Hi there,

 

As the title sounds my Unraid Crashed today and webgui unresponsive and found this when I plugged monitor into it. This is the first time in a little bit since my last crash due to power and other issues. This time different though since those issues I'm pretty sure been resolved and also this glitchy log keeps spamming down my monitor which I can tell doesn't look good.

 

My question is what could be causing this if anything is familiar and/or what can I do to figure it out and solve it? 

 

Since this I have force restarted the server so unsure if I can now actually provide anything useful for this other than the video attached I captured.

 

Edit: For some reason it won't let me upload my attachment

Ok so I managed to snip shots from the video I took seems doesn't like my mp4 file

Enable the syslog server and post that after a crash together with the diagnostics.

 

I can do the syslog to the flash drive for when it crashes but how can I collect the diagnostics for after the crash?

I actually did have it active it seems but the syslog it has doesnt have anything related the server just all to the proxmox server container I have running. 

 

EDIT: I just uploaded it incase

syslog-127.0.0.1.log

Edited October 11, 20232 yr by Nath2125

21 minutes ago, Nath2125 said:

ow can I collect the diagnostics for after the crash?

Diags can be after a reboot.

 

20 minutes ago, Nath2125 said:

EDIT: I just uploaded it incase

Nothing relevant that I can see.

This is a diagnostics I just pulled now, since this is the reboot from the crash. Doesn't seem to have time logs from before the crash, maybe I missed it. It does have something in the end of the syslog.txt that looks similar to the screenshots I sent above when I plugged a monitor into it.

mediaserver-diagnostics-20231011-2126.zip

Diags were more to see the hardware used, but the syslog already has a clue:

 

Oct 11 19:55:08 MediaServer kernel: macvlan_broadcast+0x10a/0x150 [macvlan]
Oct 11 19:55:08 MediaServer kernel: macvlan_process_broadcast+0xbc/0x12f [macvlan]

Macvlan call traces will usually end up crashing the server, switching to ipvlan should fix it (Settings -> Docker Settings -> Docker custom network type -> ipvlan (advanced view must be enabled, top right)).

Ok, il give that a go tomorrow and report back.

So I havent yet had the opportunity to do this change, but in that time Ive ran into another issue which Im providing the diagnostics below, it has caused all my docker containers to stop and error when i try to restart them, it hasnt crashed the server but it seems containers no longer work.

mediaserver-diagnostics-20231015-1042.zip

Docker image went read only due to corruption being detected, corruption is also being detected on the pool, suggest running memtest.

After a memtest what other procedure should I be doing to resolve this? Like, do I need to rebuild the drive or something? Simple restart and find what's been corrupted?

After memtest scrub the pool and recreate the docker image.