On Mon, Sep 15, 2008 at 12:29:34PM -0400, Clark, Patti wrote:
> I had a saveset backing up last night when we took a power hit.  The UPS
> failed and the backup server et al went down.  As a consequence, the
> tape that was being written to produces an error:
> Fsf 84: Input/output error
> 
> I understand that this is because there wasn't an EOF marker created.  
> 1.  Is there any method of fixing this tape for continued appending?

If you really wanted to, I would be tempted to try deleting the
saveset(s) that were writing to the tape at the time.  Should be in the
database as (i)ncomplete savesets.  Once deleted, it shouldn't try to
forward to past those savesets, but just past the last good one.

> 2.  Is there any hope of recovery for the previous content if needed?

I don't think you can ask Networker to do a recovery of a partial
image.  But if the data isn't overwritten, and you're on unix, you can
have scanner read the data and pipe that into uasm for recovery to a
filesystem.  That should work until it hits the end of the data stream.

> 3.  Or do I need to take corrective steps and perform a full backup for
> CYA?

My first temptation would be to validate that I could read the last
savesets on the tape that were not in use, then mark the tape full and
let it expire normally.  

If there were a lot of tape capacity left and I was really cheap, I'd be
tempted to clone/stage the valid data on the tape to another tape.  Then
I could relabel this one.

Powering down the server is really unlikely to do anything to physically
damage the tape, so I don't see any reason it shouldn't be just as good
to use as before.

-- 
Darren

To sign off this list, send email to listserv AT listserv.temple DOT edu and 
type "signoff networker" in the body of the email. Please write to 
networker-request AT listserv.temple DOT edu if you have any problems with this 
list. You can access the archives at 
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER