> I'm posting this message again, since i had no answers to the
> previous post and i couldn't solve this problem using the tools
> supplied for this (gbak/gfix) nor following the IBPhoenix docs on
> how to restore corrupt databases.
>
> I'm feeling a little bit unconfortable because it's the first
> problem i had with FB/IB that i couldn't solve making a backup and a
> restore or using gfix, and because it makes 15 days that my customer
> is working without a backup of his data.
>
> When i try to backup this database, i have the following error:
>
> gbak: writing data for table REM1
> gbak: ERROR: message length error (encountered 58, expected 54)
> gbak: ERROR: gds_$receive failed
> gbak: Exiting before completion due to errors
>
> If i run gfix it doesn't report nothing.
> I tryed to follow all steps of IBPhoenix doc, without success.
> The server is a IBM eServer xSeries 200, working with no-breaks, so
> it's not supposed to have problems caused by abnormaly server
> shutdowns.
> It's running Linux RH 9.0, was running FB RC6 but now i installed
> RC7, and the problem persists.
>
> Apart this problem, the database seems to be ok, the users are
> working on it without problems.
>
> I'd like to have a clue on what can cause this type of erros and
> what can i do to solve it.
>
> Thanks again.

I would recommend creating another database, and pumping metadata and data
to the new shell. This will mean some downtime (maybe only short if you
practice on an old backup).

Are you backing up the database regularly? Do you test the restore
regularly?

This sort of corruption is rare - maybe there is something you are doing
like taking tape backups of the file?

Alan