Running SeAT 5 in Docker: every problem I hit and how I fixed it

I run SeAT, the EVE Online corporation management tool, for our corp in Docker on a Proxmox LXC, managed with Dockge. Getting it from SeAT 4 to SeAT 5 and then keeping it healthy turned into a long series of problems. This post collects all of them in one place, with the fixes that actually worked, so the next person (or future me) doesn’t have to dig through logs again.

The stack is a split compose setup, so every command needs the full -f chain:

Services are front, worker, scheduler, cache (Redis) and mariadb.

Read more

Network Monitoring with Prometheus & SNMP Exporter in the Homelab

Why SNMP Exporter instead of LibreNMS?

Anyone who’s done network monitoring before knows the usual suspects: Nagios, Zabbix, LibreNMS. These tools come with everything — SNMP discovery, device inventory, maps, alerting — but also their own database, their own visualization layer, and their own operational overhead.

In a homelab that already runs Prometheus, Grafana, and Loki, there’s little reason to add LibreNMS on top. The Prometheus SNMP Exporter fits seamlessly into an existing stack: it’s a single container that accepts SNMP queries and returns them as Prometheus metrics.

In this post I’ll walk through how I’m monitoring OPNsense, a Ruckus R720 AP, managed switches, Proxmox nodes, and a Brother printer with it.


Preparing devices: enabling SNMP

OPNsense

Go to Services → SNMP: enable the service, set a community string (e.g. homelab), bind to the LAN interface.

Ruckus R720

In the web UI under System → SNMP: enable SNMPv2c, enter the community string.

Proxmox nodes

On each node in the shell:

Edit /etc/snmp/snmpd.conf:

Managed switch & Brother printer

Varies by vendor — enable SNMP in the management UI and set the community string. On the Brother DCP-L8410CDW the option is under Network → Protocol → SNMP.


Docker Compose (Dockge)

Place snmp.yml in the same directory as the compose file.


snmp.yml — configuration

The official snmp.yml from the repository ships with ready-made modules for hundreds of device types and is the recommended starting point:

For a leaner custom config (compatible with snmp_exporter v0.26+):

Note on the Brother page counter: OIDs under 1.3.6.1.4.1.2435 are model-specific. The best way to find the right one for your printer is to run snmpwalk -v2c -c homelab <IP> 1.3.6.1.4.1.2435 and look for a counter that increments with each print job. The standard Printer-MIB (1.3.6.1.2.1.43) is model-independent and reliably covers toner levels.


Prometheus scrape config


Testing it

Once the container is running, query a device directly to verify everything works:

You should see Prometheus-formatted metrics. An empty response usually means a firewall issue — UDP 161 needs to be open from the exporter host to the target devices.


Grafana dashboards

Ready-made dashboards on grafana.com that work with these metrics out of the box:

  • 11169 — SNMP interface monitoring (generic)
  • 21491 — OPNsense
  • 10347 — Proxmox (SNMP)

For the printer, a simple hand-built dashboard works well — toner colors come through as labels directly from the SNMP walk.


Wrap-up

The SNMP Exporter is the leanest way to bring network devices into an existing Prometheus/Grafana stack. No extra database backend, no separate web UI — just one exporter container and a few lines of config. If you already have the stack running, it’s a much cleaner solution than standing up LibreNMS or Zabbix alongside it.

Ubuntu 24.04 Server Quick Start: Sudo User, SSH Hardening, fail2ban, Docker and Dockge

Every time I spin up a new Ubuntu server I go through the same handful of steps before anything else touches it. This post is that checklist, written down so I (and hopefully you) don’t have to piece it together again. It assumes a fresh Ubuntu 24.04 LTS install where your SSH public key is already in place for the initial login.

Throughout the post I use youruser as the new account name and 2222 as the new SSH port. Replace both with your own values.

Golden rule: keep your current SSH session open the whole time and test every change from a second terminal. That way a typo never locks you out.

Read more

Rome / Milan CPUs

Processor Model Codename Cores/Threads Base Frequency Max Boost Frequency TDP (W) L3 Cache (MB) Approximate Used Price (USD)
EPYC 7F32 Rome 8/16 3.7 GHz 3.9 GHz 180 128 $198
EPYC 7F52 Rome 16/32 3.5 GHz 3.9 GHz 240 256 $315
EPYC 7F72 Rome 24/48 3.2 GHz 3.7 GHz 240 192 $469
EPYC 7302P Rome 16/32 3.0 GHz 3.3 GHz 155 128 $149
EPYC 7402P Rome 24/48 2.8 GHz 3.35 GHz 180 128 $192
EPYC 7502P Rome 32/64 2.5 GHz 3.35 GHz 180 128 $210
EPYC 7702P Rome 64/128 2.0 GHz 3.35 GHz 200 256 $695
EPYC 7742 Rome 64/128 2.25 GHz 3.4 GHz 225 256 $749
EPYC 7313P Milan 16/32 3.0 GHz 3.7 GHz 155 128 $850
EPYC 7443P Milan 24/48 2.85 GHz 4.0 GHz 200 128 $1,337
EPYC 7502P Milan 32/64 2.5 GHz 3.35 GHz 180 128 $1,575
EPYC 7702P Milan 64/128 2.0 GHz 3.35 GHz 200 256 $1,995
EPYC 7763 Milan 64/128 2.45 GHz 3.5 GHz 280 256 $7,890
EPYC 73F3 Milan 16/32 3.5 GHz 4.0 GHz 240 256 $1,688.88
EPYC 74F3 Milan 24/48 3.2 GHz 4.0 GHz 240 256 $2,268
EPYC 7C13 Milan 64/128 2.0 GHz — 240 256 $818.80

Proxmox – GPU Passthrough / Plex LXC

Gute Anleitung hier, jedoch funktionierte es noch nicht.

Fehler gab es bei den Zahlen in der cgroup2 config.

Aktuelle Version (06.01.2025) – https://download.nvidia.com/XFree86/Linux-x86_64/550.142/

Weitere URL für das Archiv – https://launchpad.net/~graphics-drivers/+archive/ubuntu/ppa

URL des Proxmox Helper Scripts – https://community-scripts.github.io/ProxmoxVE/scripts?id=plex

Read more

Proxmox – error bei Anlage eines neuen LXC Container: unable to create session key: disk quota exceeded: unknown

Viele Container bereits deployed. Docker ließ sich nicht starten.

Edit: detailliertere Lösung: https://gkellynavarro.com/blog/2020/04/04/container-issues/

Restoring a 2017 Redmine backup into Redmine 5.0.2 (Docker)

Update 2026: this setup has since been upgraded to Redmine 6.1 on PostgreSQL 15, and the permission workaround below is no longer needed. See Upgrading a neglected Redmine: PostgreSQL 9.6 → 15 and Redmine 5.0 → 6.1 in Docker.

I had an old Redmine backup from 2017 lying around and wanted it running again on a current system. This is a quick record of how I restored it with the sameersbn Docker images, and the permission problem that kept getting in the way.

Read more