updated form 6.11.5 to 6.12.1 because i couldn't transfer any data to a share, turned out to be a dead cache pool ssd but now my server is crashing or hanging constantly.

You can enable the syslog server and post that after a crash together with the complete diagnostics to see if anything is visible.

ok this might be a dumb question, but do i have to worry about any security vulnerabilities when i post that here ? 

I don't, and most users don't as well, but cannot speak for you, the diags are anonymized by default.

hopefully i did that correctly 

syslog neo-diagnostics-20230627-0803.zip

Are you using Firefox to access the GUI?

no safari which what i always have used without any issue before. Its what my computer chose.    

 i forgot i was trying to access the server on my other laptop which runs pop os and uses firefox but i couldn't even get it to find the server or access it, if thats what your seeing. That was yesterday sometime i've rebooted the server a few times sense then i think so that should have cleared that from the logs right ?

Edited June 27, 2023 by SpaceGator

1 hour ago, SpaceGator said:

if thats what your seeing.

Possibly, the errors I see that are usually caused by that are at the beginning of the persistent syslog, but the whole syslog is only a few minutes long, and there's nothing else out of the ordinary logged that I can see.

I turned the mirror cash on in the syslog then rebooted the server which took a bit to be able to get access, then when i finally got in and turned the array back on which lead to me being kicked off and not being able to access the server for about 30 min i guess and plex was also unavailable at that time. 

 

So when i was finally able to access the server again i assumed that was all that was needed to see the problem. I downloaded the flashback up and pulled the syslog from that then sent it over. 

 

Does it need to be left on for longer ? or was there another mistake i could have made trying to collect the data you requested? 

 

You seeing the firefox dosen't make sense to me if the syslogs are cleared after reboot. that laptop is powered off and unplugged form power and i was attempting to access the server a day ago and many reboots ago with it. Is it possible ive been hacked or something. 

11 minutes ago, SpaceGator said:

You seeing the firefox dosen't make sense to me if the syslogs are cleared after reboot.

Persistent syslog is not cleared after a reboot, that's the point of it.

 

Syslog starts at Jun 27 04:18:37 and ends at Jun 27 04:38:58, and the array was not even started in the last boot.

 

 

I'm confused. If the persistent syslog is what i enabled and what was active from Jun 27 04:18:37 and ends at Jun 27 04:38:58 then you shouldn't have seen firefox. the only firefox activity from me was on the jun25 or jun26. I'm positive i turned the array on in that time i enabled the Mirror to Flash at least it showed me the array was on maybe it didn't. actually turn on but it was on in the gui when i was able to get back in. 

 

should try again and leave it on for longer like 8 hours or so ? and should i reboot after renabling it ? is downloading the flashback up and pulling the file form that the right way to collect the syslog? 

Usually you enable the syslog server, let it run until the server crashes, then post that log.

maybe what I'm experiencing is not a crash. Does a crash require a reboot to get working again ?  ill i know is that i was kicked off the server a few times during that time period and the array said it was on after i turned it back on after enabling syslog server and rebooting and the plex docker said it was on but plex couldn't access the server as well.

I'm going to try and re enabling Mirror to Flash then start the array with out a reboot involved at all this time and leave it running until i can respond again. I will also post the diagnostic again at that time as well. hopefully it shows something because the server is unusable as it is now.     

It seems to be stable at the moment. There was no one using plex today so idk if it was in and out. I guess its a intermittent thing because i started the array before i slept and it kicked my off and was not responsive ect... like its been doing. I just turned off the array and back on again and it was fine soo idk.

 

Ill post the diagnostic i just pulled. 

neo-diagnostics-20230628-0320.zip

Everything looks normal to me so far.

Yea its still stable. Hopefully it stays that way and it was just something weird that corrected its self but thats probably wishful thinking haha. 

 

Ill keep monitoring it and if it starts again ill post another diagnostic at that point. If i remember correctly this started after a day or so of it running ok. 

 

Thanks for helping me so far i appreciate it. 

It hasn't been as bad as it was but it just acted up and plex dropped connection for a bit and i wasn't able to get in to the unraid gui ect... 

 

So i thought i would run a another diagnostic to see if anything is there but it doesn't seem like it looks any different from the others. 

neo-diagnostics-20230629-0410.zip

Unfortunately there's nothing out of the ordinary logged that I can see.

Is this warning anything out of the ordinary ? "Jun 28 00:17:41 NEO kernel: BTRFS info (device loop3): using crc32c (crc32c-intel) checksum algorithm" there also a warning for loop2. I just keep seeing that in the log. 

 

If plex wasn't being dropped for other clients on the network i would assume it may be me computer not being able to reach the server, i also would assume it would be my network causing the problem if all these problems didn't start right after the update. I guess it could be my network if the server change something during the update but idk. 

 

I checked and my docker settings are using macvlan could that because these issues ?