cancel
Showing results for 
Search instead for 
Did you mean: 

restoring - "tapemark or blank tape encountered"

ianG
Level 5

Hi there,

Encountering this error message when attempting to restore from a few tapes (on a Quantum i6000). Both master and media server are on version 7.0.1, running windows 2k3.

Trying a 'bpmedialist -mcontents -m 15832S -L' returns the following in activity monitor:

 

15-Aug-2011 13:53:51 - requesting resource 15832S
15-Aug-2011 13:53:51 - granted resource  15832S
15-Aug-2011 13:53:51 - granted resource  I6K003
15-Aug-2011 13:54:02 - started process bptm (pid=2180)
15-Aug-2011 13:54:02 - started process bptm (pid=2180)
15-Aug-2011 13:54:02 - mounting 15832S
15-Aug-2011 13:54:52 - mounted 15832S; mount time: 0:00:50
15-Aug-2011 13:54:58 - Error bptm (pid=2180) tapemark or blank tape encountered reading media header, drive index 10
media read error  (85)

 

Have attached the bptm log from the media server for this operation - 081511_00001.txt

 

We get the same error message when trying a restore:

 

15-Aug-2011 13:09:13 - begin Restore
15-Aug-2011 13:09:24 - number of images required: 1
15-Aug-2011 13:09:24 - media needed: 15832S
15-Aug-2011 13:09:51 - restoring from image virtapp172s_1296273001
15-Aug-2011 13:09:56 - connecting
15-Aug-2011 13:09:57 - started process bptm (pid=5572)
15-Aug-2011 13:09:57 - mounting 15832S
15-Aug-2011 13:10:01 - connected; connect time: 0:00:00
15-Aug-2011 13:09:57 - requesting resource 15832S
15-Aug-2011 13:09:57 - granted resource  15832S
15-Aug-2011 13:09:57 - granted resource  I6K003
15-Aug-2011 13:10:55 - Error bptm (pid=5572) tapemark or blank tape encountered reading media header, drive index 10
15-Aug-2011 13:10:56 - Error bptm (pid=7428) The following files/folders were not restored:
15-Aug-2011 13:10:57 - Error bptm (pid=7428) UTF - /E/BE89DD/AAB1DF/A5D800/NT.08C
15-Aug-2011 13:10:58 - restored from image virtapp172s_1296273001; restore time: 0:01:07
15-Aug-2011 13:11:09 - Warning bprd (pid=6796) Restore must be resumed prior to first image expiration on 28/01/2012 03:50:01
15-Aug-2011 13:11:18 - end Restore; elapsed time 0:02:05
the restore failed to recover the requested files  (5)

 

Had thought initially it was something to do with buffer sizes, but I dont think this is the case now - read on this forum somewhere that the tapes buffer size is automatically detected, so the time spent playing with the SIZE_DATA_BUFFERS files has been pretty fruitless.

Any ideas?

1 ACCEPTED SOLUTION

Accepted Solutions

ianG
Level 5

Thanks guys, we got to the bottom of it :)

 

The libraries were configured to encrypt the tapes, and during a repartition of the library a few months ago, this encryption configuration was lost. So when the libraries then went to read the older encrypted tapes they just appeared to be blank/corrupted. Just needed to reconfig the connection to the encryption servers, and these tapes were readable again.

Thanks again all, your input was was much appreciated

View solution in original post

10 REPLIES 10

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

I found a TN that points to problematic/outdated firmware on tape drive and/or library causing similar errors:      http://www.symantec.com/docs/TECH51486   

Double-check HCL for recommended firmware on your tape drives and library:   http://www.symantec.com/docs/TECH76495     

ianG
Level 5

Thanks for the reply Marianne

 

We're using a Quantum i6000, on f/w 607A.GS05301.

The drive we're having issue with is a HP Ultrium, LTO4 on f/w H62Z (same issue however is experienced on one of our LTO5, f/w I3HZ)

 

Worryingly, neither f/w appears to be be listed in the second document you linked

Nicolai
Moderator
Moderator
Partner    VIP   

Have these tape been expired or relabeled lately to release new media for backup ?

It *cloud* be the tapes has been overwritten by re-labeling them. This would put a empty header in the beginning of the tape. The tapes however hold valid backup images.

kbytes = 1342166651, NumberOfImages = 24, ValidImages = 24

See info about tape marks in  : http://www.symantec.com/docs/TECH143890

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

The firmware on your devices seem to be newer than the tested versions in the HCL, but that should not be a problem. The intro section under 'Tape Libraries' section says:

" The firmware versions listed below represent the firmware that has been tested. Firmware versions that are newer than what is listed are also supported. This means that if a newer firmware version is causing a problem, Symantec will work with the hardware vendor to resolve the problem. Symantec and most hardware vendors listed are members of TSANet." The same statement of support for newer firmware is repeated in the Tape Drive section.

As per Nicolai's post, the other possibility is that the tapes could have been re-labelled/erased on some other environment. This would not be allowed in this environment, as EMM contains info about valid images on the tape.
You will know if the tapes are at  any stage removed from the current environment....

ianG
Level 5

Thanks for the replies guys

These tapes havent left the tape library, and have only been written to via this environment - so shouldnt be overwritten/relabelled

 

There was a f/w upgrade of the library - it used to be an i2000, but had an inplace upgrade to the i6k, at which point we added in the lto5 drives. We're following up with the vendor to see if this may be related. I believe it may be, as some tapes are reading fine, and other (slightly older) appear to suffer this issue.

 

I see theres a number of ways to check the header info on *nix machines (manually mounting, and reading off the header) - is there any similar process for windows?  Can a corrupted header be recreated?

ianG
Level 5

Oh, we also upgraded from NBU 6.5.5 to 7.0.1 in this time.

 

The tapes themselves are not multiplexed

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Please log a support call with Symantec so that the two vendors (Symantec and Quantum) can work together as per this statement in the HCL:

" ....  if a newer firmware version is causing a problem, Symantec will work with the hardware vendor to resolve the problem. "

Please let us know - it might be a lengthy process.....

ianG
Level 5

Yep, calls open with the pair of them - just at the log gathering stage at the moment, so yeah, dont expect this to be a quick one alright!

Nicolai
Moderator
Moderator
Partner    VIP   

Has the 2003 box been patched to support a SCSI block size larger than 64K ?.  Both the OS and the miniport driver for the HBA need to support large SCSI IO. The OS support is included with SP1 but the miniport driver need to support it as well.

http://www.symantec.com/docs/TECH64105

ianG
Level 5

Thanks guys, we got to the bottom of it :)

 

The libraries were configured to encrypt the tapes, and during a repartition of the library a few months ago, this encryption configuration was lost. So when the libraries then went to read the older encrypted tapes they just appeared to be blank/corrupted. Just needed to reconfig the connection to the encryption servers, and these tapes were readable again.

Thanks again all, your input was was much appreciated

View solution in original post