Backup Central

Wrote blk_block=320000, dev_blk_num=3517 VolBytes=77,656,227,840
rate=11172.0 KB/s
16:13:03 Flush block, write EOF
05-Mar 16:13 btape: End of Volume "TestVolume1" at 80:472 on device
"VXA-172" (Tape0). Write of 245760 bytes got -1.
05-Mar 16:13 btape: btape Error: Error writing final EOF to tape. This
Volume may not be readable.
../../stored/dev.c:1687 ioctl MTWEOF error on "VXA-172" (Tape0).
ERR=Input/output error.
btape: ../../stored/btape.c:2343 Last block at: 80:471
this_dev_block_num=472
btape: ../../stored/btape.c:2357 Block not written: FileIndex=2372212
blk_block=320472 Size=245696

btape: ../../stored/btape.c:2377 End of tape 80:472.
VolumeCapacity=77,772,226,560. Write rate = 11169.4 KB/s
05-Mar 16:13 btape: End of medium on Volume "TestVolume1"
Bytes=77,772,226,560 Blocks=316,456 at 05-Mar-2007 16:13.
05-Mar 16:13 btape: 3301 Issuing autochanger "loaded? drive 0" command.
05-Mar 16:16 btape: 3302 Autochanger "loaded? drive 0", result is Slot 2.
Wrote Volume label for volume "TestVolume2".
05-Mar 16:18 btape: Wrote label to prelabeled Volume "TestVolume2" on
device "VXA-172" (Tape0)
05-Mar 16:18 btape: New volume "TestVolume2" mounted on device "VXA-172"
(Tape0) at 05-Mar-2007 16:18.
Done writing 0 records ...
Wrote End of Session label.
Could not create state file: /tmp/btape.state ERR=No such file or directory
16:18:46 Done filling tapes at 0:13. Now beginning re-read of first tape ...

05-Mar 16:20 btape: Ready to read from volume "TestVolume1" on device
"VXA-172" (Tape0).
Rewinding.
Reading the first 10000 records from 0:0.
05-Mar 16:21 btape: End of file 1 on device "VXA-172" (Tape0), Volume
"TestVolume1"
10000 records read now at 1:1334
Reposition from 1:1334 to 80:471
Reading block 471.

The last block of the first tape matches.

Reposition from 0:2 to 0:11
Reading block 11.

The last block on the second tape matches. Test succeeded.

Hi,

On 3/5/2007 10:53 PM, Brian Debelius wrote:
Wrote blk_block=320000, dev_blk_num=3517 VolBytes=77,656,227,840
rate=11172.0 KB/s
16:13:03 Flush block, write EOF
05-Mar 16:13 btape: End of Volume "TestVolume1" at 80:472 on device
"VXA-172" (Tape0). Write of 245760 bytes got -1.
05-Mar 16:13 btape: btape Error: Error writing final EOF to tape. This
Volume may not be readable.
../../stored/dev.c:1687 ioctl MTWEOF error on "VXA-172" (Tape0).
ERR=Input/output error.
btape: ../../stored/btape.c:2343 Last block at: 80:471
this_dev_block_num=472
btape: ../../stored/btape.c:2357 Block not written: FileIndex=2372212
blk_block=320472 Size=245696

btape: ../../stored/btape.c:2377 End of tape 80:472.
VolumeCapacity=77,772,226,560. Write rate = 11169.4 KB/s
05-Mar 16:13 btape: End of medium on Volume "TestVolume1"
Bytes=77,772,226,560 Blocks=316,456 at 05-Mar-2007 16:13.

End-of-tape detection seems not to work correctly. I haven't seen a VXA
drive for a really long time now, but I'm sure it should be possible to
get the end of tape notification in time.

Might also be a tape error, though.

05-Mar 16:13 btape: 3301 Issuing autochanger "loaded? drive 0" command.
05-Mar 16:16 btape: 3302 Autochanger "loaded? drive 0", result is Slot 2.
Wrote Volume label for volume "TestVolume2".
05-Mar 16:18 btape: Wrote label to prelabeled Volume "TestVolume2" on
device "VXA-172" (Tape0)
05-Mar 16:18 btape: New volume "TestVolume2" mounted on device "VXA-172"
(Tape0) at 05-Mar-2007 16:18.
Done writing 0 records ...
Wrote End of Session label.
Could not create state file: /tmp/btape.state ERR=No such file or directory

This one looks more interesting... are you sure your user account may
write to /tmp, and there is no file btape.state owned by someone else?

16:18:46 Done filling tapes at 0:13. Now beginning re-read of first tape ...

05-Mar 16:20 btape: Ready to read from volume "TestVolume1" on device
"VXA-172" (Tape0).
Rewinding.
Reading the first 10000 records from 0:0.
05-Mar 16:21 btape: End of file 1 on device "VXA-172" (Tape0), Volume
"TestVolume1"
10000 records read now at 1:1334
Reposition from 1:1334 to 80:471
Reading block 471.

The last block of the first tape matches.

Reposition from 0:2 to 0:11
Reading block 11.

The last block on the second tape matches. Test succeeded.

Looks ok up till here, alsthough I would investigate the errors reported.

It would be helpful if you told us which OS, which Bacula version, and
which tape drive exactly you use. Also, try another tape to test for end
of tape detection problems.

Arno


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Bacula-users mailing list
Bacula-users < at > li...
https://lists.sourceforge.net/lists/listinfo/bacula-users

--
IT-Service Lehmann al < at > it...
Arno Lehmann http://www.its-lehmann.de

Arno Lehmann wrote:
Hi,

On 3/5/2007 10:53 PM, Brian Debelius wrote:

Wrote blk_block=320000, dev_blk_num=3517 VolBytes=77,656,227,840
rate=11172.0 KB/s
16:13:03 Flush block, write EOF
05-Mar 16:13 btape: End of Volume "TestVolume1" at 80:472 on device
"VXA-172" (Tape0). Write of 245760 bytes got -1.
05-Mar 16:13 btape: btape Error: Error writing final EOF to tape. This
Volume may not be readable.
../../stored/dev.c:1687 ioctl MTWEOF error on "VXA-172" (Tape0).
ERR=Input/output error.
btape: ../../stored/btape.c:2343 Last block at: 80:471
this_dev_block_num=472
btape: ../../stored/btape.c:2357 Block not written: FileIndex=2372212
blk_block=320472 Size=245696

btape: ../../stored/btape.c:2377 End of tape 80:472.
VolumeCapacity=77,772,226,560. Write rate = 11169.4 KB/s
05-Mar 16:13 btape: End of medium on Volume "TestVolume1"
Bytes=77,772,226,560 Blocks=316,456 at 05-Mar-2007 16:13.


End-of-tape detection seems not to work correctly. I haven't seen a VXA
drive for a really long time now, but I'm sure it should be possible to
get the end of tape notification in time.

Might also be a tape error, though.


05-Mar 16:13 btape: 3301 Issuing autochanger "loaded? drive 0" command.
05-Mar 16:16 btape: 3302 Autochanger "loaded? drive 0", result is Slot 2.
Wrote Volume label for volume "TestVolume2".
05-Mar 16:18 btape: Wrote label to prelabeled Volume "TestVolume2" on
device "VXA-172" (Tape0)
05-Mar 16:18 btape: New volume "TestVolume2" mounted on device "VXA-172"
(Tape0) at 05-Mar-2007 16:18.
Done writing 0 records ...
Wrote End of Session label.
Could not create state file: /tmp/btape.state ERR=No such file or directory


This one looks more interesting... are you sure your user account may
write to /tmp, and there is no file btape.state owned by someone else?


16:18:46 Done filling tapes at 0:13. Now beginning re-read of first tape ...

05-Mar 16:20 btape: Ready to read from volume "TestVolume1" on device
"VXA-172" (Tape0).
Rewinding.
Reading the first 10000 records from 0:0.
05-Mar 16:21 btape: End of file 1 on device "VXA-172" (Tape0), Volume
"TestVolume1"
10000 records read now at 1:1334
Reposition from 1:1334 to 80:471
Reading block 471.

The last block of the first tape matches.

Reposition from 0:2 to 0:11
Reading block 11.

The last block on the second tape matches. Test succeeded.


Looks ok up till here, alsthough I would investigate the errors reported.

It would be helpful if you told us which OS, which Bacula version, and
which tape drive exactly you use. Also, try another tape to test for end
of tape detection problems.

Arno


Hi,

OS is Windows 2003r2
Bacula is v2.0.2 (director and sd are on Windows)
Tape is a VXA-172 (its the same as a VXA-320, just a firmware change to
handicap it to half capacity)

I get these state errors occasionally, and can't figure out why.

So, Why does the test say that it failed with the EOF write, but then at
the end, say that it was able to read the block, and that the test passed.

Hi,

On 3/6/2007 3:47 PM, Brian Debelius wrote:
... possible problems with end-of-tape detection with a VXA drive...
Hi,

OS is Windows 2003r2
Bacula is v2.0.2 (director and sd are on Windows)

Oh, windows... perhaps something to take to the developer list, too. In
any case, when you encounter serious problems with the windows versions,
try to contact Rober Nelson, too.

Tape is a VXA-172 (its the same as a VXA-320, just a firmware change to
handicap it to half capacity)

Ok. Perhaps someone else has a working config for these drives, so that
you could compare your settings?

I get these state errors occasionally, and can't figure out why.

In that case I'd guess the pre-end-of-tape notifications come too
shortly before the actual end of tape, so that sometimes the last block
Bacula writes only just fits onto the tape. Or something :-)

So, Why does the test say that it failed with the EOF write, but then at
the end, say that it was able to read the block, and that the test passed.

Because the end-of-tape condition does not definitely indicate that data
could not be written. Rather, the tape drive thus indicates that it's
quite sure the tape will be at the end really soon.

I can't explain it much better, but looking through the unix manual
pages for the tape interfaces, that's what I distilled out of it once.

The problem is that a tape drive can't know for sure when exactly a tape
will be full. That, and large blocks of data to write suggest the
strategy to report a nearing end of tape condition before data can not
be writen anymore.

Thus, a last block can be written to tape, but the write returns an
error because the tape drive then is really sure it can't put more data
onto the tape, and the final EOF mark either will not be written at all
or returns the error state "just in case".

This is the reason for Bacula checking the last block. When that check
is ok, it's not really a problem there is no filemark at the end of tape
because Bacula will never try to read beyond that tape position.

I hope that helps, and I hope I've got this more or less right, and
finally I hope windows handles tape drives similar :-)

Arno


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Bacula-users mailing list
Bacula-users < at > li...
https://lists.sourceforge.net/lists/listinfo/bacula-users

--
IT-Service Lehmann al < at > it...
Arno Lehmann http://www.its-lehmann.de