One of the problems with the hardware approach is the inability of MVS to bring the CHPID back online automagically when this is done.
Case in point: I played the hardware side in order to trace cables to a disk device. CF'd one at a time OFF/ON looking for the (non)blinking lights. Got to the last cable, and the MVS's came down. (whoops!). VM responded as you would think MVS should have. <g> On Fri, 15 Dec 2006 17:16:50 -0800, Edward Jaffe <[EMAIL PROTECTED]> wrote: >Shane wrote: >> On Fri, 2006-12-15 at 11:56 -0800, Edward Jaffe wrote: >> >> >>> Performing routine configuration tasks from the SE can be problematic. >>> >> >> Mmmmm - since when has yanking channels out from operating LPARs been >> "routine". >> > >The so-called "yanking" of channels is not routine. But, configuring >them on/offline is. We do it _at least_ twice a day -- sometimes more >often. And remember, it was not I that suggested the hardware function >be used instead of the MVS CONFIG command -- it was the z/OS MVS System >Commands book! All of the hardware books advise against doing this as >does the APAR referenced by Roy Hewitt (OA17152). It states: > >"The Hardware Management Console (HMC) can be used to > configure a CHPID online or offline. IBM does not recommand > this practice when the operating system is available, as it > physically configures the CHPID online or offline but does > not notify the hosts of this change." > >The words I use to describe the recommendation in z/OS MVS System >Commands are not for use in polite company. An RCF may not be strong >enough. Perhaps a doc APAR is in order... ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
