CPU stall? Can someone explain what this is?

This AM I found tower web non-responsive and looking at the syslog I see this:

Feb  9 03:40:01 Tower logger: mover started

Feb  9 03:40:01 Tower logger: moving My Pictures/

Feb  9 03:40:02 Tower logger: ./My Pictures/Jennifer college/10942673_10152538847631783_7561285069982606248_n.jpg

Feb  9 03:40:02 Tower logger: .d..t...... ./

Feb  9 03:41:02 Tower kernel: INFO: rcu_sched self-detected stall on CPU { 0}  (t=6000 jiffies g=4495183 c=449518

2 q=27910)

Feb  9 03:41:02 Tower kernel: Task dump for CPU 0:

Feb  9 03:41:02 Tower kernel: shfs            R  running task        0  3589      1 0x00000008

Feb  9 03:41:02 Tower kernel: 0000000000000000 ffff88021fc03de8 ffffffff8105cc09 0000000000000000

Feb  9 03:41:02 Tower kernel: 0000000000000000 ffff88021fc03e00 ffffffff8105f2c4 ffffffff81822d00

Feb  9 03:41:02 Tower kernel: ffff88021fc03e30 ffffffff810766a5 ffffffff81822d00 ffff88021fc0e0c0

Feb  9 03:41:02 Tower kernel: Call Trace:

It goes on and on.

I can telnet in, things seem to be running but the web page is non-responsive.

Ideas?

thanks

david

Please post more of the log.

Are you on u raid version 6?

If its similar to what otthers have been experiencing it seems to be an issue with using ReiserFS and SHFS [the /mnt/user/ filesystem]. It causes system instability and can even cause filesystem corruptions. Thr only way prople have worked around it is by convertong sll of their drives off RFS to XFS. Search the forums for more information on how to proceed.

Another day, another RFS SHFS issue...

Yes this is unRaid 6, I upgraded last week.

I only have RFS as that is what I had before and didn't do any conversions.

The system had gone completely unresponsive now.  I can't even telnet in, I guess copying the syslog to /boot was a bad idea.

Guess it is time for a hard reboot, ouch. . .

and a parity check. . .

Is the console responding? Accessing the syslog before rebooting is the only way to determine what has occurred.

LimeTech staff,

I hope this is now getting your attention.  These CPU Stalls are becoming rampant...

lovingHDTV,

I went through this a few months back...which is here.  http://lime-technology.com/forum/index.php?topic=37311.0

Also here is another more recent case.  http://lime-technology.com/forum/index.php?topic=38019.0

Good Luck!

I was surprised that there isn't a warning in the front of the beta12 thread saying this has happened before.

I was already getting ready to add a new drive, so I'm in the middle of converting everything to XFS.  From what I've read here that is the only way to truly fix the issue.

I couldn't get the syslog off, because I tried to access /boot and that hung the machine completely.  I hard reset, did a parity check (no errors detected) and am now almost done copying my first 2GB over to the new drive.  This will take a while to complete.

I also changed all my docker from /mnt/user/apps to /mnt/cache/apps hoping that using the user share less may help while I'm converting.

david

I was surprised that there isn't a warning in the front of the beta12 thread saying this has happened before.

Part of the problem is that this issue does not affect everyone.

Both of my servers are running 6b12, and there is only one xfs drive in each, the rest are reiserfs, and I have never had a cpu stall at all.