After upgrading the Mac client 0.68.3 -> 0.71.3 (matching mgmt) and clearing
its NetBird resolver cache (netbird service restart), the managed-zone
deletion works: with 0 managed-zone records, the Mac resolves all
hubris.network names by forwarding to Technitium (192.168.8.2). iPhone
confirmed on cellular (no LAN path -> proves mesh-forward).
The first deletion "failure" was a misdiagnosis: the old 0.68.3 resolver
cache held stale answers and wouldn't clear on down/up (needs daemon
restart); the Mac's dual LAN+mesh paths muddied it. A direct
dig @100.122.255.254 of an unsynced name had shown forwarding working.
Done:
- Deleted all 23 NetBird managed-zone A-records.
- Removed the */10 dns-sync cron. Kept /opt/dns-sync/sync.py + token +
pre-deletion backup as an emergency-restore tool only.
End state: Technitium is the single DNS source. Mesh peers forward to it
(Core route -> 192.168.8.0/24); LAN/household query it directly. No replica,
no sync. Requires mesh clients on 0.71.x+.
Docs: dns.md + 107-dns.md updated to single-source; subdomain recipe no
longer references the sync.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Deleting the NetBird managed-zone replica broke mesh-peer DNS: the Mac
(NetBird 0.68.3) could not resolve hubris.network via the home-lab-dns
nameserver group (-> 192.168.8.2) even though it shows "Available" and the
192.168.8.0/24 route is present. Forwarding to the routed-LAN IP does not
actually serve queries on the current client. Restored the managed zone via
dns-sync.py and re-enabled the cron; resolution recovered.
Correction to Phase 2: the route fix delivered roaming-peer *service
connectivity* (the real iPhone win) but did NOT enable DNS forwarding. The
original "NetBird won't forward to Technitium for mesh peers" finding
stands; managed zone + sync are retained as load-bearing.
To finish single-source later: upgrade clients to 0.71.x, or point the
nameserver group at a mesh-native DNS IP (join CT 107 to the mesh).
Docs: dns.md + 107-dns.md corrected to reflect retained managed zone.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Root cause of "NetBird won't forward to Technitium" was NOT a nameserver
bug — it was a missing route. The home-lab-dns nameserver group
(-> 192.168.8.2, domain hubris.network) was applied to all peers, but the
192.168.8.0/24 route (home-lab-network resource) was distributed only to
the Services group. Roaming peers (Core: iphone + laptops) had no route to
192.168.8.2, so forwarding silently failed (Networks: -).
Fix: added Core to the home-lab-network resource distribution via the
NetBird API. Roaming peers now get the subnet route + the already-applied
nameserver forwarding -> *.hubris.network resolves off-LAN. Also grants
roaming devices full homelab service access. No CT 107 mesh-join needed
(original Phase 2 hypothesis obsolete).
Docs: corrected dns.md + 107-dns.md root-cause claims. Managed zone kept
as fallback pending Phase 4.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 1 of the DNS-redundancy cleanup (keep NetBird, collapse toward one
zone) — the safe, mesh-independent half:
- Every homelab LXC now resolves via Technitium (192.168.8.2). Fixed 8
boxes on a dead resolver (.180), the router (.1), or Tailscale MagicDNS
(100.100.100.100): 101,102,104,105,106,114,119,126.
- Removed the redundant /etc/hosts auth/mcp/secrets overrides (Technitium
returns identical-or-better answers); disabled hubris-hosts-override.
- Net effect: on-prem DNS (LXCs + household via Fritz!Box->Technitium) is
now NetBird-independent, so dropping the managed zone later can't break
on-LAN resolution. Phases 2-4 still pending.
Tailscale decommissioned (was legacy/being-phased-out):
- Removed from the 6 LXCs still running it (101,103,104,105,114,119):
logout, disable tailscaled, apt purge, state cleared.
- inventory.yaml: dropped tailscale from accepted + all mesh blocks;
regenerated hosts/*.yaml (also pruned orphan authentik/claudio-bot).
- Tightened secrets-issuance MESH_SUBNETS: removed the now-vestigial
Tailscale CGNAT range 100.64.0.0/10.
- Updated narrative docs (mesh, dns, network, README, AGENTS,
agent-enrollment, homelab-context, 105-apps, 107-dns).
Live infra changed on the fleet + Mac; this commit records the docs/inventory.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Caddyfile updated on caddy LXC with new backend IPs
- inventory.yaml updated to reflect current DHCP assignments
- hosts/{paperless,haos}.yaml regenerated from inventory
- New investigation doc: 2026-06-03-moonlight-sunshine-wifi-jitter.md
- Updated investigations/index.md with link and status
- Added .gitignore for .DS_Store
- Saved Hermes planning docs from recent sessions
- scripts/dns-sync.py: reconcile Technitium named A-records -> NetBird managed
zone via API (cron */10 on dns LXC 107). Single authoring source; kills the
manual drift behind the auth/sso/nfs-export saga.
- secrets/netbird-pat.yaml: sops-encrypted NetBird API PAT for the sync.
- dns.md / 107-dns.md: document the sync model + why forward-to-Technitium was
abandoned (NetBird self-IP / nameserver-group quirks).
- Cleanup: removed inert Mac secondary; reverted primary AXFR; home-lab-dns ->
[192.168.8.2] (1/1 Available); deleted vestigial Proxmox Names group.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
hosts/hubris.md:
- Update At a glance network section: vmbr1 uplink (192.168.178.10),
vmbr0 portless internal bridge with 192.168.8.1 alias
- Remove Phase 1 WiFi failover section (wlp3s0 disabled 2026-06-02)
- Changelog: Slate AX retired, SODOLA added, Proxmox as subnet router
containers/121-caddy.md:
- Changelog: caddy.service unit was missing from hubris1 package,
recreated manually; risk of loss on package reinstall noted
containers/107-dns.md:
- Update Who points here: Technitium DHCP hands out .2 as DNS for
homelab clients; Fritz!Box LAN clients still get Fritz!Box DNS
- Add DHCP section documenting the homelab scope (100-240, gw .1)
- Changelog: DHCP enabled 2026-06-02, replaces Slate AX DHCP
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- plans/: mark Slate AX → SODOLA migration as Done (2026-06-02)
- infrastructure/network.md: rewrite to reflect actual topology —
Fritz!OS 8.x lacks second-IP-network support, so Proxmox routes between
vmbr1 (192.168.178.10, uplink) and vmbr0 (192.168.8.x, internal portless
bridge with 192.168.8.1 alias). Static route on Fritz!Box, no NAT on Proxmox.
Technitium DHCP enabled for 192.168.8.100–240 (gateway .1, DNS .2).
Caddy service unit was missing on CT 121 — recreated and enabled.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- tools/setup-hermes-soul.sh: new auto-setup script (no-op on non-Hermes)
- HERMES.md: canonical persona for all agent types, agent-type table
- AGENTS.md: auto-setup section lists both caveman and hermes-soul
- post-pull.sh runs it automatically on every sync
