0.6.6 - Operations hardening (R2)
Added
- Prometheus metrics at
/metrics. Request latency (by route template),
terminal ingestion-job counts, and live printer status by provider are now
exported in Prometheus text format for Grafana/Prometheus scraping. The
endpoint is open by default on the trusted internal network; set
VAULT_METRICS_TOKENto require a static bearer token. - Richer
/api/v1/healthoutput. The health probe now reports background
ingestion-job counts and external-library scan status alongside the existing
database, storage, backup, and printer-provider sections. - Unraid Community Applications support. Added CA-ready repository metadata
(ca_profile.xml, rooticon.svg) and Docker templates undertemplates/
(printstash-api.xml,printstash-frontend.xml) plus a step-by-step
unraid/README.mdfor installing PrintStash on Unraid (create the
printstashnetwork, install the API then the frontend, finish in the setup
wizard).
Fixed
- External library scans no longer get stranded by a restart. If the backend
was restarted mid-scan, the library stayed markedrunningforever and the
scheduler skipped it indefinitely. Orphaned scans are now reset toerrorat
startup so they are picked up again on the next scheduled tick.