On Mar 12, 2013, at 6:01 PM, Andrew Rowley wrote:
On 13/03/2013 9:41, Paul Gilmartin wrote:
Presumably this is to avoid regressing user mods to members of /etc.
But there's a way to deal with this. Years ago I took a class in
SMP/E
software maintenance (Amdahl taught it). One of the exercises was
to APPLY a USERMOD then attempt to APPLY service that conflicted with
the USERMOD. It failed. RESTORE the USERMOD; APPLY the PTF; then
re-APPLY the USERMOD. Why is /etc/magic not handled this way?
(I'm confident that we have no user mods to that file.)
/etc is the standard location for system configuration information.
In the same way you don't use SMP/E for SYS1.PARMLIB, you wouldn't
want to use it for /etc.
I am semi surprised that there wasn't and ++HOLD Action on the Function.
Anyone?
Ed
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
