It got me wondering...
I have a test LPAR, which z/VM 5.1 has been up for about a week along
with SLES9 images.
I haven't bothered to do anything on this lpar relating to the time
change, as it will just correct itself at the next IPL.
Even though CMS still reports CDT and is a hour off, the Linux images
all show CST and the correct time. I'm assumming there was a cron task
that fired up and set the correct timezone.
Since Linux is automagically doing this, it would imply to me that most
all applications under Linux, are not sensitive to the timezone being
changed.
Just....interesting.....
Tom Duerbusch
THD Consulting
>>> [EMAIL PROTECTED] 11/01/05 9:53 AM >>>
For the benefit of all the folk on this list who're new to IBM
mainframes,
and with apologies to the, "old folk" for stating what is to us so
obvious
that it's like breathing, can I please point out that:
The Commercial IBM Mainframe Operating Systems (and, I believe Linux,
but
I'm happy to be corrected on this one) are all DESIGNED to have the
hardware TOD Clock set to GMT, with a software timezone-offset being
applied by the operating system in order to derive, "Local Time".
Setting the system ("hardware/LPAR/TOD") clock to anything other than
GMT
is incorrect. As with many, "incorrectnesses", the impact can vary from
zero to quite significant but at the very least be aware that doing
this
is non-standard and has the potential to cause problems. Preferably,
don't
do it - but if you choose to, at least make a conscious decision to
walk
out of step with the standard, and don't do it just out of innocence /
ignorance (as I did for many years back in my callow youth).
Jeff
