:SEL-RESTORE THUSTK
Restore from F)ull/Account, I)ncremental or T)ransaction log (f/i/t):F
account name on tape:WBUK
file name:STK
mds 2
dm 3
sql 1074
WBEU 49414
WBUK 51968
VID.DEBT 51999
VID.DEBT 52000
peqs.backup 52565
peqs.backup 52566
tape format err @ rs.skip.index:0C9
err "d" segment @1Es0ps14h9v0s0b4101Ta2925h5895V 5a3150h5895V
0a3285h5895V @ rs.dq.type:0B5
enter level (1=acct,2=dict,3=data)? 1
tape format err @ rs.skip.index:0C9
tape format err @ rs.skip.index:0C9
err "d" segment ISCOUNT ........TAX......... GROSS$ ORD.^TYPE NO.
NO.... NAME............. @ rs.dq.type:0B5
enter level (1=acct,2=dict,3=data)? 1
tape format err @ rs.skip.index:0C9
tape format err @ rs.x.dcode:046
spooler 249
peqs 250
err "d" segment er for that attribute.^^ @ rs.dq.type:0B5
enter level (1=acct,2=dict,3=data)? 1

A LITTLE more information would be useful - like the platform and
revision you are using, and that was used to generate the tape.

I'm guessing (from the peqs.backup file) that you are on a D3 platform,
and yet the files you show as being created in the newly restore MBUK
account don't have a modulo?

Do you know how the "peqs.backup" file was created & populated? as it
would appear to contain some dodgey data - if you can repeat the backup
& exclude this file (DX) then I would think you will have success (I
don't think PEQS items are meant to be restored, but am probably wrong
..... maybe just an unfortunate name choice? [doubt it])

Anyway, I think you have gone past the account on the tape (pseudo? how
transferred? binary FTP?), and the restore processor has found another
file (D pointer) and wants you to indicate at what "level" it is ...
account, dict or data level file

HTH

Ross Ferris
Stamina Software
Visage > Better by Design!

jmemb... (AT) hotmail (DOT) com wrote:

The restore is attempting to skip an index.

Data is placed onto tape with some control characters to tell the system
what any particular bit of info is. Every "data" or "dict" item has a file
number associated with it to say "I go into THAT file".

Whenever there is a problem it can't recover from on its own, it will stop
(tape format error: expected data not found; or unexpected data found at the
wrong place) and then it gives you a chance to skip to the next "good" data
it can find.

The choices, acct, dict, data mean just that: do you want to skip the bad
data and everything else up to the next data pointer, the next dict pointer
or the next acct pointer. Since you are doing a selective restore of data
from a particular file, I would suggest that you don't want to skip to the
next acct pointer. That would skip everything else in THIS account and
that's not what you want. Try 3, to skip to the next dictionary definition.

The error itself is probably caused by attempting to save indexes on the
file save along with the data. This can be a major time saver, since you
don't have to generate the indexes after everything is done. On the other
hand, as you see, it can be a major PITA if there's any kind of problem.

Other suggestion: if possible try this:

Sel-Restore MD Crapola
F
acctname: Craporama
file: Total_foul_up

And watch for your file to go by. Note the file "number". Then, once
you're past your STK file, kill the process, rewind the tape and do:

SEL-RESTORE THUSTK (N

and enter that file number when it asks. That way, it skips a lot of stuff,
but reads everything looking for that file number.

If it's possible to recreate the save (from another machine?), use the
option that says 'don't save indexes'.

Mark Brown

<jmembrey (AT) hotmail (DOT) com> wrote