frecover problem with LTO Ultrium

[BRUEL Pierre]

We have problem in restoring a fbackup tape, made with a LTO-1 drive with default fbackup options.
Sometimes, the restore works, but sometimes, it doesn't work and we get this message :
frecover(2114): read error from input device (I/O error)
frecover(2114): read error from input device (I/O error)
frecover(2113): unable to resync backup media.
I've seen identical messages already posted in the forum, but without any definitive response.
Is it important to adapt the parameters in the /etc/sam/br/fbackup_config ?

Thanks for your help !

[Eric Antunes]

Hi,

Check if you have this patch installed:

#swlist -l patch PHCO_28998

Best Regards,

Eric Antunes

[BRUEL Pierre]

Yes this patch is installed !

[Eric Antunes]

And is the tape ok?

#ioscan -fnCtape

[Managed Services]

Hi

I'm having the exact same problem. fbackup reports no errors, but when running a verification it gives the message:
frecover(2114): read error from input device (I/O error)
frecover(2114): read error from input device (I/O error)
frecover(2113): unable to resync backup media.

As it is a nightly backup I have noticed a pattern over the past 3 weeks (was using cpio before that) where an average of 3 out of 7 backups fail to verify. The tapes are on a strict 2 week rotation and they are changed every 6 months.

I checked if the afforementioned patch exists and it does not. Also, when running ioscan on the tape I get the following:
Class I H/W Path Driver S/W State H/W Type Description
=====================================================================
tape 0 0/0/1/0.3.0 stape CLAIMED DEVICE HP C1537A
/dev/rmt/0m /dev/rmt/c0t3d0BESTn
/dev/rmt/0mb /dev/rmt/c0t3d0BESTnb
/dev/rmt/0mn /dev/rmt/c0t3d0DDS
/dev/rmt/0mnb /dev/rmt/c0t3d0DDSb
/dev/rmt/c0t3d0BEST /dev/rmt/c0t3d0DDSn
/dev/rmt/c0t3d0BESTb /dev/rmt/c0t3d0DDSnb

Any ideas?

[Managed Services]

I just realised that the patch is for version 11.11 and we're running version 11.00...

Also, forgot to mention the tape drive is a DDS.

[Eric Antunes]

Managed Services,

Check the tape state as root:

#ioscan -fnCtape

Eric

[Eric Antunes]

If the tape is ok, check post here the /etc/sam/br/fbackup_config file...

[Eric Antunes]

If the tape is ok, post (not check as previous post) here the /etc/sam/br/fbackup_config file...

[Managed Services]

Eric,

I run the backup/verification routines via a script in crontab, I don't use SAM. Here are the contents of the config file:
blocksperrecord 512
records 64
checkpointfreq 1024
readerprocesses 6
maxretries 5
retrylimit 5000000
maxvoluses 200
filesperfsm 2000

Thanks
- Peter

[Eric Antunes]

Hi,

I've this on the config file:

blocksperrecord 32
records 32
checkpointfreq 32
readerprocesses 2
maxretries 5
retrylimit 5000000
maxvoluses 100

And no problem at all....

[Managed Services]

Found the problem!

It was the tape drive itself. HP went out today and replaced it. I tested it several times over with tapes that had failed on other days. All was well.

I do have a question about the exit error code, which frecover gives. Does it give out an exit code 0? I have a variable in the backup script to pick this up ($?) and it simply is not seeing the exit code as anything other than 0... Any idea?

The initial command is:
/usr/sbin/frecover -r -vN -f $TAPE_DEV > $TAPELIST 2>&1
rc2=$?

Followed by:
if [ $rc2 != 0 ]
then
VBACSTAT1="verification_incomplete"
else
VBACSTAT1="backup_success"
fi

etc.

However this has not reported the backup as failed.

Thanks
- Peter