Can not read old CompacTape II tapes

[Jur van der Burg]

Make sure that you have a real TZ87 tape unit, and not a TZ87N which does not understand TK70.

If you get have a TZ87 and you get a parity error then the only hope is another tapeunit.
No mount command will help.

Jur.

[Jon Pinkley]

Lluis,

When the tape gets loaded, what LEDs on the drive light up? Normally, the LEDs will indicate what type of tape and at what density it was initialized. A real TZ87 will have an LED labeled TK50/TK70 at the top left of the drive (see page 4 of http://deathrow.vistech.net/~cvisors/DEC94MDS/xtz87ina.pdf )

Are you sure there is data written to the tapes? Do you have any tapes with VMS distribution or patches on them? You stated that "all the tapes" (in the cabinet?) had the same problem. Do you have access to any known "good" tapes with data that you can try in the drive?

As Jur stated, if you do a mount/for followed by a dump/blo=cou=1 and get a parity error, the drive isn't seeing any good data on the tape. That could be because the tape was bulk erased as suggested by Steven Schweda, or it is possible the TK50/TK70 part of the drive is broken (or absent if a TZ87N or a DLT2000)

Backup/list will error with BACKUP-F-LABELERR apparently without reading the tape if getdvi(dev,"VOLNAM") is "". To get it to actually try to read the tape you have to specify /ignore=label

I would guess that the commands:

$ mount/for mkc500
$ backup/log/list=info.txt mkc500:*.*/save/ignore=label

would result in a parity error as well.

p.s. mount/foreign and mount/nolabel are equivalent.

$ pipe verb mount | search/nowin sys$pipe foreign,nolabel
qualifier FOREIGN, syntax=MOUNT_WO_LABEL, nonnegatable
qualifier NOLABEL, syntax=MOUNT_WO_LABEL, nonnegatable
disallow (LABEL and NOLABEL)
disallow (MULTI_VOLUME and (not FOREIGN or LABEL))
disallow (POOL and (CACHE or SYSTEM or GROUP or FOREIGN))
disallow ((FOREIGN or NOLABEL) and (ACCESSED or
disallow (FOREIGN and SHADOW)
disallow ((FOREIGN or NOLABEL) and (ACCESSED or

Good Luck,

Jon

[Lluís Aunós]

Hoff,
we have all backups on DLTIII and DLTIV, and these old backups on CompacTapeII. Now, we are changing to LTO-4, and we want to move all data to new format LTO-4.

Jon,
the model tape is â TZ87 â TAâ , and the Ownerâ s Manual (Digital part number EK-OTZ87-OM.A01), pag. 1-7, specifies that CompacTape II/TK70 is â Read onlyâ . I have tested to read DLTIII tapes and it is ok, but these old tapes is not possible.

we have a when the tape gets loaded, there are 4 leds light up:
- TK50/TK70
- Write protected
- Tape in use
- Use cleaning tape
state â use cleaning tapeâ is because unit canâ t read. Iâ ve put the cleaning tape five times, and if I put a DLTIII tape, the unit can read it.

execution of mount and back commands:
----------
$ mount/for mkc500:
%MOUNT-I-WRITELOCK, volume is write locked
%MOUNT-I-MOUNTED, mounted on _DIDO$MKC500:
$ backup/log/list=info.txt mkc500:*.*/save/ignore=label
%BACKUP-F-LABELERR, error in tape label processing on MKC500:[000000]*.*;
-BACKUP-F-NOTANSI, tape is not valid ANSI format
----------

Thank you,

Lluís

[Lluís Aunós]

I think it's a hardware problem, so we'll open a new case with HP.

Thank you.

Lluis

[Richard Jordan]

Lluís
it is entirely possible that the tapes themselves are the problem. We've had similar issues with a number of 15-20 year old TK50 cartridges, even "used only once" tapes like VMS distributions. Known good drives (TZ30 and TK50, which were the native drives for these) will often fail to read tapes this old, and the attempts often further damage the media; the oxides are 'loose' on the tape backing, and can even come off while being unspooled inside the cartridge. We saw this with 9-track magtapes back in the day too; the oxide layer would be partially stripped off by the back of the tape that was the next layer out.

With magtapes, you could try a low level bake in an oven; that would allow one attempt to read with a lower probability of the oxide layer stripping. We never found out if this was useful to try, or safe to attempt, with TK cartridges, so we didn't try.

We ended up trying to do raw copies off the tapes, and managed to recover less than 10% of the ascii data (source code) from them.

When I asked about this online, two programs were recommended; one called VMSTPCE which I believe was either on a freeware disk or a DECUS distribution (did not work for us, but who knows in your case), a tape utility provided by Emulators International (now part of softresint.com) that was supposed to be able to read VMS tapes via a SCSI tape drive attached to a PC (we didn't try this because our tapes were too far gone).

Finally Glenn Everhart recommended (and we tried) a simple program issuing QIO reads to the foreign mounted tape, with IO modifiers set to disable retries and error checking. Thats what got us the little bit we did recover, and only because it was recognizable ASCII in the savesets. No binary data was recovered. And sorry but that program was not saved.

Best of luck to you.

[Volker Halle]

Lluis,

note that there are also professional data recovery services, which might be able to help you recover data from those old TK50s, but they'll charge you - of course.

Volker.

[GuentherF]

I doubt that this is a problem with your current tape drive. A parity error returned by MKDRIVER means a SCSI Media Error was detected.

As mentioned earlier an older TK drive might be able to read these cartridges.

/Guenther