I'm moving files around so I can start making use of ZFS. This issue has been going on for most of this year, but I've rarely needed to move large amounts of data around so haven't spent much time on troubleshooting.

 

I started doing large moves using rsync via CLI and had problems, so moved to unBALANCE as it gives better visibility of what's going on. The crashing appears to be the same when I use either method.

 

I find that the move completes correctly, but the main Unraid GUI becomes unreachable after a while. So for example, the whole move may take 5 hours, but the GUI becomes unreachable after 2 hours. If I try to connect with the "Array - Unraid Monitor" Android app while the GUI's borked, it displays a "Sever error" message. When the Unraid GUI is down, the unBALANCE GUI is not affected and still runs fine.

 

I created the attached diagnostics while unBALANCE was running. It was a 172GB move, took nearly 50 mins and completed successfully.

 

This is what top looked like while the GUI was crashed after an earlier, larger move. This was many hours after the move had finished. No dockers or VMs running:

tower-diagnostics-20230820-1400.zip

You can enable the syslog server and post that after a crash, to see if there's something relevant logged, don't see anything in the diags posted.

 

I also use rsync and had an issue with my GUI crashing, but I never made the connection. For me, the issue seemed to stop when I upgraded to 6.12.3, but that could also be coincidence. I don't put a ton of data on my server to need to sync.

9 hours ago, JorgeB said:

You can enable the syslog server and post that after a crash, to see if there's something relevant logged, don't see anything in the diags posted.

 

Thanks for checking the diags- I have some more moving to do and will be sure to post the logs.

 

7 hours ago, sphbecker said:

I also use rsync and had an issue with my GUI crashing, but I never made the connection. For me, the issue seemed to stop when I upgraded to 6.12.3, but that could also be coincidence. I don't put a ton of data on my server to need to sync.

Interesting. I've not been able to find another example of someone having the same issue, so somewhat reassuring to know I'm (possibly) not the only one.

Then again, this was happening when I first started using Unraid last year (6.11.3). I didn't upgrade until the 6.12 releases came out and have kept up to date since then. Unfortunately the problem has persisted across all of the versions I've been on.

I ran a large unBALANCE transfer and, typically, it all went without a hitch.

 

 

Not sure why it took so long, but it completed successfully and when finished the Unraid GUI was still running fine. Will continue to monitor the situation and will post if things go awry again.

This is happening to me now just upgraded to 6.12.0 and I'm trying to do a rsync from ARRAY to external zfs drive with 15TB of data, through a user script running in background. Last I peaked at progress was around 1.5TB done of the 15TB. GUI has become unresponsive after 2 Hours. Can't access GUI locally or through WireGuard Remote Access tunnel. The Tunnel establishes but can't access the GUI. Ping still works. I don't have SSH enabled so I don't know if that no longer works. Also all docker containers with separate IPs their GUIS load, but they can't reach the internet.

 

Update: 2024.07.22

So it turns out I had a corrupted ZFS Pool on my NVME pool which was in raid 1. I just re-formated and recreated the pool and this error seems to have gone away. So I thought. I now found another issue that crashed the UI for me once. I had WireGuard off, I had docker containers running with one ipvlan pool running fine. I then turned on WireGuard Tunnel and it crashed the UI. I then hard-restarted and then clean restarted. I then turned on WireGuard First, then turned on my dockers and everything seems to be ok now. If I turn on docker first then Wireguard it crashes the GUI. This is reproducible. I have the logs saved in case this can help a developer. However I glanced through the system log and no kernel panics or errors jumped out at me. I also upgraded to 6.12.11. I did run meatiest with no errors. I also went through a hardware change while this corruption happened. I migrated to Asus w680-ace ATX motherboard with 64GB ECC ram. ECC Ram passed mem-test also. So ill keep an eye out if the system remains stable you won't hear from me again  else ill post more updates, here and any additional findings.

Edited July 22, 20241 yr by alfredo_2020