Discussion:
Problems with tar and Seagate DAT drive
Matthew Henkler
2001-01-21 19:02:35 UTC
Permalink
I'm trying to back up a large (2.7gb) directory to tape with verify
using the command

tar -cWvf /dev/st0 <DIRECTORY> > /root/tape_log 2> /root/tape_error

And, upon completion, I check the tape_log, and it lists all the files
as being backed up, but the tape_error file has the following lines:

...
tar: /dev/st0: Cannot read: Input/output error
tar: Too many errors, quitting
tar: Error is not recoverable: exiting now

These errors do not appear until the verify section, and then, only
towards the last few files being verified.

My tape drive is a Seagate Scorpion 8 STD28000N external DDS2 drive. I
am running kernel version 2.4.0 with a Adaptec 2940U2W SCSI card and
everything seems to be terminated correctly.

I've tried this on various tapes, and it happens every time. If anyone
could shed any light on why this is happening and how I can fix it, I
would appreciate it.

Thanks,

matt henkler
Matthew Henkler
2001-01-23 05:39:27 UTC
Permalink
Well, when I went and backed up /home tonight (300 megs or so), it seems
to work fine, but when I backed up /etc (a separate job, maybe 10 megs),
the verify cycle fails on the 5th file from the end with the same errors
as before. I have all SCSI support built directly into the kernel, not
as modules, and no, my tapes are not write-protected. I wonder if it is
something to do with setting a buffer size somewhere.... I have always
been able to do full backups with verify without error using Backup Exec
under windows.

Thanks for the help,

matt henkler
The following message is a courtesy copy of an article
that has been posted to purdue.student-org.plug as well.
Post by Matthew Henkler
I'm trying to back up a large (2.7gb) directory to tape with verify
using the command
tar -cWvf /dev/st0 <DIRECTORY> > /root/tape_log 2> /root/tape_error
This won't fix the problem for you, but it will (should) help you
figure out where the errors are occuring with respect to stdout.
Post by Matthew Henkler
/root/tape_log 2> /root/tape_error
1>>/root/tape_log 2>>/root/tape_log (for bash, ksh, zsh)
or
Post by Matthew Henkler
&/root/tape_log (for csh and tcsh)
These both assume that /root/tape_log do not exist. Now, you can look
at /root/tape_log and see where error messages correspond with output
messages.
Post by Matthew Henkler
And, upon completion, I check the tape_log, and it lists all the files
...
tar: /dev/st0: Cannot read: Input/output error
tar: Too many errors, quitting
tar: Error is not recoverable: exiting now
These errors do not appear until the verify section, and then, only
towards the last few files being verified.
What happens if you try to do something smaller, like /home or /etc?
Post by Matthew Henkler
My tape drive is a Seagate Scorpion 8 STD28000N external DDS2 drive. I
am running kernel version 2.4.0 with a Adaptec 2940U2W SCSI card and
everything seems to be terminated correctly.
Just, out of curiosity, you do have SCSI tape suppor either builtin to
your kernel or through a module, don't you?
Post by Matthew Henkler
I've tried this on various tapes, and it happens every time. If anyone
could shed any light on why this is happening and how I can fix it, I
would appreciate it.
The tapes aren't write-protected are they?
Charles Fultz
fultz
Loading...