TheDragon

Members
  • Posts

    389
  • Joined

  • Last visited

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

TheDragon's Achievements

Contributor

Contributor (5/14)

4

Reputation

  1. I'm trying to setup custom permissions on disk and user shares which will be mounted via NFS4 on a Debian server. I need them to match a UID/GID I use on that Debian server. I've set permissions for the local folders (via ssh as root) using chown -R UID:GID and chmod -R xxx Then setup NFS for those user/disk shares, and mounted them on Debian - everything works perfectly at this point. However if I stop the array and restart it, all the custom permissions on the local files on unRAID are reset back to nobody:user 777 Is there somewhere I can either prevent this action of resetting my custom permissions, or change the user/group/permissions that are being set automatically? Or is there a way to have all files/folders appear as owned by a specific UID/GUD on my Debian server once mounted, without having to change the ownership/permissions locally on unRAID?
  2. Aah thanks for sharing that. Doesn't seem logical but at least seems isolated to that attribute! Thanks for your help @JorgeB
  3. If we focus on just the power on hours for now, how could it be a device issue if its showing a value above 0? Just doesn't seem to make sense logically
  4. Does anyone have any other ideas or know more re how SMART works with NVME drives?
  5. Its the power on hours specifically that caught my eye, as its clearly not updated, which in turn made me doubt whether the percentage used attribute or other attributes were also being updated. Do only some of the SMART attributes update on this type of drive? Is there anything else I could try?
  6. Sorry for the slow reply, diagnostics attached.
  7. I'm on the current unRAID version, and generally update within a week or so of a new version release. So yeah its certainly possible, but not sure if this is correlation or causation!
  8. I have an NVME drive which up until 10 days after installation was showing updates to the attributes page - since then (when I looked today) it hasn't updated since (this is months, not days/weeks). Is there something obvious I'm overlooking, or does this indicate a problem with the drive?
  9. Is there any updated installation steps for Linkace? Tried various methods - following the steps in the docker description, and also from the official Linkace website, simple and advanced, sadly no joy whatsoever!
  10. I'm not sure if all the steps I took are totally needed, I just figured if it ain't broke don't try and fix it! Also to clarify settings in Plex Transcoder section, I have - "Enable HDR tone mapping" disabled "Use hardware acceleration when available" enabled "Use hardware-accelerated video encoding" enabled With those selected it uses HW. If I enable HDR tone mapping as well, everything is done by the CPU.
  11. Just to share my experience. Using an i3-12100 on a Z690 motherboard on unRAID 6.11.0-rc2 I now have HW transcoding working perfectly, the only thing that doesn't work is tone mapping. With this - echo "blacklist i915" > /boot/config/modprobe.d/i915.conf Intel-GPU-TOP GPU Statistics and /dev/dri in the official Plex docker container. No other changes from default configuration. I've played a 4K home movie in Plex, transcoded down to 720p without issues. I've also converted it to a lower quality and downloaded it to a android phone with no problem either. Both definitely using the GPU as can see low CPU usage and high GPU usage throughout. I've had no crashes or errors in either my system or container logs. If I enable tonemapping, I don't get any errors or crashes, but instead of using the GPU it uses the CPU. Hope this is useful to others!
  12. Quick question, have a new build with a i3-12100 on 6.10.0 where I've left a monitor plugged in and turned on but have no /dev/dri folder.. I tried using the Intel-GPU-TOP plugin with this version of unRAID which from reading I think are the only requirements - am I missing something stupid or is this expected behavior without the custom kernel?
  13. I've run across an issue where several times mover has failed to complete transfers from cache to array disks due to lack of space. The files on array disks show up on MC as .filename.flac which isn't the full file based on size. After making space i was able to get it to move over and ended up with filename.flac as well as .filename.flac Is there a way to search my array disks for other partial files to delete them all? Just as I only noticed the partials while trying to free up space, and they could be littered all over the array disks wasting space. Any suggestions would be much appreciated!