Proxmox Rename: prox → grimoire¶
Renaming the Proxmox hypervisor from prox.sqrd.link to grimoire.sqrd.link to match the homelab naming scheme (Nostradamus-themed). The grimoire is the book that summons things into existence — fitting for the hypervisor that runs everything else.
⚠️ This is not a casual hostname change. Proxmox bakes the node name into its cluster config and certificate paths. Do this deliberately, with backups verified first.
Pre-flight checklist¶
- Verify PBS backups are current (check
almanacshare on nastradamus) - Note all running VMs and LXCs — have a list ready
- Schedule a maintenance window (all VMs/LXCs will stay running, but Proxmox UI will restart)
- Take a snapshot of the Proxmox host config if possible
Step 1 — Update /etc/hostname¶
Step 2 — Update /etc/hosts¶
Edit /etc/hosts and replace every occurrence of prox with grimoire:
Change:
To:Step 3 — Rename the Proxmox node¶
This is the critical part. Proxmox stores node identity in /etc/pve/.
# Stop the pve-cluster and corosync services
systemctl stop pve-cluster corosync
# Mount the cluster filesystem in local mode
pmxcfs -l
# Rename all references from 'prox' to 'grimoire'
find /etc/pve -type f | xargs grep -l "prox" | while read f; do
sed -i 's/prox/grimoire/g' "$f"
done
# Rename the node directory itself
mv /etc/pve/nodes/prox /etc/pve/nodes/grimoire
# Kill pmxcfs and restart services normally
killall pmxcfs
systemctl start pve-cluster corosync pvedaemon pveproxy pvestatd
Verify: After restart, the Proxmox web UI should show
grimoireas the node name.
Step 4 — Regenerate certificates¶
The old TLS certs will reference prox. Regenerate them:
Or via the UI: Datacenter → grimoire → Certificates → Regenerate
Step 5 — Update AdGuard DNS¶
In AdGuard Home (adguard.sqrd.link):
- Remove DNS rewrite:
prox.sqrd.link→10.10.100.8 - Add DNS rewrite:
grimoire.sqrd.link→10.10.100.8
Both AdGuard instances (.1 and .2) — adguard-sync should propagate automatically, but verify.
Step 6 — Update Netbox¶
In Netbox (netbox.sqrd.link):
- Rename the device
prox→grimoire - Update the primary FQDN to
grimoire.sqrd.link - Check any tags or config context that reference the old name
Step 7 — Update Ansible / Semaphore¶
- Check
ansible/netbox.yml— if inventory is pulled dynamically from Netbox, this updates automatically once Netbox is updated - Check any hardcoded references to
proxin playbooks underansible/playbooks/ - Update Semaphore task templates if any reference
proxdirectly
Step 8 — Update the_codex repo¶
Once the monorepo restructure is done:
- Ensure the host folder is named
hosts/grimoire/(nothosts/prox/) - Update README references
Step 9 — Update Proxmox Backup Server¶
PBS (10.10.100.7) may have storage/backup jobs referencing the node name:
- Log into PBS UI at
10.10.100.7:8007 - Check Datastore config and backup jobs for any
proxreferences - Update if needed
Step 10 — Final verification¶
# Confirm hostname
hostname
# → grimoire
# Confirm Proxmox sees itself correctly
pvesh get /nodes
# → should list 'grimoire'
# Check certificate CN
openssl s_client -connect grimoire.sqrd.link:8006 2>/dev/null | openssl x509 -noout -subject
Also verify:
- [ ] Proxmox UI accessible at grimoire.sqrd.link:8006
- [ ] All VMs and LXCs still listed and running
- [ ] PBS backup jobs still functioning
- [ ] prox.sqrd.link no longer resolves (or resolves to the same IP — either is fine, just don't leave stale DNS pointing somewhere wrong)
Rollback¶
If something goes wrong before step 3, just revert /etc/hostname and /etc/hosts and reboot — no harm done.
After step 3, rollback means repeating the process in reverse. The node rename is the point of no return — make sure backups are verified before crossing it.
Related¶
- [[Docker Host Migratie Plan LXC naar VM]]
- Netbox:
grimoire.sqrd.link - Proxmox UI:
https://grimoire.sqrd.link:8006