Thanks Kris! I was wrong about the location of the Format-4 DSCB, it
doesn't always follow immediately after the volume label record. The
volume label record is always the third record on cylinder 0 track 0. Th
e
volume label record has a pointer to the first record in the VTOC, (which
is always a Format-4 record). The Format-5 record is always the next
record after the Format-4 record. Here is part of the volume label recor
d
from Kris's z/VM 5.3 install: (I'm posting from the web interface, which
doesn't allow very long lines, so I'm only posting parts of the DDR TYPE
output.)
CYL 00000000 HD 00 REC 003 COUNT 0000000003 04 0050
00004 0004 KEY LENGTH
00000 0000 E5D6D3F1 *VOL1
00080 0050 DATA LENGTH
00000 0000 E5D6D3F1 E5E3C5F5 F3D9F000 00000005 *VOL1VTE53R0.....
The DATA portion of the label starts with VOL1, then the 6 character
volser, (VTE53R), a one byte "security" code, ("0" - x"F0"), and a five
byte pointer to the first record in the VTOC, (Format-4), in CCHHR
format. This label says the VTOC starts in cylinder 0, track 0, record 5
.
Here is part of the Format-4 DSCB, (record 5):
CYL 00000000 HD 00 REC 005 COUNT 0000000005 2C 0060
00044 002C KEY LENGTH
00000 0000 04040404 04040404 04040404 04040404
00096 0060 DATA LENGTH
00000 0000 F4000000 00050000 00000000 00000001 00000D0B 000F
The KEY area actually contains 44 bytes of x'04' and the first byte of th
e
DATA area is x'F4', which indicates that this is a Format-4 DSCB record.
The x'0D0B' is the number of cylinders, (3339), and x'000F' is the number
of tracks per cylinder, (15).
Here is part of the Format-5 DSCB, (record 6):
CYL 00000000 HD 00 REC 006 COUNT 0000000006 2C 0060
00044 002C KEY LENGTH
00000 0000 05050505 00010000 00000000 00000000
00096 0060 DATA LENGTH
00000 0000 F5000000 00000000 00000000 00000000
The x'05050505' in the KEY area indicates that this is a Format-5 DSCB.
Immediately following the key identifier, the x'0001' is the first free
track available, (relative to the beginning of the volume). The x'0000'
after that is the number of free cylinders in the extent, in this case
zero. The x'00' after that is the number of additional free tracks, also
zero. Other free extents would follow this one, (there are none). This
Format-5 DSCB indicates that there is NO free space on this volume, which
is what z/OS would believe also.
Kris's CPVOL formatted T-Disk is very similar to what is above, except
that the volser is TESTRS and it is only 10, (x'0A'), cylinders in size!
(If anybody wants to see the data from the formatted T-Disk, let me know
and I will post it.)
It looks to me like the VM install volumes are formatted correctly and
that ICKDSF CPVOLUME also formats them correctly, (Format-5 DSCB shows no
free cylinders/tracks available). My guess is that DITTO may be looking
for Format-1 and Format-3 DSCBs, (they describe space used by allocated
datasets), and not finding any on a VM volume, it is a assuming that the
entire volume is free. z/OS will not allocate a dataset on a properly
formatted VM volume, regardless of what DITTO shows.
--
Dale R. Smith
"Always acknowledge a fault. This will throw those in authority off
their guard and give you an opportunity to commit more."
- Mark Twain
On Mon, 21 Jan 2008 16:35:57 +0200, Kris Buelens <[EMAIL PROTECTED]>
wrote:
>I sent Dale a TYPE of two cyl 0 head 0 (one of z/VM 5.3 and one of an
>"ICKDSF CPVOL"-ed Tdisk)
>--
>Kris Buelens,
>IBM Belgium, VM customer support
>========================
=========================
========================
