Timed shutdown and restart of QNAS TS-212P with QTS 4.2 gets problematic with VirtualHere

<p>Hi,
</p>
<p>I had a ticket at the QNAP support for my NAS box TS-212P running QTS 4.2, because every morning after a timed shutdown/startup of the box in the night, I got an email with a warning, that the file system of the RAID-1 volume is not clean.
</p>
<p>The QNAP support advised to disable the CloudLink app and to uninstall the VirtualHere app.
</p>
<p>I tried first by disabling both applications and than to re-enable first VirtualHere and then CloudLink.
</p>
<p>With a disabled CloudLink app and an enabled VirtualHere app I still get the file system warning, while disabling VirtualHere, too, the file system warning does not occur anymore.
</p>
<p>So - is there anything I can do to support you supporting me?<br />
Is there somethin I can do to prevent file system corruption with timed shutdown/startup of my QNAP TS-212P?
</p>
<p>Best regards,
</p>
<p>Martin Lemburg</p>

#2

Hi Martin, some questions:

1. Does it happen if you just restart your nas normally (i.e not timed shutdown)?
2. Is the raid-1 volume composed of disks inside the qnap only, or are some disks of the raid connected via usb?

VIrtualHere doesnt interact with the RAID system at all so im not sure yet what is causing the issue...one thing might be that virutalhere takes a few seconds to shutdown and perhaps the qnap doesnt want to wait for this period when its auto-shutdown, hence the question 1 above...

Hi Michael,

  1. This file-system-not-clean warning occurs only after a planned(timed shutdown-startup.
  2. Restarting the NAS from its WebUI does not cause this error.

  3. The RAID-1 is inside the NAS, composed of to identical Western Digital hard disks.

There is no device connected to the NAS.

VIrtualHere doesnt interact with the RAID system at all so im not sure yet what is causing the issue...one thing might be that virutalhere takes a few seconds to shutdown and perhaps the qnap doesnt want to wait for this period when its auto-shutdown, hence the question 1 above...

VirtualHere has no file open, which might cause file-system troubles, if not correctly closed, but gets closed forcibly by the shutdown of the QNAS, planed in the energy settings?

Best regards,

Martin

#4

yeah thats a good point, i logged into my test qnap and got this result when listing the open files of the virtualhere process:

Checking on my qnap with virtualhere running, i see:

[/proc/26414/fd] # ls -al
dr-x------ 2 admin administ 0 Dec 3 22:00 ./
dr-xr-xr-x 7 admin administ 0 Dec 3 21:58 ../
lr-x------ 1 admin administ 64 Dec 3 22:00 0 -> /dev/null
l-wx------ 1 admin administ 64 Dec 3 22:00 1 -> /dev/null
lrwx------ 1 admin administ 64 Dec 3 22:00 10 -> socket:[1412769]=
lrwx------ 1 admin administ 64 Dec 3 22:00 11 -> socket:[1412770]=
l-wx------ 1 admin administ 64 Dec 3 22:00 13 -> /proc/bus/usb/001/043
l-wx------ 1 admin administ 64 Dec 3 22:00 14 -> /proc/bus/usb/001/001
lrwx------ 1 admin administ 64 Dec 3 22:00 19 -> socket:[912930]=
lrwx------ 1 admin administ 64 Dec 3 22:00 2 -> /dev/null
lrwx------ 1 admin administ 64 Dec 3 22:00 20 -> socket:[912931]=
lrwx------ 1 admin administ 64 Dec 3 22:00 3 -> socket:[4411]=
lr-x------ 1 admin administ 64 Dec 3 22:00 4 -> /dev/null
l-wx------ 1 admin administ 64 Dec 3 22:00 5 -> pipe:[912456]|
lrwx------ 1 admin administ 64 Dec 3 22:00 6 -> socket:[912923]=
lrwx------ 1 admin administ 64 Dec 3 22:00 7 -> socket:[912932]=
lrwx------ 1 admin administ 64 Dec 3 22:00 9 -> socket:[1412765]=

This shows the existing open files that the virtualhere server process has. All those files there are virtual files created by the linux operating system (QTS) and are not really present on a real disk. Im certain there is no RAID corruption, but i dont know why it would think that unless that process is "kill -9" by QTS and the raid is sensitive to that

Can you tell me the exact error message?

i will search google to see what is actually generating it (kernel or an open source raid software system)...