3d4ee3818d
Last missing core piece of KB phase 4: kb-query no longer hard-fails /search when Ollama@SOLARIA is unreachable. app/fallback.py implements the plan's circuit-breaker exactly (30s cached health probe, 3s hard embed timeout on SOLARIA, one-shot same-request switch to a new local ollama-piha@PIHA container on timeout/error). sol_status in /healthz and /search now reflects the real breaker state instead of a hardcoded "up". New services/ollama-piha (bge-m3, OLLAMA_KEEP_ALIVE=0, arm64/no-GPU) is the local fallback leg. Live calibration on PIHA (2026-07-27, normal load): embed latency 4.2-5.2s, RAM peak ~983MiB against a 2.5GiB ceiling -- both inside the plan's go-bar, so the fallback is enabled by default rather than gated behind a flag. Calibration also surfaced and disabled (not removed) a previously-undocumented orphaned native ollama.service on PIHA that had been conflicting with the container's port. The embed-model invariant (query embedding == document_chunk.model) still enforces once at startup, since both fallback legs share one EMBED_MODEL constant by construction; a redundant per-request DB check was deliberately skipped and the invariant is instead proven structurally by test. retrieval_eval.py gains --transport http (plan §2 decision 6/§9), previously unimplemented. Verified live: HTTP transport is bit-identical to direct transport against the same live SOLARIA (0 mismatches), and a live sol-down simulation (kb-query's own OLLAMA_URL pointed at a dead address, no other Ollama consumer touched) shows the PIHA fallback answering with the same hit@3 gate outcome and dist within ~3e-4 of the SOLARIA baseline. Zero changes to DB schema or kb_retrieval's retrieval logic -- only the embed + health layer, per task constraints. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
33 lines
1.7 KiB
Plaintext
33 lines
1.7 KiB
Plaintext
# kb-query secrets + host-local binds — copy to .env (gitignored) next to
|
|
# docker-compose.yml and fill in real values. Never commit .env.
|
|
|
|
# LAN IP of PIHA. The published port (8230) binds ONLY to this interface —
|
|
# never 0.0.0.0. Verify after host rebuilds: ip -4 addr.
|
|
LAN_BIND_IP=192.168.31.5
|
|
|
|
# asyncpg DSN for kb-postgres@PIHA. kb-query runs in its own Docker network
|
|
# (separate compose project from kb-postgres), so it reaches kb-postgres's
|
|
# published port over the host's LAN interface, not "localhost" — same
|
|
# reasoning as paperless-worker@SOLARIA reaching paperless@PIHA over LAN.
|
|
KB_DSN=postgresql://kb:CHANGE-ME@192.168.31.5:5433/kb
|
|
|
|
# Ollama upstream (SOLARIA, over Tailscale MagicDNS — same trick as
|
|
# llm-gateway's OLLAMA_URL). Optional: defaults to this value if unset.
|
|
# OLLAMA_URL=http://solaria:11434
|
|
|
|
# Local fallback Ollama (ollama-piha@PIHA, plan §2 decision 2 / §5) — used only when SOLARIA
|
|
# is unreachable/times out. kb-query runs in its own Docker network (separate compose project
|
|
# from ollama-piha), so this must be PIHA's LAN IP + published port, not "localhost" — same
|
|
# reasoning as KB_DSN above. Optional: defaults to http://localhost:11434, which fails closed
|
|
# (nothing listens there in this container) until you set the real value below.
|
|
# OLLAMA_PIHA_URL=http://192.168.31.5:11434
|
|
|
|
# Embedding model kb-query enforces as a startup invariant (plan §2 decision
|
|
# 2) against document_chunk.model / document_summary.embedding_model.
|
|
# Optional: defaults to bge-m3.
|
|
# EMBED_MODEL=bge-m3
|
|
|
|
# document_summary.model kb-query's cascade path pre-filters on (the
|
|
# compilation track, plan §2 D3). Optional: defaults to claude-haiku-4-5.
|
|
# SUMMARY_MODEL=claude-haiku-4-5
|