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>
34 lines
1.3 KiB
YAML
34 lines
1.3 KiB
YAML
service:
|
|
name: ollama-piha
|
|
owner_node: piha
|
|
role: kb-embed-fallback # module 5 phase 4 plan §2 decision 2 / §5: local fallback embed
|
|
# when Ollama@SOLARIA is unreachable/times out. NOT the
|
|
# default embed path -- kb-query only reaches this via its
|
|
# circuit breaker (services/kb-query/app/fallback.py).
|
|
exposure: private # LAN bind (LAN_BIND_IP), consumed only by kb-query@PIHA today
|
|
dependencies: []
|
|
ports:
|
|
- container: 11434
|
|
host: 11434
|
|
protocol: tcp
|
|
healthcheck:
|
|
type: http
|
|
endpoint: http://localhost:11434/api/tags
|
|
interval: 1m
|
|
timeout: 10s
|
|
retries: 3
|
|
restart_policy: unless-stopped
|
|
persistence:
|
|
paths:
|
|
- /opt/homelab/data/ollama-piha
|
|
runtime:
|
|
directories:
|
|
- /opt/homelab/data/ollama-piha
|
|
env_vars:
|
|
- LAN_BIND_IP # required — compose port-bind interpolation
|
|
config_files:
|
|
- .env # LAN_BIND_IP (gitignored, from env.example)
|
|
# Same model as document_chunk.model / document_summary.embedding_model (bge-m3) -- kb-query's
|
|
# startup invariant (app/startup.py) enforces this once for both the SOLARIA and PIHA legs.
|
|
# `ollama pull bge-m3` is a manual deploy step (see README.md), not baked into the image.
|