sounds like the tape is bad.

Try to scan the tape from command line with verbosity using the scanner command. If the tape is bad you will get an error.

Thank You. I have checked the messages logs (1, 2, 3, 4) and these error messages are popping since June 2007 that I can trace back. This is coming for different tape drive (st1, st2, st3, st4).

st0: Failed to read 65536 byte block with 32768 byte read.
st0: Failed to read 65536 byte block with 32768 byte read.
st0: Failed to read 65536 byte block with 32768 byte read.
st1: Failed to read 65536 byte block with 32768 byte read.
st0: Failed to read 65536 byte block with 32768 byte read.
st3: Failed to read 65536 byte block with 32768 byte read.

Do you have any other storage nodes running? Maybe it's just one bad tape, that was mounted in every drive. You could try to verify which tape was mounted in which drive at the time of the error and see if it's the same tape.

This is a block size issue.

All you need to check is:
1. Are tape drive configured for the correct media type?
2. OS drivers for this media type are proper?
3. If nothing helps, set the block size in drives configuration. New media labelled with a set block size should not give this kind of messages.

Thank You.

I'm using "handler default" for block size under devices.

I'm able to run backups and restores from the tapes successfully. What is the recommended block size?

We are using LTO 2 Ultrium tape drives. LTO 2 tapes.

Red Hat Linux Enterprise 3 O S.

I think this would be beeter answered by the Hardware vendor but LTO-2 should work good with a 128 KB block size.

Please recheck with your hardware vendor on this.

Hi,

Do you have DDS configured on the drives i.e. the same drive is accessed by Linux and Windows system ?

I assume you have stinit.def defined too?

Thanks everyone. I haven't defined stinit.def

We are not using DDS.

I will check stinit.def settings

Thanks again