Hi Andreas.

Thanks for your quick reply.  I appreciate it.

I reckon that the misuse of init scripts is my responsibility
however the service integration isn't perfect yet.  A lot of
scripts/packages still use init scripts.  Monit comes immediately
to my mind.  It relies heavily on start/stop/restart.

And I was in a state of panic.  I know full well how catastrophic
the results of setting the clock too far away in time could do on this
system.  I couldn't remember the command name 'service' so I relied
on what has been working faithfully for years, initd.

I'll complete the transition to service whenever I have spare time.

That being said, I saved the console so you can see each and every
command I typed yesterday and the system reaction or absence of:

shizuma@gt5232h-a3e401d:~$ sudo /etc/init.d/hwclock.sh
[ ok ] Usage: hwclock.sh {start|stop|reload|force-reload|show}.
[ ok ] start sets kernel (system) clock from hardware (RTC) clock.
[ ok ] stop and reload set hardware (RTC) clock from kernel (system) clock.
shizuma@gt5232h-a3e401d:~$ sudo /etc/init.d/hwclock.sh stop
shizuma@gt5232h-a3e401d:~$ date
mercredi 11 novembre 2015, 23:26:56 (UTC-0500)
shizuma@gt5232h-a3e401d:~$ sudo /etc/init.d/hwclock.sh show
jeu 12 nov 2015 00:28:19 EST  -0.084876 secondes
shizuma@gt5232h-a3e401d:~$ date
mercredi 11 novembre 2015, 23:27:30 (UTC-0500)
shizuma@gt5232h-a3e401d:~$ sudo /etc/init.d/hwclock.sh reload
[info] Saving the system clock.
[info] Hardware Clock updated to mercredi 11 novembre 2015, 23:28:07 (UTC-0500).
shizuma@gt5232h-a3e401d:~$ sudo /etc/init.d/hwclock.sh show
mer 11 nov 2015 23:28:24 EST  -0.062948 secondes
shizuma@gt5232h-a3e401d:~$ sudo /etc/init.d/hwclock.sh stop
shizuma@gt5232h-a3e401d:~$ sudo /etc/init.d/hwclock.sh show
mer 11 nov 2015 23:35:14 EST  -0.969351 secondes
shizuma@gt5232h-a3e401d:~$ date
mercredi 11 novembre 2015, 23:37:35 (UTC-0500)


I also tried the same commands on another system I maintain, with
identical results:  stop producing no output at all:

root@NC-PH-0657-10:~# /etc/init.d/hwclock.sh show
Thu 12 Nov 2015 06:47:32 PM MST  -0.406578 seconds
root@NC-PH-0657-10:~# date
Thu Nov 12 18:47:34 MST 2015
root@NC-PH-0657-10:~# /etc/init.d/hwclock.sh stop
root@NC-PH-0657-10:~# /etc/init.d/hwclock.sh show
Thu 12 Nov 2015 06:47:56 PM MST  -0.922150 seconds


I understand that you would not maintain initd scripts.  However an 
issue is still present - and maybe this very bug description should be
changed accordingly.  It's that on shutdown for reboot, the hardware clock
wasn't properly updated.  This should be addressed.

I'd be more than happy to provide you with all the assistance you need.

Is there something particular that I should look for in dmesg|system logs?

Best regards.

* Andreas Henriksson (andr...@fatal.se) wrote:
> Control: tags -1 + moreinfo
> 
> Hello shizuma.
> 
> Thanks for your bug report.
> 
> On Thu, Nov 12, 2015 at 11:17:47AM -0500, shiz...@teksavvy.com wrote:
> > Package: util-linux
> > Version: 2.25.2-6
> > Severity: important
> > 
> > Dear Maintainer,
> > 
> > Daylights saving are on since last week-end.  A security linux kernel was 
> > also
> > issued.  I rebooted after upgrading the kernel and the time was 1h ahead of 
> > real
> > time.
> 
> Works for me.
> 
> > 
> > I remembered that there was a script to adjust the hardware clock to the 
> > kernel
> > clock in the '/etc/init.d' folder that should run whenever the machine is
> > shutwown.
> 
> Thanks for using reportbug to file this bug report. At the bottom we can
> see that you're using systemd. This tells me that the init script you're
> trying is irrelevant!
> 
> First of all, hwclock is managed via udev hooks (unless your system
> does not have udev, which is very rare nowadays). The init script
> will detect this and bail out on 'start'
> (Additionally the udev hook will bail out if detects you're using systemd.)
> 
> The 'stop' function of the script is only run if you're using sysvinit.
> This is because systemd ships an override disabling the hwclock.sh
> script.
> 
> Also, you should not directly invoke init scripts (ever!). As an admin
> you should use the 'service' command (or systemctl if you only care about
> systems which use systemd).
> 
> In short, you're trying to use an obsolete script which you should not
> be touching.
> 
> > 
> > So I set the date 1 hour earlier, and issued an '/etc/init.d/hwclock.sh 
> > stop'
> > and nada, nothing happened.  No output, no message at all.
> 
> See above.
> 
> > 
> > Strangely enough '/etc/init.d/hwclock.sh reload' do work as intended.  A 
> > message
> > is displayed as it runs to confirm it.
> 
> Please always describe what you did, what happened and what you expected.
> Unfortunately 'work as intended' tells me nothing of what you expect or
> what happened.
> 
> > 
> > This is an important bug. 
> 
> Given that you're the only one affected and noone else has mentioned a thing
> I'm not sure I agree, but additional details from you would be useful to
> determine if there's a bug at all.
> 
> > It occurred at 11:02 PM, which put the kernel clock
> > at 12:02 AM the next day.  It broke a lot of scripts relying on timespamp to
> > run properly.  Munin is broken for 24h, mythtv recorded wrong programs and 
> > refused
> > to re-record them, the log is plagged with error messages, etc.  Please fix 
> > ASAP.
> > 
> > It's 100% reproductible
> 
> I'm not sure what you expect me to do about this. You're doing several things
> wrong. You haven't really described any details about what happens.
> You're not using the proper commands to execute services.
> 
> This leaves me very much in the dark and without any information on how
> to reproduce this. Please feel free to provide additional information
> that might help me understand if there's a bug to be fixed.
> 
> > 
> > -- System Information:
> > Debian Release: 8.2
> >   APT prefers stable-updates
> >   APT policy: (990, 'stable-updates'), (990, 'stable'), (100, 'oldstable')
> > Architecture: amd64 (x86_64)
> > Foreign Architectures: i386
> > 
> > Kernel: Linux 3.16.0-4-amd64 (SMP w/6 CPU cores)
> > Locale: LANG=fr_CA.UTF-8, LC_CTYPE=fr_CA.UTF-8 (charmap=UTF-8)
> > Shell: /bin/sh linked to /bin/bash
> > Init: systemd (via /run/systemd/system)
> > 
> > Versions of packages util-linux depends on:
> > ii  initscripts    2.88dsf-59
> > ii  libblkid1      2.25.2-6
> > ii  libc6          2.19-18+deb8u1
> > ii  libmount1      2.25.2-6
> > ii  libncurses5    5.9+20140913-1+b1
> > ii  libpam0g       1.1.8-3.1
> > ii  libselinux1    2.3-2
> > ii  libslang2      2.3.0-2
> > ii  libsmartcols1  2.25.2-6
> > ii  libtinfo5      5.9+20140913-1+b1
> > ii  libuuid1       2.25.2-6
> > ii  lsb-base       4.1+Debian13+nmu1
> > ii  tzdata         2015g-0+deb8u1
> > ii  zlib1g         1:1.2.8.dfsg-2+b1
> > 
> > util-linux recommends no packages.
> > 
> > Versions of packages util-linux suggests:
> > ii  dosfstools          3.0.27-1
> > ii  kbd                 1.15.5-2
> > ii  util-linux-locales  2.25.2-6
> > 
> > -- debconf information:
> >   util-linux/noauto-with-nonzero-passnum:

Reply via email to