I'm reading APAR OA46493 at http://www-01.ibm.com/support/docview.wss?uid=isg1OA46493 and it states twice, both for SMS managed datasets (when there's no dataclass-based volcount) and non-SMS managed, in the below wording:
"If the volume count is 1 through 5, the system allows 5 volumes. If the volume count is greater than 5, the system allows 5 plus a multiple of 15 volumes". So, you have to specify 6, 21, 36, etc. (up to 255) to be able to write up to 20, 35, 50, etc. volumes, correspondingly... On Wed, 10 Oct 2018 09:19:16 +0000, Beesley, Paul <[email protected]> wrote: >That makes sense. But seemingly that is no longer true with z/OS 2.3 and >VOL=20 means just that... try and use 21 and you get an abend. > >Regards and thanks >Paul > >-----Original Message----- >From: IBM Mainframe Discussion List <[email protected]> On Behalf Of >Vernooij, Kees (ITOPT1) - KLM >Sent: Wednesday, October 10, 2018 10:12 AM >To: [email protected] >Subject: Re: z/OS 2.3 and VOLCOUNT > >That is also how I recalled it, but the description of APAR OA46493 mentions a >start value of 6, with additions in blocks of 15. That could explain why 21 >volumes were possible with a volume count of 20. > >Kees. > > >> -----Original Message----- >> From: IBM Mairframe Discussion List [mailto:[email protected]] >> On Behalf Of Giliad Wilf >> Sent: 10 October, 2018 10:48 >> To: [email protected] >> Subject: Re: z/OS 2.3 and VOLCOUNT >> >> On Tue, 9 Oct 2018 12:53:45 +0000, Beesley, Paul >> <[email protected]> >> wrote: >> As far as I remember, specifying no volcount at all lets you write up >> to five volumes. >> If you need more - specifying volcount 6 will let you write up to 20 >> volumes, because DFSMS lets you have 5 volumes plus another whole >> multiple of 15 volumes per each "extension". >> So, if you expect 21 volume or more, up to 35 inclusive - you have to >> specify the vocount as 21. >> >> >Did something change with VOLCOUNT parameter handling between z/OS >> >2.1 >> and 2.3? >> >We have a job that uses 21 tapes to backup an application's datasets, >> and has this parameter: VOL=(,RETAIN,,20), >> > >> >Following the upgrade to 2.3 at the weekend, this job failed S837-08, >> indicating it has run out of tape volumes, which is quite correct. >> >However, looking back at the previous runs, they all have used 21 >> tapes. So they should really have abended as well, but all worked ok. >> > >> >Hence, wondering if something has changed in volume count processing, >> or whether there was a bug in 2.1 which was fixed in 2.3. >> >Would appreciate any suggestions before I face the Spanish >> >inquisition >> who are claiming that my upgrade broke their job. >> > >> >Paul >> > >> ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
