MowMdown

The GPU throttles after the transcode buffer is filled. I can't remember what the default setting is but it's either 60 or 120 seconds. Once that buffer is met, the transcoder throttles and doesn't need to encode/decode but in small segments. When you start seeking, the buffer is nuked and has to rebuild. Based on the screenshots, everything looks normal. At the very start of a transcode session you will see the Encoder/Decoder stats climb up but after the buffer is full, it will drop to 1% or 0% until more buffer is needed. Pay attention to the throttle state: P0 is most power <---> P8 is idle. During Transcode Buffer Filling: Buffer is full:

Say your Local unraid IP is 10.0.0.2, tailscale will forward any requests for Vaultwarden to 10.0.0.2, since 10.0.0.2 is a local IP it will not traverse the WAN to contact 10.0.0.2. That would be a waste of resources. When you are not on you LAN, it will send 10.0.0.2 over the tailnet. It's called "Split-Tunneling" And for whatever reason if it doesn't work, you can just have tailscale disable itself when you're on your LAN. It supports "On Demand" mode

Inside your own LAN won't require an internet connection. Only when outside your LAN.

The most beginner mistake is that unraid's "cache" pool has nothing to do with caching in a traditional sense. It's just a storage pool outside of the unraid array that you typically write data to before it gets "moved" to the array via the mover system. However when setup the idea is that when you send data to unraid, you don't have to think about it, it will just automatically go to the SSD cache pool, then later the system will move that data to the storage array. 1. If you're using it for a download drive, the filesystem on it really won't matter. For a single drive in a pool by itself, you can go with any of them, xfs, zfs, btrfs. 2. Not really. There are some plugins that have been integrated into the OS in Unraid 7 which if you downgraded you would need to install them on 6.12. Dockers are 100% not dependent on unraid OS. IMO the only differences between 6.12 and 7.0 is the intel gpu driver support and running the OS without an array configured. 99% of the rest of the OS are the same features and 7.0 is quite stable. Any issues you'd have with 7.0 you're probably going to have on 6.12.

I mean yeah if you were putting a gaming load or LLM load on it sure it could be using 100+ Watts but if you're just doing encoding/decoding say in plex, it probably pulling <50W and I can't imagine it would be enough to cause crashes. But it's something to 100% consider too.

A 500W PSU should be plenty for 6 HDDs... A bigger PSU isn't necessary better for more HDDs if your PSU doesn't have enough 12v rails to handle them. You need to look at the +12V rail on the current PSU sticker to see how many Amps are on it. This example PSU has 600W on the 12V Rail and is rated for 600W. That's up to 60 10W HDDs as long as you don't exceed the Amps. (Look at drive start up amps)

Try switching to the Linuxserver container from binhex

if you’re not using an exit node, only the data that your PC is requesting from vaultwarden goes through Tailscale.

I’d go unraid 7 personally.

Cores 0-15 are your Performance cores which are hyperthreaded (core+thread) Cores 16-27 are your Efficiency cores which are slower and consume less power Quick Sync uses the iGPU not CPU cores so is not relevant. If you pass your iGPU to your VM you won't be able to use it for unraid dockers.

Yeah, I'm not really sure where to go from here honestly. I've really never had any of my own containers go rogue and fill up like that.

Go to the settings page, click docker, you might have to enabled the advanced view. There should be a default size setting. Alternatively you can see what the docker.img file's reported size is by checking the file from the webgui under your docker share.

I did a test, I wrote a 10GB file inside one of my dockers to show you what it would look like:

It appears frigate is only using ~2GB of space. Are you 100% sure that your docker image isn't set to 10GB? because 86% of 10GB is 8.6GB which is how much space is currently being used. Something isn't adding up here and nothing appears to be ballooning out of control from what I see here.

what does the container size button show on the docker containers page?

Was trying to see if there was a specific directory ballooning out of proportion but I don't see anything like that.

what does "du -h" and "df -h" report when ran inside the containers?

Well that's not how unraid is meant to be used. You're better off using the "Unassigned Devices" plugin to mount and unmount drives you don't plan on keeping attached to unraid 24/7. In your case you'd mount it, perform your backup and simply unmount it, you'd never have to stop your array remove a pool, and start it up again. You're just making everything more difficult for yourself doing it the way you're doing it. It's fine if you prefer doing it your way but be aware there is a much more simple way of doing it.

I would post your findings in the intel_gpu_top plugin support thread. Maybe the i5-8500 is too old of an iGPU (the UHD 630) for the plugin to properly get the statistics from? But we know it's working which is all that really matters.

post the "intel_gpu_top" as well as "top" while this transcode is active

You're going to need to force a transcode by changing the bitrate of the active stream if you want it to transcode. Right now it appears that it's direct playing which will not use the iGPU (or CPU).

Just to confirm, it should have been transcoding right? the media was not direct playing? There should be a status monitor on the jellyfin dashboard showing the user streaming something at the resolution/bitrate/etc. Can you screenshot that?

what happens when you play something and check intel_gpu_top in the console? Should look something like this:

Sorry, that's my fault, the numbers are supposed to be 128 not 218.

You need the entire path with it. "/dev/dri/renderD128" (no quotes) just like it says in the subtext. edit: dyslexic numbers