Veritas-bu

Re: [Veritas-bu] Netbackup 8X errors

2008-04-14 16:12:39
Subject: Re: [Veritas-bu] Netbackup 8X errors
From: "Ed Wilts" <ewilts AT ewilts DOT org>
To: "Floyd Endsley" <Floyd.Endsley AT sun DOT com>
Date: Mon, 14 Apr 2008 14:45:43 -0500
On Mon, Apr 14, 2008 at 2:16 PM, Floyd Endsley <Floyd.Endsley AT sun DOT com> wrote:
Is it acceptable to get 8X errors?

No.  This means that your job has failed.
 
Is there an  acceptable number before we should start to get concerned?

0 is acceptable.  Anything else suggests that something isn't working.
 
We had a high number of 8X errors and resolved most of them. (the bulk
was from a bad tape drive).
Now when we come across any 8X errors we would do a long erase on the
media and put it back into the scratch pool.
This has corrected some of the 8X errors as well but not all of them.

Anyone have any other Ideas or suggestions on this issue?

Not all 8x errors are caused by media.  Some are caused by tape drives and some are caused by communication issues between the host and the tape drive.

If the bulk of your errors were caused by a bad tape drive, why are you doing a long erase on your media?  How do you really know it corrected anything at all?  If the tape media wasn't the issue in the first place, it certainly won't fix anything.  And if it did fix anything, you've got flaky media and it should be destroyed - doing more I/O on the media won't make things better.

Media is CHEAP compared to the cost of data it's hosting.

   .../Ed

--
Ed Wilts, Mounds View, MN, USA
mailto:ewilts AT ewilts DOT org
_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
<Prev in Thread] Current Thread [Next in Thread>