SUMMARY: cannot read spaced tapes

From: Ayrton Sargusingh (asargusi@sbrsim.ed.dreo.dnd.ca)
Date: Thu Jun 15 1995 - 18:27:52 CDT


My original post:

> Date: Mon, 12 Jun 95 12:23:33 EDT
> Subject: cannot read spaced tapes
> Content-Length: 1731
>
> Dear Sun managers:
>
> I am in an awful bind. I hope someone can assist me very soon!
>
> Problem Description
> -------------------
> I have a problem reading my backup tapes. I have been using dump to
> backup several volumes onto an 8mm 8505XL device (24 volumes). In
> the past this has worked fine. No recent changes have been made to
> the SCSI configuration or device configuration. Here is my dump
> command (part of my backup script) for your reference:
>
> # dump 0ubdsf 126 54000 13000 /dev/nrst24 <volume>
>
> I can perform a status on the tape device, tar files to it, etc. The
> only thing I can't seem to do is to space to the next volume. I have
> attempted this on multiple tapes. My tapes are not corrupted since I
> regularly verify backup tapes, and I just recently restored a volume
> last week. This is something else.
>
> I have attempted to space the tape in two ways.
>
> # mt -f /dev/rst24 rew
> # mt -f /dev/nrst24 fsf 22
> /dev/rst24 fsf 22 failed: I/O error
>
> # mt -f /dev/rst24 rew
> # restore -ifs /dev/nrst24 22
> ioctl MTFSF: I/O error
>
> What gives! I can restore the first volume but once I try to space to
> the next, I get the errors above. The only thing that happens is that
> the drive light will flash for several seconds or minutes before giving
> the error. Rebooting didn't clear the problem either.
>
> How sure am I that there are 25 volumes archived on those tapes? Dead
> sure!
>
> System Configuration
> --------------------
> Solaris 1.2, SPARC-20, 8505XL 8mm SCSI tape device, 160m 8mm tapes,
> tape in compressed format.
>
> -------
> Ayrton Sargusingh, SBR System Manager, RSD | phone: 613-998-2932
> Defence Research Establishment Ottawa (DREO) | fax: 613-998-4560
> 3701 Carling Ave., Ottawa ON, Canada K1A 0Z4 | email: asargusi@sofkin.ca

Thanks to all who responded, including the following people. I hope I
didn't forget anyone:

pluta@cgc1.mcae.mke.ab.com (Rick Pluta)
mrs@cadem.mc.xerox.com (Mike Salehi)
nobroin@esoc.esa.de (Niall O Broin x 3619)
Paul_Lupa@galileo.tracor.com (Paul Lupa X4184)
Ian MacPhedran <macphed@engr.USask.Ca>
mike@trdlnk.com (Michael Sullivan)
Ross.Stocks.INSDRS01@nt.com

Resolution
----------
A few people mentioned running tcopy to see how many savesets were
stored on the tape (Ian and Mike Sullivan). Good point, I didn't think
of that, but the answer was 25. That means all the save sets were there
intact.

Others said the tape heads musthave been dirty. That wasn't it either.
Some thought nrst0 and rst0 were interchanged. No so. There were no
messages in /usr/adm/messages, /var/log/syslog, or on the console.

A couple of people had similar experiences (Mike Salehi & Ross).
Ross had this to say:

        I once had a strange situation very similar to yours, possibly
        identical to yours, but on a 8500 drive. I ended up doing a mt
        eom, then mt nbsf back to the desired file. I have no idea
        what the problem was. Future attempts to use the same tape
        were problem free.

Unfortunately, I didn't have a chance to try any of these since
everything magically worked this morning! I must have prayed hard last
night 'cause miracles do happen. I wish I knew what went wrong and what
caused it to behave like this. This is serious 'cause you're almost
ready to trash all your valuable backups or overwrite them.

Follow-up Question
------------------
If no one has had found out why this happens, this is a problem that
should be addressed to Sun or Exabyte. Does anyone have an e-mail
address for Exabyte?

-------
Ayrton Sargusingh, SBR System Manager, RSD | phone: 613-998-2932
Defence Research Establishment Ottawa (DREO) | fax: 613-998-4560
3701 Carling Ave., Ottawa ON, Canada K1A 0Z4 | email: asargusi@sofkin.ca



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:10:26 CDT