On Mon, 6 Jun 2005 10:28:22 -0400, Bruce Black <[EMAIL PROTECTED]> wrote:
>Are you certain that the VTOCIX is full? Pretty sure. Or at least it was. It got a little better after a migration run over the weekend. On Friday, an attempt to rename a dataset on the volume got: IEC603I VTOC ERRORS MAY EXIST ON 0B89,PROD01,8,027 IEC331I 042-002(0812041B),SYSMATT ,$TSOUSER,RNAM,IGG0CLH2 IEC331I VOL,PROD01,NAME,SMFMVS.NEWSTAT IEC614I RENAME FAILED - RC 008, DIAGNOSTIC INFORMATION IS (0812041B),598 Looking up the error message, the doc seemed to say it was caused by not being able to create a VTOCIX record. >Generally the VTOCIX takes >only a fraction of the size of the active VTOC, so this would be true >only if you have a large VTOC with many datasets or VSAM components, and >a small VTOCIX. According to the numbers that IBM quotes (in an >appendix of the ICKDSF manual), the VTOCIX needs to be only about 5% of >the size of the VTOC, so a 5 track VTOCIX is adequate for 100 track VTOC. That doesn't seem to be true here. We have several 3390-3 volumes with 75-track VTOCs and 14-track VTOCIXes. Running the report you suggested shows VTOCs ranging from 85-95% full, and VTOCIXes running from 97-99% full, so it looks like in our case, it's closer to 20% than 5%. I'm not sure why. I know we have some weird applications that create a bunch of empty 1-track datasets (and maybe even 0=track), so we tend to use a lot more DSCBs on a 3390-3 volume than saner installations would, but I don't know if that affects the VTOC-to-VTOCIX ratio. > >If they are truly full, the jobs that others have suggested can be used >to expand the VTOCIX without moving the VTOC. On today's control units, >there is no particular performance advantage to having the VTOCIX next >to the VTOC, only a space fragmentation advantage. I'm aware that placement is no longer a performance issue, and I don't really care about fragmentation, so moving the VTOCIX is okay. Unfortunately, since the VTOCs are also close to full, they'll probably fill up next. And, with the VVDS right next to the VTOC, the VTOC can't be expanded without moving either the VTOC or the VVDS, and I don't think either of those can be moved with the volume active. (I realize there's also no reason for the VVDS and the VTOC to be adjacent, but they are, so I'm stuck with it.) ---------------------------------------------------------------------- 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
