R710. Been running for years with no issues.

Recently installed duplicacy and am working to configure backups to a separate Unraid connected via tailscale with an NFS share.

I was able to get a couple backups completed, but my main Unraid server keeps going off the network, when I plug in an external monitor to the server, I get no video. The only way to get it back up and running, is cold booting the server, which as you can imagine is unacceptable. Can anyone tell me how I can figure out what is going on here?

 

Today when it locked up, no backups were running, I was remoted into the server trying to get the NFS share to mount again, an poof, offline and crashed. Need to get this sorted ASAP. Unraid Pro running 6.12.6

Enable the syslog server and post that after a crash.

Hey @JorgeB, thanks for replying!

 

So I do have syslog server enabled, and it was pointed to my media share. I checked that directory, and there isn't anything in the syslog there.

I've since created a share on the cache drive called syslog, and attached the syslog server to it. 

In the off chance this doesn't fix this issue, is there anything I can do to test the functionality of the syslog server? Cause it doesn't appear to be doing anything.

These are my current setting for the syslog server:

 

Doesn't appear to do anything.
Syslog itself appears to be working fine, I can logger <MSG> and it displays in syslog.
Apr 12 17:24:32 Unraid Server root: Test message for syslog

But it's not writing to the directory on the cache drive I specified.

Edited April 12, 20242 yr by Original_Vecna

Those setting only set the server up in listening’ mode.     You need to set either the “mirror to flash” option or the ‘remote server” field to your servers address to get something logged as mentioned in the syslog server link.

I’m willing to bet you started having these issues around the time you upgraded to unraid 12. If you search a bit, You’ll see many peoples seem to be experiencing these weird crash or server becoming unresponsive issues.  I reverted to 11 for a while and it was rock solid again. Then I updated again to one of the newer 12.x build and started getting them again. At first it was supposed to be caused by docker’s networks settings, then it seem nobody knew whats causing theses things and the blame got shifted to mysterious hardwares issues or incompatibility that can’t really be verified. How can AMD suddenly not be recommended? I have been using AMD hardwares for years with unraid prior to these issues starting and seldomly ever had issue or crash that weren’t caused by a bad hard drive or something like that. It then started with the first time I upgraded to 12 within the firt 2 or 3 days. I personnaly will be moving back to 11 in the next few days as it doesn’t seem like this is going to be addressed.

Edited April 13, 20242 yr by SkinnyV
Correction

@SkinnyV  I have never seen any suggestion that AMD is NOT recommended - just that there are a few things you need to take care of when using AMD.   I would think a high proportion of Unraid users are using AMD and most of them without issues.

Server is an intel dell R710, the only thing that has changed recently was an upgrade to a single port Realtek 8215 from the on board 4 port 1G nic. 

One of those crashes could have certainly been due to the card, but I've since found the Realtek 8215 drivers on the community app store ( didn't know I needed to install drivers specifically in unraid and rebooted ). Thank You @itimpi for the tip. So...these settings can often times get confusing.
I don't have a remote server, and I don't want it written to the flash drive. I've created a share on the cache drive that I want the logs written to. Common sense says selecting that share in the "Local Syslog Folder" would mean your syslog is now being written to that share.

 

Is it possible to have it behave this way?

50 minutes ago, Original_Vecna said:

Is it possible to have it behave this way?

If you read the syslog server link you will see that it specifically mentions putting your Unraid servers address into the remote server field to get it to log to itself.

Ok, I'm ASKING because I don't know how to do this, and that article, which I've gone through, doesn't specify what I'm asking. OR, I'm completely bereft of all mental faculties. Which in this ask, because of your reply, I'm guessing is the likely culprit. Could you PLEASE be more specific? I'm extremely frustrated at this point, dealing with major outages for over a week now on my own, and just need the answer.  The portion about remote server is exactly for that. A remote server. So why would I assume A) I need to fill that out, and B) WHAT I need to fill out if the server isn't remote, and is actually local? I don't understand why Unraid needs remote server info for a local syslog server.

Edited April 13, 20242 yr by Original_Vecna

For the time that you are trying to diagnose this, writing the syslog to flash would be most likely an easier and better option than messing around with the syslog server. Plus, if the server start crashing, you might not get everything logged if it is affecting your share and that is where the logs get saved.

Thanks, but I don't have the space to fill up on my flash device.

For anyone who finds this in the future, the fix was incredibly simple, and could have been shared in two seconds.

Enter your Unraid IP into the remote server address field. 

 

That's it. It goes against all common sense, but that's all it was.

Thanks again for the super useful help @itimpi

 

Additional: localhost also works.

Edited April 13, 20242 yr by Original_Vecna
Additional info