zenodotus280

SYSLOG 25-W08

Network troubleshooting; Jackett is not really needed; Naming conventions for automations requires some planning


Network Troubleshooting for KRONOS

My previous method of keeping kronos up to date was to initiate a syncoid task from CAIRO. Since CAIRO is no longer active I now need to initiate from kronos directly. This is preferred anyway since backups should be pulled. And this allows it to be fully autonomous by using the syncoid script as part of a boot>backup>shutdown process.

I was unable to connect to the router from KRONOS. I ran a slew of commands to no avail despite SSH-ing from my desktop.

  • Ping the router: I was unable to ping the router (10.10.10.1), but I could ping other devices on the LAN (e.g., 10.10.10.201). bash ping 10.10.10.1

  • Review the routing table: The routing table appeared fine, with the default route pointing to the router (10.10.10.1) via vmbr0. bash ip route show

  • Check ARP entry: Found that the router's ARP entry was marked as DELAY, indicating your system was waiting for an ARP response from the router. bash ip neigh show

  • Flush the ARP cache and attempted a ping to the router again, but the ARP entry then showed as INCOMPLETE and the ping remained unreachable. bash ip neigh flush all ping 10.10.10.1

  • Verify network interface status: Confirmed that the vmbr0 interface was up and had no errors or dropped packets. bash ip link show vmbr0 ip -s link show vmbr0

  • Test different port or cable to rule out a faulty port or connection.

That was it... And I don't even know why. Lesson learned... try a different port earlier in the troubleshooting process! It's a quick and easy check.

Revisiting My NAS Upgrade Plans

I returned to thinking about the future of AVALON, my virtualized FreeBSD NAS. I regularly see about 50% CPU load when saturating a 1 Gbps link to the NAS. If I ever upgrade to 2.5 Gbps or 10 Gbps, the drives (and CPU) become potential bottlenecks. My plan is to move to a 10G card eventually, letting me hook up the NAS, the backup server (KRONOS or a future node), and my desktop all at once. This would preserve the full throughput of an SSD-based pool if I incorporate one. But that's a project for another day/year.

Jackett and FlareSolverr for Torrents

Most of my Usenet-based "arr" stack is stable (SABnzbd + Sonarr/Radarr + Lidarr), but I also like adding torrent options for certain private trackers. That's where Jackett comes in. Jackett acts as a translator or "proxy" between torrent indexers and the typical "arr" apps. You can connect all sorts of trackers to it and feed them into Radarr/Sonarr in a uniform format.

Jackett Setup:

  • Created /srv/jackett/{config,blackhole} directories.
  • Used a Docker Compose file for linuxserver/jackett:latest.
  • Once it spun up on port 9117, I could access the Jackett UI at jackett.internal:9117---super straightforward.

FlareSolverr is sometimes required for trackers that use Cloudflare's anti-bot pages. I spun up a second container for it (ghcr.io/flaresolverr/flaresolverr:latest), then told Jackett to route certain trackers through flaresolverr.internal:8191. That way, if a tracker uses tricky captcha or browser checks, FlareSolverr attempts to solve them behind the scenes. I had a few timeouts and it never really worked right so I just gave up on it and skipped trackers that require it.

Movie Naming Conventions in Radarr

Up until now, I used TinyMediaManager for local re-tagging. But once you trust Radarr to upgrade or rename your library, you might as well let it handle everything.

A couple things worth remembering: - If you lose details like release group, editions, or "quality", Radarr might re-download something you already had. - Identifying "Director's Cut", "Theatrical", or "Extended" versions becomes much more difficult if you strip that info from the filenames.

Jellyfin Compatibility:

  • Some guides recommend embedding the TMDB ID in folder names, like {Movie CleanTitle} ({Release Year}) [tmdbid-{TmdbId}]. Jellyfin can do extremely accurate matching that way.
  • For the actual file names, you can append [Quality Full], [MediaInfo VideoDynamicRangeType], or [AudioChannels]. That's optional but helpful if you want to see at a glance which encode you have.

I settled on a moderate approach: keep enough file- and folder-level info to avoid re-downloading the same release (especially if it's a certain group or a repack). This also helps me keep track of HDR vs. SDR. TinyMediaManager is no longer necessary for manual downloads either since those can be imported and processed in Sonarr or Radarr.

Thoughts? Leave a comment