alerting is handled by grafana, not alertmanager. removed the
stale reverse proxy block from caddyfile template and updated
caddy + prometheus docs to reflect grafana-only alerting.
- New zfs role with cron-based scrub scheduling for Linux and FreeBSD
- Weekly Sunday scrubs at noon (matching existing manual crons)
- Add zfs_hosts inventory group with london-a and london-b
- Configure zfs_pools per host: zroot (london-a), hdd (london-b)
- Add Prometheus alert rules for degraded/faulted/offline pools
- Add zfs.yml playbook for targeted deploys
Captures the previously untracked scrub cron on london-a and
re-enables the commented-out scrub on london-b.
Refs: PESO-93
* Configure UFW firewall rules in common Ansible role
Add UFW configuration to the common role for Debian hosts:
- Default deny incoming, allow outgoing
- Allow all traffic on tailscale0 interface (mesh comms)
- Allow SSH port 22 as safety net
- Per-host allowed ports via ufw_allowed_ports variable
- Enable UFW after rules are applied
helsinki-a gets ports 80/443 for reverse proxy traffic.
Other Debian hosts only need Tailscale + SSH.
Closes PESO-79
* Remove unused alerting and rule_files from prometheus.yml
Alerting is handled by Grafana, not Prometheus Alertmanager.
The empty alertmanagers and rule_files sections were just noise.
Resolves PESO-74
Replace local network IPs (192.168.1.x) with Tailscale IPs for
london-a and london-b in all scrape configs. This ensures consistent
connectivity via Tailscale mesh regardless of network topology changes.
Refs: PESO-80
