====== Netbox - ip.hackeriet.no ======

===== Deployment =====

  * Deployed in Docker
  * Lives on [[infra:hosts:app-01|app-01]] (''/root/services/netbox-docker'') in the klynge001 cluster
  * Is deployed with plugins (custom Docker image setup)
    * [[https://github.com/netbox-community/netbox-docker/wiki/Using-Netbox-Plugins|More info]] at Netbox Github
  * Exposed through [[infra:hosts:ingress|ingress.hackeriet.no]]
    * Docker compose -> nginx on app-01 -ssl-> nginx on ingress -ssl-> Internet

==== Update ====

<code>
docker compose build --no-cache
docker compose up -d
</code>

===== Troubleshooting =====

==== 502 Bad Gateway from ip.hackeriet.no ====

A 502 from ''https://ip.hackeriet.no/'' means nginx is reachable, but the NetBox upstream is not answering correctly.

On ''app-01'':

<code>
cd /root/services/netbox-docker
docker compose ps
docker compose logs --tail=120 netbox postgres redis redis-cache netbox-worker
df -h /storage /var/lib/docker
</code>

If the containers are confused after a restart, a full compose stop/start has worked better than restarting selected containers:

<code>
cd /root/services/netbox-docker
docker compose down
docker compose up -d
</code>

NetBox can take a little while to become healthy after ''docker compose up -d'', especially while Postgres and the NetBox container settle. Do not treat the 502 as final immediately after starting the stack; watch the compose status and retry the service after a short wait.

Useful checks:

<code>
docker compose ps
curl -v http://127.0.0.1:8000/
curl -v http://127.0.0.1:8000/api/status/
sudo tail -n 100 /var/log/nginx/error.log
</code>

Interpretation:

  * If ''http://127.0.0.1:8000/'' fails locally on app-01, focus on the NetBox Docker stack.
  * If the local app-01 check works but ''https://ip.hackeriet.no/'' still returns 502, focus on nginx or the path through [[infra:hosts:ingress|ingress]].
  * If Postgres logs mention ''No space left on device'', free space before restarting the stack.