Three permanent test hosts for validating deployment and bootstrapping
workflow. Each host configured with:
- Static IP (10.69.13.20-22/24)
- Vault AppRole integration
- Bootstrap from deploy-test-hosts branch
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Remove host entries from terraform/vault/approle.tf on --remove
- Detect and warn about secrets in terraform/vault/secrets.tf
- Include vault kv delete commands in removal instructions
- Update check_entries_exist to return approle status
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
The regex patterns expected 6 spaces of indentation but flake.nix uses
8 spaces for host entries. Also updated generated entry template to
match current flake.nix style (using commonModules ++).
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Update homelab-deploy input to get metrics support
- Enable metrics endpoint on port 9972
- Add scrape target for prometheus auto-discovery
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Compare VictoriaMetrics and Thanos as options for extending
metrics retention beyond 30 days while managing disk usage.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add homelab.deploy.enable option (requires vault.enable)
- Create shared homelab-deploy Vault policy for all hosts
- Enable homelab.deploy on all vault-enabled hosts
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add system/homelab-deploy.nix module that automatically enables the
listener on all hosts with vault.enable=true. Uses homelab.host.tier
and homelab.host.role for NATS subject subscriptions.
- Add homelab-deploy access to all host AppRole policies
- Remove manual listener config from vaulttest01 (now handled by system module)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Update homelab-deploy to include bugfix. Add CLI to devShell for
easier testing and deployment operations.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Ensure homelab-deploy-listener waits for the NKey secret to be
fetched from Vault before starting.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add homelab-deploy flake input and NixOS module for message-based
deployments across the fleet. Configure DEPLOY account in NATS with
tiered access control (listener, test-deployer, admin-deployer).
Enable listener on vaulttest01 as initial test host.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add a shared `homelab.host` module that provides host metadata for
multiple consumers:
- tier: deployment tier (test/prod) for future homelab-deploy service
- priority: alerting priority (high/low) for Prometheus label filtering
- role: primary role of the host (dns, database, monitoring, etc.)
- labels: free-form labels for additional metadata
Host configurations updated with appropriate values:
- ns1, ns2: role=dns with dns_role labels
- nix-cache01: priority=low, role=build-host
- vault01: role=vault
- jump: role=bastion
- template, template2, testvm01, vaulttest01: tier=test, priority=low
The module is now imported via commonModules in flake.nix, making it
available to all hosts including minimal configurations like template2.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
MCP exposes two tools:
- deploy: test-tier only, always available
- deploy_admin: all tiers, requires --enable-admin flag
Three security layers: CLI flag, NATS authz, Claude Code permissions.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Support deploying to all hosts in a tier or all hosts with a role:
- deploy.<tier>.all - broadcast to all hosts in tier
- deploy.<tier>.role.<role> - broadcast to hosts with matching role
MCP can deploy to all test hosts at once, admin can deploy to any group.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add plan for NATS-based deployment service (homelab-deploy) that enables
on-demand NixOS configuration updates via messaging. Features tiered
permissions (test/prod) enforced at NATS layer.
Update prometheus-scrape-target-labels plan to share the homelab.host
module for host metadata (tier, priority, role, labels) - single source
of truth for both deployment tiers and prometheus labels.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Increase duration from 5m to 10m and demote severity from critical to
warning. Brief degraded states during nixos-rebuild are normal and were
causing false positive alerts.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Merge Prometheus metrics and Loki logs into a unified troubleshooting
skill. Adds LogQL query patterns, label reference, and common service
units for log searching.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Reference guide for exploring Prometheus metrics when troubleshooting
homelab issues, including the new nixos_flake_info metrics.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Update nixos-exporter to 0.2.3
- Set system.configurationRevision for all hosts so the exporter
can report the flake's git revision
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add nixos-exporter prometheus exporter to track NixOS generation metrics
and flake revision status across all hosts.
Changes:
- Add nixos-exporter flake input
- Add commonModules list in flake.nix for modules shared by all hosts
- Enable nixos-exporter in system/monitoring/metrics.nix
- Configure Prometheus to scrape nixos-exporter on all hosts
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
RemainAfterExit=true kept the service in "active" state, which
prevented OnUnitActiveSec from scheduling new triggers since there
was no new "activation" event. Removing it allows the service to
properly go inactive, enabling the timer to reschedule correctly.
Also fix ExecStart to use lib.getExe for proper path resolution
with writeShellApplication.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add proposed dns_role label to distinguish primary/secondary DNS
resolvers. This addresses the unbound_low_cache_hit_ratio alert
firing on ns2, which has a cold cache due to low traffic.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Backups to the shared restic repository were all scheduled at exactly
midnight, causing lock conflicts. Adding RandomizedDelaySec spreads
them out over a 2-hour window to prevent simultaneous access.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
The homeassistant override key should match the entity type in the
MQTT discovery topic path. For battery sensors, the topic is
homeassistant/sensor/<device>/battery/config, so the key should be
"battery" not "sensor_battery".
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Displays FQDN and flake commit hash with timestamp on login.
Templates can override with their own MOTD via mkDefault.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
