[p4] Daylight Savings Time Changes Affecting Perforce Servers

[JEREMY OVERESCH]

I looked at the files that I had issues with again, and it appears that
for the entire summer of last year, the DST change was completely
nullified.  The keyword expansion is one hour off from the Date
Submitted field in P4V for the entire summer.  Also, all MD5s for
revisions made during the summer are incorrect.  Has anyone else seen
this?

>>> "Shawn Hladky" <p4shawn at gmail.com> 2/26/2007 4:31 PM >>>
We noticed a problem after applying a Microsoft patch for DST on the
p4 server.  It appears that after the latest DST patch, Windows is
using 2007 DST rules for all date conversions... even historical
dates.  So far this has caused issues with Perforce for files that use
keyword expansion and were submitted in one of the 4 weeks that are
now DST.

I happened accross this when I diffed 2 files (one synced before the
update and one synced after)... and I expected them to be identical.
They differed only by 1 hour in the keyword expaned DateTime variable.
 I checked the backup logs, and sure-enough p4 verify failed on a ton
of files.  All appear to be keyword expanded and occured in one of the
4 culprit weeks.

The verify is easily fixed with p4 verify -qv, but I'll have to live
with the fact that MS has just changed history :-)

On 2/20/07, Ken Williams <ken.williams at thomson.com> wrote:
>
>
>
> On 2/20/07 3:06 PM, "LockeyD at lakeviewtech.com"
<LockeyD at lakeviewtech.com>
> wrote:
>
> > We are running this version of the server: Server version:
> > P4D/NTX86/2006.1/109255 (2006/10/17), on Windows of course.
> >
> > With the change in Daylight Saving Time  this year our IT staff is
> > concerned there may be unknown issues with DST and the perforce
server.
> >
> > I searched the archive and the web-site, can anyone tell me
anything to be
> > concerned about with the Perforce server and DST?
>
> I have no special knowledge about it, but I would be very surprised
if any
> DST logic were actually built into p4d.  It probably gets all its
time
> information from system calls like localtime(3).  So if you update
the
> server OS itself, and its associated TZ databases, you should
probably be
> fine.
>
>  -Ken
> _______________________________________________
> perforce-user mailing list  -  perforce-user at perforce.com 
> http://maillist.perforce.com/mailman/listinfo/perforce-user 
>
_______________________________________________
perforce-user mailing list  -  perforce-user at perforce.com 
http://maillist.perforce.com/mailman/listinfo/perforce-user