5

I have a QNAP TS-412. It has 4x 4TB WD Red drives in RAID10.

The poor thing works pretty hard. Virtually 24/7 it runs at 20-30MB/sec as a backup device from a bunch of office computers and a handful of servers.

In the last 2 weeks, seemingly out of nowhere, it has been rebooting every 16 hours or so. It has been under this sort of workload for about 3 years, with only breaks to upgrade the drive capacity every 18 months or so.

Is there anywhere I can go in the console (or via SSH) to see the reason why it is rebooting?

5 Answers 5

2

Connect to your QNAP with a terminal and look in /mnt/HDA_ROOT/.logs. Specifically, kmsg. You should see a lot of deep information for your perusal, interpretation, and ultimate edification.

Another option is going to the Web UI >> Select System Administration >> System Logs >> tabs across the top will give you UI access to various files.

As for what to look for, it's usually a crap shoot of what looks bad, and what Google tells you is worse.

4
  • 2
    Meow meow meow meow meeoowww meow (translation: There's a lot of stuff in there; anything specific I should look at?) Oct 3, 2014 at 6:08
  • Meow meow meow meow meeeooowwww meow meow meow meooowwww (I had a look at the section where the timestamps go backwards, indicating a reboot, but can see nothing of interest. I will keep looking). Oct 3, 2014 at 6:16
  • Mew mew rowww SKREEEE! (QNAP sucks)
    – Wesley
    Oct 3, 2014 at 6:18
  • Meow meow meeoowww hhhuuukkkkkkkkkkkkkk meow siiiigh (Then why do I keep buying them?! Sigh.) Oct 3, 2014 at 6:19
2

if you enable ssh login then

ssh admin@<qnap_ip_addr_here>
password: admin

then look at /var/log

grep ERR *log

look at hal_lib.log and hal_lib.log.bak specifically

look for stuff like this. I don't think my HDD was 60C and am suspicious about whether sometimes HDD temp is read incorrectly. It's possible these are setting thresholds not the actual readings. Not sure.

hal_lib.log:Get_Temp_Threshold() called, SYS_ERROR_TEMP=70
hal_lib.log:Get_Temp_Threshold() called, HDD_ERROR_TEMP=60
hal_lib.log:Get_Temp_Threshold() called, CPU_ERROR_TEMP=85
hal_lib.log:Get_Temp_Threshold() called, CPU_ERROR_TEMP=85

I'm also curious if other people see this:

manaRequest.cgi:Tue Oct 21 16:45:48 2014
PD_Is_Exist: can't retrieve port_id value!
hal_daemon:Tue Oct 21 16:45:59 2014
root_get_fan_speed(978): NOT implement.
se_sys_get_SAS_HBA_info(3606):System Error.
disk_manage.cgi:Tue Oct 21 16:46:03 2014

you may want to go inside the QNAP qui (hardware) and disable smart fan speed setting and switch to manual speed (medium always?)

when it says "root_get_fan_speed(979): NOT implement." I'm suspicious. Since fan speed seems correct in the gui. ~9k when I choose medium, and >10k when high speed is chosen

0

I could not find anything in the logs as per Wesley's answer. As a result I have gone on-site and swapped the QNAP chassis for a spare chassis I had and put the old drives into the new chassis.

It seems to have been stable for over 2 days now, whilst still under the same sort of load. I suspect the hardware has failed due to an overheating issue, as it was in a reasonably warm location for a very long time.

0

I wonder if there was some unexpected failure that caused a reboot, then QNAP was deciding it had to resynchronize the drives. And normal operation is allowed while resynchronizing. But I've had problems with activity during the resync. If you have a "constant load"...and it rebooted every 16 hours, I wonder if it was the interaction of the load with the resync.

I had a similar issue, and couldn't get out of the reboot loop until I stopped my backup jobs and let the resync complete (you can find the resync status on the appropriate place in the gui (in storage manager I think). Also the log tells you when a resync starts. Not sure if it logs when it's done. The gui definitely will show while it's going (I think as a task/job at the top of the main page also).

summary: check if it's resync'ing. If so, stop outside activity till the resync completes.

You might also set the smartfan setting to manual high speed, to get best cooling.

0

You can also check if rebuild is busy, and how long it will take in mins, by ssh into the NAS and perform:

# cat /etc/mdstat

You will see the RAID rebuild including remaining time.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .