Hi Guys,

 

I have been putting off posting a message for some time now because every time I thought it would be oké. But actually things are getting worse now.

 

Occasionally I experienced that the Unraid web GUI is not working anymore. The dockers are usually not working anymore when that happens. I unfortunate did not properly write down what would happen en when. So I though because at this very moment I experience problems again, lets get some help.

 

This issue happened a couple of times now but every time after 1 or 2 days the issue would resolve by itself. Last week however the server got completely stuck (no ping, no SSH, no GUI etc) so I did a hard reset. After that I did a diagnostics (20241210) which I will post. This is not the diagnostics off the ongoing problem this very moment. Just wanted to share something already because I think the issues are related.

 

Status this very moment 2024-12-15 20:00

 

• Dockers are working
• Web GUI is not working
• Shares not accessible (SMB), can be accessed through Krusader docker though
• SSH is not working (getting a port 22: connection refused

Cannot get a diagnostics because SSH and GUI are not working. Will post when I have acces to them again and tomorrow I will have access to the monitor attached to the server. If I remember correctly I started up with GUI last time. From past experience this 'issue' should be automatically disappear.

 

Thanks in advance for any help. This is driving me nuts and because I am a sailor I do not always have (physical) access to my server. Of course I could ask the wife but better not.

 

gr Poet

tower-diagnostics-20241210-0902.zip

Edited December 15, 20241 yr by poeterdebier
clarification

Looks like you were using an incorrect password to login via SSH

 

Does the webGUI work on the locally attached monitor if you boot via GUI mode?

 

Good morning,

 

I just checked. The monitor connected to the server is full black (you see the "_" in the top left corner). No response. I can still ping the server. I still get a connection refused error when I try:

 

SSH [email protected]
ssh: connect to host 192.168.2.10 port 22: Connection refused

 

I do not get the option to write a password how you normally do.

I decided to to a power cycle to get things working again. After the server was up I downloaded the diagnostics again. I also setup the syslog server. Although, I do not have access to a second server. I setup the syslog as described in

 by Frank1940. I've used the third option. So if I need to do a power cycle again I will at least have something to look at. Attached the latest diagnostics.

tower-diagnostics-20241216-1052.zip

gr Poet

Enable the syslog server and post that after a crash.

49 minutes ago, JorgeB said:

Enable the syslog server and post that after a crash.

Will do. Thanks.

It happend again somewhere last night.  Complete freeze of system. No ping, nog SSH. Nothing to see on the local attached screen (frozen on Unraid web GUI). Attached the syslog from the syslog server and latest diagnostics.

 

gr Poet

syslog-192.168.2.10.log tower-diagnostics-20241217-0904.zip

Nothing obvious logged, but btrfs is detecting data corruption on both pool devices, start by running memtest, also make sure this has been taken cred of:

 

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/#findComment-819173

 

Memory test I did yesterday after restart. That passed without problems. Will take a look at the link you provided. Thanks. Will post if something happens again. I have to note that this is a 'new' occurrence. The server was running fine for quite some time in the past.

 

gr Poet

Edited December 17, 20241 yr by poeterdebier

Hi,

 

found the 'power supply idle control' and set it to 'typical current idle'. As mentioned in link you provided in the last post. Let's see if this improves the situation.

 

Any difference if I postpone the parity check for now?

 

gr Poet

You can cancel, just remember to run it later.

so the server crashed again. At least this time it said that I had a corrupted flash drive so no more searching. See attached syslog. I rebooted the server with a new USB although this was just a generic 8Gb USB that I had laying around. I know for sure it was not used before (so basically new).

 

I guess this is a topic for somewhere else but I do not understand why my USB sticks keep failing. I must have replaced around 6 of them for the past 4 years. Is it some setting that I got wrong? Something to do with the fact that I use a Ryzen? The last USB was a Verbatim 8GB - USB 2.0. Some tips (links) for this issue would be greatly appreciated.

 

Is it a good idea to keep this syslog server running? Or is this just extra wear and tear on the cache?

 

gr Poet

 

syslog-192.168.2.10.log

That syslog server is still showing flash drive issues, is that after the replacement?

Hi Jorge,

 

that is the syslog just before I replaced the USB. I wanted to compare the syslog at this moment (19-12-2024)  but I noticed that the syslog was disabled again. Probably because the flash backup was from before I setup the syslog server. I enabled it again and will keep an eye on it just in case I get errors again.  attached the diagnostics just in case (so that is diagnostics after the restart of the server with a new flash drive).

 

gr Poet

tower-diagnostics-20241219-1510.zip

No flash issues so far, there are SMB related crashes, possibly related to the UD disks you are mounting.