No, ML1 datasets cannot span volumes. Use a larger volume. Also, there is a parameter that says if the datasets is over a certain size it has to migrate directly to ML2. As your free space on ML1 goes down, this will need to be adjusted. Monitor your migration log for failures and note the size of the datasets that fail for insufficient space.
On Thu, Dec 6, 2012 at 2:36 PM, Judith Nelson <[email protected]> wrote: > Hi, > We recently converted some of our production data from tape to disk, for a > different DR strategy. I setup MC's to send the -1 of the generation datasets > to ML1 disk volumes, keep the 0 generation on disk and send the older GDS's > to ML2 tapes. This works great, with the exception of the largest datasets. > During Primary Spacemanagement I get following message: > > ARC1030I DATA SET PTP.DCD.DC80.IFMR.PURGE.ARC.G5261V00 OF 331660 TRACKS > ELIGIBLE FOR MIGRATION/BACKUP TO AN OVERFLOW VOLUME HAS BEEN REDIRECTED TO A > NOOVERFLOW VOLUME > ARC0757I MIGRATION OF A DATA SET OF 331660 TRACKS TO ML1 VOLUME HSM004 FAILED > FOR INADEQUATE SPACE > > I added 5 mod9 volumes as HSM ML1 overflow volumes and tried the migration of > the dataset to ml1 again. It had the same result. > A mod9 has 149519 trks to use, so the dataset above would have to span over 3 > volumes. It just won't do it. > I was trying to find information to see if datasets can't span over volumes > if they are ML1, but had no luck. > > Can someone tell me if there are restrictions on dataset size, larger then a > single volume, or how I can accomplish getting these large datasets to ML1? > > Any help would be appreciated. > > Judith Nelson -- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
