I have been been having a lot of issues recently with my new unraid build, and despite my best efforts to lurk and fix it myself, I really need to help trying to figure of my syslog.

Firstly my Unraid is virtualised using ESXi 5 and followed the Atlas fairly closly and it was working great until now.

 

Recently my server has been just crashing and I am unable to access the Web management, or my samba shares except for flash. I can still telnet in and when I try to fuser -mv /mnt/disk* or /mnt/user*, I get left with a blank line and nothing happens until I Ctrl+C. Finally if I try to use the powerdown script or manually kill off all process, neither of them allow me to safely powerdown my unraid which forces me to powercycle. After i powercycle I can access the web GUI just fine and my system start to do a parity check as I expect, but I have been met with this issue a few times today and during the last parity check it crashed again, so I am now going to ask for help.

 

I've read the syslog and my guess is that either my samba share is causing this issue or its hardware related.

 

Any assistance/insight would be greatly appreciated

syslog.txt

Well I have continued testing this myself and I've a few interesting/worrying things.

 

Firstly if I restart the machine I can access the Web gui and samba shares fine, and if I stop the array and have nothing running at all, about 10-15 mins in I come across issue again, everything stop responding. So even doing nothing it all breaks.

 

I'm going to test it out but I think it could have something to do with the spinning down of my disks now they don't wake back up or are freezing my system.

 

I have half of my data backed up I just really hope that it isn't completely broken.

 

Is there anyway I can test indervidual disk via console, quickly to see if they are an issue, as any process that requires a length of time wont work as the array will stop responding by then.

I thought stopping the array stops samba which means your system is no longer able to resolve via NETBIOS or SAMBA lookups. After you stop the array, you will need to access the system using the IP address. The 10-15 minutes is likely how long your other client systems will have NETBIOS/SAMBA name resolutions cached.

I thought stopping the array stops samba which means your system is no longer able to resolve via NETBIOS or SAMBA lookups. After you stop the array, you will need to access the system using the IP address. The 10-15 minutes is likely how long your other client systems will have NETBIOS/SAMBA name resolutions cached.

 

Samba is restarted for flash drive access.

Well after doing more testing i'm fairly certain its faulty hardware. If I remove a disk then it doesn't start and complains I have removed a disk. I can read the disk and reboot and it comes back fine. The issue is as soon as a start a parity check then the whole system freezes, web and samba access both go, which leads me to believe that a disk is freezing up and causing my problems, but I haven't figured out a way to test to find out which one.

 

When I start a parity check also none of my hard drive activity light come on so I can't tell if its any harddrive causing the issue cause none of them are doing anything.

So I finally figured out what it is. The cause of the crashes was the parity check and today when I rebooted after leaving it running for 48 hours with no changes was three of my harddrives where missing so I hope I can recover anything otherwise somehow i've had a huge crash and i have no idea how it started.