Hello Helge, On Mon, Jul 18, 2016 at 08:24:38PM +0200, Helge Wiemann wrote: > Issuing... > > echo +150 > /sys/bus/i2c/devices/0-0030/rtc/rtc*/wakealarm > > ...would not work as I don't have the "rtc" directory. But you have /sys/bus/i2c/devices/0-0030? Even if you didn't unbind the driver with the command line that I mentioned in the bug report? > But I have attached > my log file and it clearly indicates there is indeed an RTC issue. I only see the following related to rtcs: > Jul 12 21:22:16 MediaCenter kernel: [ 2.815689] rtc-mv rtc-mv: internal RTC not ticking > Jul 12 21:22:16 MediaCenter kernel: [ 2.828086] rtc (null): invalid alarm value: 1900-1-12 19:19:0 > Jul 12 21:22:16 MediaCenter kernel: [ 2.834150] rtc-s35390a 0-0030: rtc core: registered rtc-s35390a as rtc0 > Jul 12 21:22:16 MediaCenter kernel: [ 2.861368] rtc-s35390a 0-0030: setting system clock to 2016-07-12 19:22:02 UTC (1468351322) The first is harmless, the second is also harmless and fixed in when my patches go in. The two last lines are obviously ok. One thing that I consider strange is: > Jul 10 14:28:58 MediaCenter systemd[1]: Starting [Cron] "00 23 * * * /sbin/poweroff"... I guess something (anacron?) tries to catch up that the machine was already off at 23:00 yesterday? That could explain why you have to boot twice and is independant of an rtc issue. > I do not have an answer to your questions, but let me explain again what > happened before I applied your patch: > > Every other boot or so when I pushed the ON button on my NAS the system > would try to boot (first beep) and then immediately shut down (another beep) > while in the middle of the process. I think you will see the exact same > thing in the log files. > > It would not matter how I would shut down the system the night before > (either by pushing the button or sending an SSH command), my NAS would > consistently show this pattern every second time I would try to boot it > again. > > Whether it has anything to do with RTC I do not know, but I doubt it. All I > can confirm is that your fix did the magic and I have not had to start the > NAS twice ever since. If my suspicion above is right, it could only be that the rtc is relevant because it influences the check "Given the last shutdown time and the current boot time, was the job at 23:00 last night done or should it be done now instead?". Best regards Uwe
Attachment:
signature.asc
Description: PGP signature