EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  <20072008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024  Index 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  <20072008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
<== Date ==> <== Thread ==>

Subject: RE: Daylight savings time
From: "Thompson, David H." <[email protected]>
To: Andrew Johnson <[email protected]>
Cc: [email protected]
Date: Mon, 12 Mar 2007 23:54:21 -0400
General time is involved.  Not sure where the culprit is.  If it is just us then it narrows it down,
 

________________________________

From: Andrew Johnson [mailto:[email protected]]
Sent: Mon 3/12/2007 6:34 PM
To: Thompson, David H.
Cc: [email protected]
Subject: Re: Daylight savings time



Thompson, David H. wrote:
> Something very weird just happened here and we are still trying to
> figure out why.

Are you guys running the standard R3.14.x version of
base/src/libCom/osi/os/vxWorks/iocClock.c or does General Time just
register itself as a time provider using iocClockRegister()?  If the
latter then the code it's running is all yours...

> Running iocs were updated by just setting the TIMEZONE environment
> variable and all is fine.
>
> Iocs that were rebooted today had the TIMEZONE variable set in a common
> startup script that is called before everything else.  Those IOCS seemed
> to display the correct time but camonitor was reporting the time an hour
> earlier.  The hardware timing module displayed the time an hour later
> than it was. The timing system was not rebooted and had the correct time
> on the hardware link.
>
> We fixed the problem by moving the TIMEZONE setting to the post-startup
> script.
>
> Has anyone else had similar problems?

We had R3.14 IOCs with both the correct and the old settings for
EPICS_TIMEZONE, and both behaved as I expected them to, but we're not
using a timing system with these, they're just soft timing slaves.  The
vxWorks version of the iocClockInit() routine that runs at iocInit()
looks to see if the TIMEZONE variable is set and if so it uses that, but
if not it copies the setting of EPICS_TIMEZONE to the TIMEZONE
environment variable.  After iocInit() EPICS_TIMEZONE will be ignored,
and any adjustments at runtime have to be made directly to TIMEZONE.

If you set TIMEZONE before iocInit(), does it still have the same value
afterwards or is it being overwritten somehow?  Is General Time or your
hardware event receiver driver making any use of local time?  It
probably shouldn't be.

- Andrew
--
The right to be heard does not automatically include
the right to be taken seriously. -- Hubert H. Humphrey




References:
Daylight savings time Thompson, David H.
Re: Daylight savings time Andrew Johnson

Navigate by Date:
Prev: EPICS meeting in April at DESY - first agenda and news on trainings/ workshops Matthias Clausen
Next: Re: Easy EPICS Installation for the mortal user Stefan Heim
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  <20072008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
Navigate by Thread:
Prev: Re: Daylight savings time Andrew Johnson
Next: Re: Daylight savings time Lawrence T. Hoff
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  <20072008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
ANJ, 10 Nov 2011 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·