On Mon, 2 Jun 2008 11:57:14 -0500, Chris Mason <[EMAIL PROTECTED]> wrote:
>Pat > >I'm reluctant to be dogmatic about there being no connection >between "generic resources" and USS. "... >... That said, I can't see how "generic resources" >and USS can interfere with one another given that the USS >process needs to be complete, that is, the owning VTAM needs > to have created a formatted request from the unformatted >request. The formatted request leads to a >CDINIT being sent off and arriving at one of the members >of the generic resource group before the CDTERM is returned >with the selected member name for use in the final and - one >hopes - successful CDINIT being sent to the selected member. >... And I'm reluctant to trust anything I "know" about POLUIO. Some time in the distant past (maybe 15-20 years ago) we had some gear that regularly gave us this status, but I can't remember any details. It was brand-x gear (HP, maybe?). The LU's code providing status information (NOTIFY, LUSTAT, or whatever I'm tryig to remember) and the USS support code were not well coordinated. It would send "READY" before it was able to handle a USS message a/d/or stop being able to handle USS messages before it would send "NOT READY". The result was a handful of LUs being in POLUIO. My gut feeling is that in Magen's case this is more a result of the various VARYs rather than anything directly related to GR. But I could be very (vary?) wrong. (Every time I see a VARY FORCE I expect to see an out-of-sync condition, but that rarely happens these days.) Pat O'Keefe ---------------------------------------------------------------------- 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
