Raki72

Hello fellow Unraiders, I was about to unbalance my HDDs to follow invaderone's video to convert disk1 to ZFS. By doing so, I realized that when trying to move all data from disk1 to disk2, disk3, disk4,... the performance drops from 100MB/s at the beginning to 15MB/s after 10 hours. I then stopped the unbalance / rsync task and started some investigation. 1. I disabled Docker, VMs and auto-start array to be on the safe side, thenrestarted my server (22 data and 2 Parity HDDs) 2. using top, I found this task consuming 60+% CPU: /usr/local/bin/shfs /mnt/user -disks 8388607 -o default_permissions,allow_other,noatime -o remember=0 This tasks never seems to end. 3. there are also several find commands running like: root 9382 15493 0 13:33 ? 00:00:00 /bin/timeout 7102 find /mnt/user/Exchange -noleaf -maxdepth 10 root 9383 9382 9 13:33 ? 00:00:10 find /mnt/user/Exchange -noleaf -maxdepth 10 But these find commands end after a while. Any idea what this shfs task is doing and why it consumes almost one CPU core? Thanks! unraid-diagnostics-20230716-1329.zip

Thanks, it works with v6.11.3 again. Now I just have to wait until the array has been rebuilt for two empty HDDs. 😒

Hi, my server allows up to 24 HDDs. Currently I use 21 of the HDD slots: 2 parity drives and 19 data drives. Now I had the idea to add 3 more drives (Disk 20, 21 and 22) at the same time and something went wrong. What happend? 1. after adding the three HDDs, I received the normal message that new devices have been discovered and need to be cleared first. This was done without any problem. 2. next step was to shut down the array, assign the 3 new drives to the array and start up the array again. 3. now I got the message that the drives must be formatted. I acknowledged this and the formatting started. 4. for whatever reason the formating went wrong because one drive (drive 20) showed multiple sector errors. Now I had my old array with 21 HDDs plus 3 drives that showed "unmountable: unsupported partition layout". 5. I removed Disk 20, that had errors, and started the array again. Now Disk 20 showed as being emulated although there never was any data on it. Disk 21 and 22 still showed "unmountable: unsupported partition layout", so I tried to format them. 6. the format stopped after some minutes without error message. 7. I removed Disk 21 from the array. Disk 20 and 21 now are shown as emulated although there never was any data or any parity build for them. 8. Disk 22 is now the only new HDD in the existing array (Disk 1-19) but it shows "unmountable: unsupported partition layout" again. When trying to Format Disk 22, the format command stops immediately. The "old" array (Disk 1-19) is still working fine and no data is lost. But now I have the old array plus two drives (Disk 20 and 21) that I removed and that are now emulated and Disk 22 that is still in the server but I cannot format. I think, the best way to solve this issue would be to get rid of Disk 20, 21 and 22 assigned to the array and then add them disk by disk (instead of all at once). But does anybody know, how I can delete Disk 20, 21 and 22 from the array without losing any data? Thanks for your support! 😀 unraid-diagnostics-20221114-1324.zip

Not sure, how to do this. How can you exclude disks from shares?

Hi, on my Unraid server two data drives were disabled due to an error. I have two parity drives and two spare drives. But I wonder how I can rebuild the two data drives most effectively. Can I just put the two new drives in the system, add them to the Array and let Unraid do its magic with two disks in parallel at the same time. Or do I have to sequentially add one new drive first, let Unraid rebuild the array. Then add the second drive, and let Unraid rebuild the array for the second drive. Any sound idea is greatly appreicated. 😀

Maybe you can elaborate this? The only way I know is to replace disk-by-disk the smaller one with the bigger one, for all drives of the existing array. Butwhen upgrading 21 HDDs from 6 to 16 TB, the new capacity would increase by 190 TB... way too much in terms of new capacity and also price. I am searching for a way to replace the 21x 6TB drives with 10x 16TB drives (2x parity plus 8x 16TB). This would increase the overall capacity to 128TB, compared to the current 114TB, and I could add more drives later on. Is there a way to achive this without having a second array?

+1 for multiple arrays. i am also running out of space and would like to replace my 6TB disks with 16TB. with two arrays, this would be super easy.

In the past, I used to use the rtcwake command to suspend my server for a given time à la "sudo rtcwake -m mem -s 3600". Now, after some months without using this, I tried it again and all I get is a "write error" without the server going to sleep. Any idea what has changed here?

Hi, I used to run lancache-bundle 2 years ago, without much configuration. Now I wanted to reinstall it on Unraid v6.10.3 and I am not getting it to work after one week. I wonder if lancache-bundle is still supported under 6.10.3? Any idea what I could do to troubleshoot? Thanks for your support!

HW encoding works again. All I did was to switch off the server and switch on. No clue, why this changed anything, but now its working again. ⁉️

I already reinstalled the Nvidia drivers and rebooted the server - didn't help. Please let me know, if you find something.

I just realized that in my Plex docker container (plexinc/pms-docker) the transcoding is not using my Nvidia graphics card anymore. I didn't change anything in the configuration or so. The only change was to upgrade from the latest RC to the final 6.10 Unraid version. But I am not sure, if the problem already occured with one of the RCs. Anyway, the problem is that the Plex transcoder is not using the dedicated graphics card anymore, but falls back to CPU coding. I can't find any error message, so I tried to identify the problem with the following steps: - changed the Nvidia driver from v510.73.05 to v515.43.04, v495.46 with reboot and back - no impact - installed binhex/arch-plexpass and repeated the Nvidia driver changes there - no impact Any idea how I could solve this problem? Or anyone having the same problem? Thanks for any feedback. 🙂 unraid-diagnostics-20220527-1508.zip

Is there a way to run the Docker container or the binary therin with LOW priority? Thanks for your support! Edit: okay, I found this - add --cpu-priority 0 to the Additional XMRig Arguments of the Docker and it runs on Idle Priority.

Hi! Starting /sbin/haveged takes 4-5 minutes on my server. I know how to start / stop this service _after_ reboot, but I want to disable it entirely, so it's not taking all the time during boot. It is not generating added (randon number entropy) benefit for my server. The usual Linux randnom number generator is good enough for me. So the waiting time during a reboot is just painful for me. I haven't found a way to disable the service permanently. If it exists, then any advise here would be great. If not, I'd like to request that this Services will be made optional through the GUI in future releases. Thanks! Regards Raki72

I am trying to use the plugin and to bind a GTX 970 graphics card. However, all I get is Group 0 as "BIND=vices/*". Nothing to selct from. Only this. Any suggest what I should try? Thanks for any support!