Hi,

I am on Unraid 7.1. I have an Intel skylake i7, 32 GB of ram, 6 HDs with 2 of them acting as parity drives. 

 

My Unraid box’s web UI is unresponsive after starting and entering my encryption phrase. I have tried reimagining the flash drive from a backup, but that did not help. The issue seems to occur almost immediately after the race starts and parody check has begun. If I boot and let unread come up, the box will sit there and I can navigate the unread – no problem. The issue seems to only occur after the array starts.

 

I can SSH in but the documentation I have found from Google seems to he depreciated which is frustrating probably my fault for not finding the right link but old commands were on Unraid wiki that Google returned. 

 

I can not access any SMB share. My docker container appear to be running but syncthing doesn’t appear to be able to sync any data. 
 

I do not use cloud services for data syncing and rely heavily on this box with sync thing to sync my data. I have spent 2 weeks trying different things on and off so as a last ditch effort posted here for help. Hopefully somebody has seen this it’s been very frustrating being without my data automatically syncing for 2 weeks. 

See if you can get the full diagnostics over SSH, if not, get and post the syslog:

 

cp /var/log/syslog /boot/syslog.txt

 

Hey Jorge,

 

Thank you for the response sorry for the delay. I have the sys log here in this post.

syslog.txt

I cannot see an array start attempt logged, was the syslog saved after that?

 

 

syslog.txt

 

Yes sorry here is the booted version.

 

One note. I removed a few plugins that I had hoped were causing the issue. Doesn't look like it fixed it.

 

So, I booted to safe mode and it seemed to be ok there. So I booted back to normal, disabled docker via settings, then started array and it seemed ok. I then started docker and it was ok but updated some containers and now it's unresponsive. This syslog is from the unresponsiveness.

Mar 25 13:47:19 server kernel: nvme nvme0: I/O 194 (I/O Cmd) QID 1 timeout, aborting
Mar 25 13:47:19 server kernel: nvme nvme0: I/O 192 (I/O Cmd) QID 3 timeout, aborting
Mar 25 13:47:19 server kernel: nvme nvme0: I/O 1 (I/O Cmd) QID 7 timeout, aborting
Mar 25 13:47:19 server kernel: nvme nvme0: I/O 32 (I/O Cmd) QID 8 timeout, aborting
Mar 25 13:47:19 server kernel: nvme nvme0: I/O 33 (I/O Cmd) QID 8 timeout, aborting

 

Logs show issues with the NVMe pool device, do you have a different device you can try with? Or just try using the array only and see if it's stable, but note that docker image/appdata is likely in the pool.

 

 

Yes after that error I disabled docker again and, it has been over 24 hours and everything is still running. 

 

So it looks like the cache drive and or the docker image could be the issue? How can I trouble shoot this?

 

I have only one nvme slot on the board I would have to replace the entire drive if switching the drive is recommended.

Edited March 26, 20251 yr by kernelpaniced

Those errors above are hardware related, so it's not a docker image issue, I would recommend replacing the NVMe device.

Gotcha can do. Right now I am using nohup and rsync to copy the from /mnt/cache to a new share I setup with array only. Then once I install the new drive I can just copy them back to /mnt/cache pool correct?

Correct, make sure docker and VM service are stopped before starting the copy.

 

It's possible that the docker image will also need to be rebuilt, but that can be done later, with the new device.

Just wanted to share a quick summary of an issue I had and how it was resolved, in case it helps someone else down the line.

 

My Unraid WebUI became unresponsive, and when I SSH’d into the box, I noticed /mnt/usr/ was empty. After some great troubleshooting help from JorgeB, we found the cache drive was faulty—it was throwing errors and preventing the array from mounting properly.

 

Replacing the cache drive fixed everything. Hope this helps anyone facing a similar problem.