[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian Jessie on QNAP TS-112P - Reboot instead of shutdown



Hello Uwe,

It took a little longer as I was out of town for two weeks.

Ever since I applied your patch the system clock would not sync and point to January 1st (1970?). Is there any way I can reverse the stuff I did to my NAS?

Once I have done that I will retest again and see if I get to the gist of the problem. I want to go through the whole procedure again and understand what the issue is.

Would you mind? Thank you.

Helge

On 07/19/2016 10:10 PM, Uwe Kleine-König wrote:
Hello,

On Tue, Jul 19, 2016 at 09:47:04PM +0200, Helge Wiemann wrote:
I do have "/sys/bus/i2c/devices/0-0030", but did not specifically follow
your instructions you described in the bug report. If it helps you I will,
please let me know.

I am running a cron job to switch off my NAS at 11 PM. However, my issue
(shutting down after a short boot) was not related to that, I did not see
any particular pattern at all. This issue did still occur when I switched
off my NAS using the power button or SSH. So I doubt either RTC or Crontab
are the culprits.

I think you're wrong here. Reading your syslog I see the following:

  - You shut down your machine at Jul 9 22:29
  - You booted it at Jul 10 14:28
  - Log has:
  	Jul 10 14:59:37 MediaCenter systemd[1]: Starting [Cron] "00 23 * * * /sbin/poweroff".
	Jul 10 14:59:37 MediaCenter systemd[1]: Started [Cron] "00 23 * * * /sbin/poweroff".
    and in the following the machine shuts down.
  - In the fo llowing boot at Jul 10 15:00 this is not mentioned and the
    machine boots up successfully.

So my suspection is that at 14:59 the poweroff job was caught up for
because it didn't run the day before. Then at 15:00 it wasn't run,
because it already was active a minute before.

Assuming your machine still has a (more or less) accurate date in its
rtc, I'm sure the issue returns if you disable the machine before 23:00
and restart it the next day. Given that you don't have the path
/sys/bus/i2c/devices/0-0030/rtc maybe the driver fails to bind now
because the rtc hardware is in a strange state and so the cron daemon
doesn't notice it has to catch up for the poweroff job?

Can you provide a boot log of the supposed fixed current state?

Best regards
Uwe



Reply to: