config, Chroma-verify, config-delta pre-sync rule, #381, sweep) → persistent-memory.d/2026-08-03-worldtree-b168-384-385-arc.md

  • [2026-08-02] mimir-inbox / #377-read-path arc — deployed + 4 bugs found/fixed/verified + a cloned voice. mimir-inbox live on corviduo-dev:8091 (#377 write+read proven, live 8ece117); worldtree-dev #380 (wing-blind index) + #381 (stale-client restart) + #382 (intermittent Mimir grounding) chased and verified 3/3 by ratatoskr-dev; muninn-gate → dispatch 0.1.5; donut voice cloned from the 65-frost Booth bundle into the Zonos gateway; Zonos streaming confirmed already-working. Full arc, procedures, and lessons → persistent-memory.d/2026-08-02-mimir-inbox-arc.md

  • [2026-07-31] muninn-gate (#377 ingestion front door) BUILT + DEPLOYED + healthy on corviduo-dev:8090. First-boot acceptance passed (watcher:running:true proves ingestion_root byte-identity); submit path deferred to the mimir-inbox era. Full wiring (uid-1000, state-volume mount, staging path-agreement, BuildKit-secret build, deferred repoint + operational guards) → persistent-memory.d/2026-07-31-muninn-gate-deploy.md

  • [2026-07-31] worldtree-sdk 1.1.0 (Python) published to vh Gitea PyPI + a durable infra-ops publish cred. memory_context pass-through; unblocked wyrd-dev. claude-bot now a write-collaborator on vh/worldtree-sdk (source pulled via the Gitea API archive — git-HTTP 403s on that repo); publishing to the vh USER namespace can't be delegated (401 reqPackageAccess even with write:package) so it needs an owner token — operator saved a FULL vh site-admin token at ~/.config/gitea/vh-token (0600) for it (⚠️ high blast radius, kept over a scoped one; org-namespace migration is the only real de-personalization, parked by wtsdk-dev). auto-memory reference_infra_ops_vh_gitea_token_and_sdk_publish.

  • [2026-07-31] kimi-k3 "output cap" root-caused = a ~16384 REASONING-token ceiling, not an output cap. heid's cross-frontier panel was silently degraded (empty content, finish_reason: stop). dvalin+bil researched (docs said deprecated-max_tokens); heid's live data refuted that (completion hit 18455) → it's a reasoning ceiling. Proven on the wire against heid's real 500KB bundle: reasoning_effort: low drops reasoning under the ceiling → content returns, on BOTH coding + general endpoints. Fix is CALLER-side (no gateway change): send reasoning_effort via extra_body (LiteLLM drop_params: true strips the top-level param — why heid's earlier attempt no-op'd). Relayed to heid to validate; backstop = allowed_openai_params on the route. → persistent-memory.d/2026-07-31-kimi-k3-reasoning-cap.md

  • [2026-07-27] jackdaw-compose.service DECOMMISSIONED (jackdaw-dev request; the JackDAW AI Composer was cut from v1 by operator decision 2026-07-27). Stopped + disabled the nh3-dev :8787 user service (no client calls it — ai/server/AiChat deleted from main, /compose proxy removed); unit archived not deleted → ~/.config/systemd/user/jackdaw-compose.service.decommissioned-20260727 (revival = rename + daemon-reload). No credential revoked — the unit used the SHARED all-agents LiteLLM key (sk-eA_XOd…, model gen), not a dedicated one. Code preserved on jackdaw origin/ai-composer-preserved; treat as permanent. The :4500 HTTPS audition bench is untouched. (Supersedes the 2026-07-23 stand-up line below.)

  • [2026-07-25] bil-smithy-dev wired as an althing zellij-window-ping (pane route). She's a driver: human dwarf peer (pane bil-smithy already live alongside eitri/dvalin/regin-smithy in the Claude zellij session) but had no delivery route → smoke messages posted to the bus but never reached her window. Mechanism (reusable for any pane-route handle): ~/.althing/config.yaml → zellij_sessions.Claude.agents[] maps handle → target (a zellij pane TITLE, matched via list-panes -j in althing/zellij.py:resolve_pane_id) → command (herald write-chars + CR into that pane). The herald loads config ONCE at startup (herald.py main()), so systemctl --user restart althing-herald.service after editing. Added bil (target: bil-smithy), restarted, verified: herald delivered the pending smoke 01KYD7W7CF… (available→attempted→delivered). ⚠️ Noticed pre-existing pane-route errors on worldtree-codex + eitri-smithy-dev ("route-error: list index out of range", empty msg_ids — likely render_command messages[0] on an empty list; NOT caused by this change, bil works) — worth a herald look.

201 older entries archived to archival-memory.md.

Tried and abandoned

  • [2026-08-15] Grafted bf16 MTP loads UNINITIALIZED (0% accept) unless re:^mtp.* is in the quant-config ignore; and W4A16=Marlin (not native FP4) costs ~20% even on decode. Cost a premature 79 GB delete of a good model (declared desync-dead off the 0%). Lessons: test MTP on bf16 FIRST, isolate before deleting; modelopt 0.43 is dependency-hell for qwen3_5 (list-vs-dict quant_cfg + transformers conflict) — use llm-compressor. Full → persistent-memory.d/2026-08-15-uncensored-gen-seat.md

  • [2026-08-03] ComfyUI --enable-triton-backend on the irv-ml1 A6000 crashes EVERY render — Ampere has no hardware e4m3. adhoc-agent's operator-approved probe: comfy_kitchen's triton backend has a FUSED int8 matmul that would beat the eager backend's ~1.9x-slower unfused int8 path (21.3s vs 11.2s fp8 on the Moody Krea2 int8 checkpoints). Flipped it (added to COMFY_CMDLINE_EXTRA, recreated) → triton.compiler.errors.CompilationError: ValueError("type fp8e4nv not supported in this architecture. supported: fp8e4b15, fp8e5") in comfy_kitchen/backends/triton/quantization.py:145 dequantize_per_tensor_fp8, failing at node 5 CLIPTextEncode. Triton's fp8 dequant kernel targets fp8e4nv (Hopper/Ada e4m3); sm_86 Ampere (A6000) lacks hardware e4m3 → the JIT compile dies. With triton on it grabs the global --fp8_e4m3fn-text-enc dequant, so every render (fp8 AND int8) dies upstream at the text-encode step — the int8 UNet path never ran, so the convrot-coverage caveat wasn't even the limiter. Reverted cleanly (~15s to healthy, image unchanged sha256:94afb8ca, sage intact, prod restored). The parked cu130 rebuild won't fix it (e4m3 = hardware format, not CUDA version). DEFERRED to the Ada refresh (operator: "ada is coming, we'll optimize then" — Ada sm_89 has native e4m3, so triton's fp8 path should compile there). Mechanics: --enable-triton-backend is a compose environment: var, so toggling it needs docker compose up -d (recreate), NOT docker restart (reuses the baked env, no-ops silently). Full: auto-memory parked_triton_backend_ampere_fp8.

  • [2026-08-03] corviduo-dev shared containerd: a concurrent-pull race fails ONE instance's deploy; DON'T "prune to fix" — the image is in-use by the instance that won the race. b169 personal deploy failed at docker compose pull (Lchown … no such file or directory on the big torch layer → looked like disk pressure / corrupt snapshot). ACTUAL: NOT disk (56G free, inodes 7%). demo + personal + pinned share ONE /var/lib/containerd on corviduo-dev; demo (from main) and personal (from staging tag) extracted b169's shared torch layer simultaneously → personal's hit a partial snapshot mid-race and aborted while demo's completed. The image 6e34a87 was FULLY VALID — demo was RUNNING it healthy. Fix = just re-run the failed deploy (image already materialized; compose pull finds it present). NEAR-MISS: worldtree-dev's suggested "prune unused images/snapshots" would have rmi'd 6e34a87 = the image the running demo depends on → demo outage. Lesson: before any prune/rmi "cleanup," docker ps the running images — an "unused" image may be a co-tenant's live one; and verify the failure's REAL cause (disk? inode? in-use? race?) before applying the suggested remedy. (Pipeline fix, deferred: serialize demo-from-main + personal-from-staging, or a per-image pull lock, to avoid the shared-layer extraction race.)

  • [2026-08-02] docker exec into worldtree containers defaults to ROOT — root writes contaminate the uid-1000 (vh) KB tree. My sudo docker exec … --reindex on personal ran as ROOT (muninn app = uid 1000); its wing git-commit + atomic note-swap left root-owned files in the worldtree-personal_worldtree-kb volume: a root-owned .old-<job> backup dir (blocked the uid-1000 retry's rmtree → Errno 13, because unlink needs write on the DIR and it was root:root 755) AND 60 root-owned loose git objects in .git/objects/. Fix (host-side, corviduo-dev): sudo rm -rf the superseded .old- dir (tar'd aside to /tmp first) + sudo find … -user 0 -exec chown 1000:1000 the objects (ownership-only, git-content-safe; the .git/objects/XX/ dirs were vh-owned so these weren't a hard blocker, but violated "clean tree"). RUNBOOK RULE (worldtree-dev, ADOPTED): any docker exec into worldtree containers that WRITES pipeline state runs -u 1000, never default-root — same genus as the mv footgun (acting without matching the target's constraints; 3rd such slip in one session). GOTCHA that hid the scope: find … -user 0 | head -20 TRUNCATED (the .old- dir alone had 153 files, so the first page was all .old-) → I "verified clean" off a partial list. Never head a scope-defining find; count first (| wc -l). Related blind-spot (muninn-dev): a root-owned job SUBDIR passes every requeue guard (job_row/dispatch/list_jobs render fine) AND /health (contract's os.access(ingestion_root, W_OK) tests only the ROOT dir, so a foreign-owned subdir under pending/ still reports ingestion_root_writable: true) — then the uid-1000 gate can't write into it. "Clean board + green /health + failure at next mutation." muninn-dev added an OWNERSHIP column to the standing post-move check to catch it; two green signals both miss a foreign-owned subdir otherwise.

  • [2026-08-02] mv <job> complete/ → failed/ RENAMED the job to failed because failed/ didn't exist. worldtree-dev's round-2 unblock command (mv /data/state/ingestion/complete/<job> /data/state/ingestion/failed/) assumed failed/ existed; on PERSONAL muninn it did NOT (fresh instance — root was active/ complete/ pending/ sources/, no failed/). mv src nonexistent/ renames src→nonexistent, so job1 became the failed dir and job2 nested inside it. Caught on post-move ls (failed/ held job contents, not two subdirs), reconstructed via complete/ as watcher-safe scratch + rebuilt failed/ (worldtree:worldtree 755) — NO data loss. Lessons: (1) before mv X into-dir/, verify the dir EXISTS ([ -d dir ]) — an empty ls dir/ 2>/dev/null is AMBIGUOUS (missing vs empty), which was the preflight miss that let it through; (2) the correct guard is mv -t <targetdir> <src> (--target-directory): it refuses a MISSING target loudly (rc=1, "No such file or directory", nothing moved) — this is the house convention for queue/state moves now. TESTED by muninn-dev on coreutils 9.1: a trailing slash does NOT protect — mv src failed/ with failed/ missing STILL silently renames to failed (rc=0); "just add the slash" is a false guard. (mkdir -p failed/ first also works, but mv -t inverts the failure from silent-wrong to loud-safe in one flag.) Container sh is dash — no ( in echo strings. SILENT failure mode (muninn-dev carry-forward): a misplaced ingestion-state move doesn't crash anything — list_jobs() stays OK, loose files are inert; the ONLY symptom is the job quietly absent from the board (job_row→None, requeue→not_found/404, looks IDENTICAL to the original block). So after ANY state move, verify the job is actually ON THE BOARD (job_row found + guards pass), don't trust mv exit codes — and confirm job.dispatch.json survived (requeue refuses a dispatch-less job with the same not_requeueable symptom). Cross-checked + all-clear'd by muninn-dev, who correctly refused to mutate ingestion_root (INV-MG-1) and flagged instead. DON'T TIDY (round-2 pending): both DCC + P&P jobs currently REST in personal failed/ with manifests reading state: complete until round-2 requeue runs — deliberate + load-bearing (requeue keys on DIRECTORY PLACEMENT, not manifest state); looks wrong to anyone cold, leave it exactly as-is. Round-2 sequencing: the requeue is mimir-dev's browser flow (pending their operator's board-vs-API ruling); muninn-dev is the gate confirmer (runs the post-move board-check inside its custody — the right split, don't reach across INV-MG-1); infra-ops = the #381 restart after both jobs go terminal, then later the supervised main-collection sweep. Guard-verified HOLD LIFTED by muninn-dev 02:36Z. ARC COMPLETE (2026-08-03 ~05:49): both books terminal — DCC mimir-6351554e8e8f 705 concepts + P&P mimir-f3887c9b97b7 667, extracted AND indexed, 5/5 phases, 0 failures/truncations (validates the #385 budget fix vs April's 785 control); #381 restart-after-ingest FIRED (personal api, healthz/readyz 200 ~25s), retrieval-visibility confirmed (search_library returns DCC+P&P from fiction post-restart); handed ratatoskr-verify go to worldtree-dev. Delete-sweep precondition NOW MET — the stale DCC rows in main are genuine duplicates of live fiction rows, so worldtree-dev's supervised sweep of the ~785 April orphans is unblocked (still comes to me supervised: snapshot + operator-in-loop).

  • [2026-08-02] donut voice multi-clip reference (onyx-58 expansion) — TRIED, REVERTED. Folded the onyx-58 bundle's 3 Donut clips (seg101/seg110/seg148) in alongside the original seg000 → a 52.0s 4-take concat reference, hoping a longer ref → more robust speaker embedding. A pinned-seed A/B (5 pairs, varied registers, booth donut-onyx58) showed the original single-clip seg000 (16.3s) sounds better — concatenating disparate takes muddied the timbre more than the extra range helped. Reverted to seg000-alone (live + build-source). Two durable lessons: (1) for a faithful clone, a single clean representative take can beat a longer multi-take concat — more reference audio is NOT automatically better when the takes vary. (2) Emotion steering pulls the output AWAY from the cloned voice fast (operator's craft rule) — keep donut (and clones) emotion-neutral for fidelity; the gateway only enables emotion when an emotion_*/preset dial is explicitly sent, so bare {input,voice} calls stay pure-clone. seg148 was diarized SPEAKER_03 but IS Donut (operator-confirmed misdiarize). onyx-58 curated bundle lives in booth onyx-58 (24h TTL — stash to /mnt/smithy/voice_clones/ if a future middle-ref experiment is wanted).

  • [2026-08-02] Verifying the INDEX is not verifying GROUNDING (#382). A search_library returning wing=fiction hits proves the content is retrievable; it does NOT prove the agent (Mimir) trusts and uses those hits vs. silently answering from training. I reported "Mimir read Austen back to you" off a grounded-looking answer; ratatoskr-dev caught that grounding was intermittent (some sessions discarded the correct hits and substituted training knowledge). Test the harder claim — are the citations note-extracted or model-knowledge? — and reading the DEPLOYED artifact beats trusting the test for "is the fix live."

  • [2026-07-30] brokkr's WebSearch "verification" CONFIRMED a hallucination — 3 phantom microsoft/Mage-Flow-{Base,Turbo,Edit} repo IDs. brokkr-smithy-dev handed 3 gated-looking repo IDs for an operator-directed model pull; they don't exist (its own web-search fabricated an arXiv ID + project page, twice). Lesson: the HF registry API is ground truth — an unauth 401 ≠ exists ({"error":"Invalid username or password"} masks private/gated/nonexistent alike), an authed 404 = phantom, and author=X&search=Y refutes existence. API-verify every repo ID before a pull; LLM-summarized web fetches confabulate. auto-memory reference_verify_hf_repo_ids_before_pull.

  • [2026-07-30] magpie TTS serving — evaluated, ABANDONED. Pulled magpie_tts_multilingual_357m (the one real repo of brokkr's batch) to NFS, stood it up on irv-ml1 (ephemeral NeMo-Speech-main container — stock PyPI/NGC NeMo can't load v2607), A/B'd vs Zonos → Zonos wins expressive English decisively, multilingual not needed. Not served; magpie-nemo torn down. .nemo KEPT on NFS as brokkr's fine-tuning base. auto-memory project_magpie_tts_eval_rejected.

  • [2026-07-25] Chaining the althing wake-listener arm orphans it. reply && althing-wake-listener & (or spawning althing-wake-listener with & inside a run_in_background task) → the &-child reparents to init, UNTRACKED by the harness: no fire-notification, and re-arms bounce rc3 off a lock nothing services (mail silently unwatched). Compounding foot-gun: re-arming after a plain operator turn (not an actual fire) collides with the still-live prior listener (rc3). FIX: spawn althing-wake-listener as its OWN run_in_background task, and re-arm ONLY after a real fire (<task-notification> completed rc0). Reclaim an orphan with althing-cli stop-monitor then re-arm.

135 older entries archived to archival-memory.md.

Reference in New Issue View Git Blame Copy Permalink