Veritas-bu

[Veritas-bu] Could it have been sunspots or solar flares?

2002-07-17 16:35:44
Subject: [Veritas-bu] Could it have been sunspots or solar flares?
From: bryan_bahnmiller AT agilent DOT com (bryan_bahnmiller AT agilent DOT com)
Date: Wed, 17 Jul 2002 14:35:44 -0600
Dale,

> NetBackup 3.4.2
> Solaris 2.6
> Separate master and media server
> 
> All of a sudden backups were failing on this media server 
> with the error "Unable to allocate new media, storage unit 
> has none available".  After some investigation it was found:
> 
> 1.  The robot definition was gone from the media server.
> 2.  The drives were renamed and the paths were switched 
> around.  Also the 4mm drive on the server was now defined. 
> And the 4mm drive on the master was now defined even though 
> it strictly controls the robots.
> 3.  The volume host was changed from the master server to the 
> local media server.
> 4.  The barcode rules changed the default pool from scratch 
> back to netbackup.
> 
> And I think that was all.  Anybody have any idea what might 
> have caused this?  Nobody was fooling with NetBackup at the 
> time this happened.

  We have seen something similar happen in a couple instances with 3.4. What we 
have found is that if _anyone_ who has administrative authority starts up the 
Java GUI the first time, it comes up in "Configure NetBackup" mode. If, by 
mistake, you don't exit fairly quickly, and happen to start in on the 
configuration of your already configured system, it can wipe out major portions 
of your configuration.

  We have not had the desire to test this extensively with our operational 
systems. So we don't know exactly at which point in the sequence things start 
getting over-written. Someone else can do so if they desire. ;)

   Bryan

Bryan Bahnmiller
IT | Information Technology
Agilent Technologies, Inc.
1900 Garden of the Gods Road, M/S C103G
Colorado Springs, CO 80907
719-590-3262
www.agilent.com



<Prev in Thread] Current Thread [Next in Thread>