The management room is end-to-end encrypted, but Draupnir ran with
experimentalRustCrypto: false and Pantalaimon off, so it could post there
(its own messages are unencrypted) but never saw commands: its last reply
to a command was ~57 days ago, and `!draupnir rooms add …` sent from Lotus
on 2026-10-13 was ignored. The same command sent unencrypted worked at
once (✅, joined dco-memes-minecraft).
Merging deploys via lxc110-draupnir.sh, which restarts draupnir.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017PCb59uY3UKqDqt7B6hzd5
- draupnir/production.yaml: Add health.healthz (port 8081) and web.abuseReporting
(port 8080) config — healthz was live on LXC but missing from repo; web server
enables Matrix client Report button forwarding to management room (Synapse module
install on LXC 151 still needed to complete the integration)
- README: Add Draupnir port map, abuse reporting setup docs, updated monitoring
section (3 new Prometheus scrape jobs, Draupnir Down alert, Grafana panel count),
add presence-disabled federation lag fix to performance checklist, document
Draupnir healthz/audit DB paths
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>