problem in backup ( fbackup)

[Shashikiran_1]

HI ,

I have a problem in our backup process.
We have mooved DLT Drive from K class server to a D class server Since there was some problem in k class machine.
It was working fine for 2-3 weeks( 3 full backups).
Now it is giving following error after 3-4 hours. I have changed the Tape cleaned the drive but still it is giving the same problem.

fbackup(3012): WRITE ERROR while writing checkpoint record, at media record 644096
fbackup(3102): attempting to make this volume salvagable
fbackup(3105): writing 2 EOFs and rewinding the tape
fbackup(3106): please mount a good tape
fbackup(3310): enter 'yes' when volume 1 is ready on /dev/rmt/c2t5d0BEST,
or 'no' to discontinue


Command we are using is fbackup -0nv -g /UnixBackup/fbackupfiles/graphs/G1 -f /dev/rmt/c2t5d0BEST

There is no problem in tape , Itried this for 4-5 nnew tapes.
could you please help me regarding this.

thanks
Shashikiran

[Dietmar Konermann]

I think you're clearly facing a hardware problem. Since the tapes seem to be ruled out, the next pice to check would be the drive itself, IMHO. Time to open a HW call to get the diag logs analyzed.

Best regards...
Dietmar.

[Jean-Luc Oudart]

Check this tread :
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=28572

or
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=178204

Regards
Jean-Luc

[clar_1]

Hi Shashikiran

Please try backing-up the same directories/filesystems using tar backup or cpio backup utilities.This would conclude your problem,if all these backup utilities report errors while backup,you can directly go for a DLT replacement/repair - whichever feasible to your hardware contract.

Best Regards
Jai

[Michael Schulte zur Sur]

Hi,

have you used a cleaning tape?

greetings,

Michael

[MarkSyder]

Hi,

Sorry to betray my AIX background, but I always apply the principles drummed into me by IBM engineers.

Clean the heads with a headcleaner a minimum of three times before you insert a brand new tape (a new tape in a dirty drive will be ruined immediately).

If there is still a fault place a hardware call. IBM will not accept a hardware call unless you have done as outlined above. I do not know if HP use the same guidelines, but I think they are good ones and I still apply them eight months after I last worked on an AIX server.

Mark Syder (like the drink but spelt different)

[Emiel van Grinsven]

Hi

How about the patch level on the system you placed the drive in?

[Steven E. Protter]

A cleaning tape has made this exact message go away for me. Eventually however the entire drive bit the big one.

SEP

[T. M. Louah]

please make sure that tar works, simple test send 1 file to tape :
# tar cvf /dev/rmt/0m /etc/hosts
now display tape content by:
# tar tvf /dev/rmt/0m
if this works the next thing to consider is patching fbackup command.

regards,
T+