Veritas-bu

Re: [Veritas-bu] Veritas-bu Digest, Vol 25, Issue 38

2008-05-19 23:07:49
Subject: Re: [Veritas-bu] Veritas-bu Digest, Vol 25, Issue 38
From: "Mark Glazerman" <Mark.Glazerman AT spartech DOT com>
To: <bob944 AT attglobal DOT net>, <veritas-bu AT mailman.eng.auburn DOT edu>
Date: Mon, 19 May 2008 21:54:22 -0500
Bob,

I'll try and summarize the process and the reasons behind it.

For the last year or so we had been running NBU 6.0 MP4 on a redhat master 
server.  We ran the usual upgrade procedure from 6.0 to 6.5.1 but just like 
getting 6.0 to work on redhat, we had some quirky issues which caused annoying 
problems.

When we saw that NBU 6.5 now supports Solaris x86 we decided that we wanted to 
run our master server on a Solaris 10 x86 OS. We built a new master server with 
the same hostname and ip as the current master server and installed 6.5.1 for 
x86 on that new box.  Once the NBU admin console was installed we kicked off a 
restore of our catalog, just like we would at a DR, so that all the storage 
units, policies, schedules etc. would be imported onto the new server.

That’s the reason that we imported our catalog, because it was a new server 
housing our master server.

As for the issues with DSU's, as I hope my post explains, our NBU environment, 
except for the upgrade to 6.5.1 and the change from redhat to x86, doesn't 
really explain why our backups would remain queued before their window closes.  
All schedules, policies, storage units have remained unchanged.

Hope that answers your questions.

Mark Glazerman
Desk: 314-889-8282
Cell: 618-520-3401
 please don't print this e-mail unless you really need to


-----Original Message-----
From: bob944 [mailto:bob944 AT attglobal DOT net] 
Sent: Monday, May 19, 2008 9:06 PM
To: veritas-bu AT mailman.eng.auburn DOT edu
Cc: Mark Glazerman
Subject: RE: Veritas-bu Digest, Vol 25, Issue 38

> Does anyone know if they made any changes to how DSU's are
> utilized in 6.5.1 ?
> 
> We recently upgraded to NBU 6.5.1.  That is the only
> difference in our backup environment.  The hardware,
> network settings, policies and schedules are all exactly
> the same as our 6.0 MP4 setup.  Everything was imported
> via full catalog restore. 

What was your process?  You seem to be saying you upgraded [from 6.0MP4]
to 6.5.1, then restored or in some way imported your [old?] catalog.
Unless I missed the memo, that isn't the correct process.  AFAIK,
NetBackup has never supported restoring/importing/whatevering an older
catalog version into a release.  You can 

o  upgrade NetBackup to the new version and run on it

o  backup the catalog and recover it to a NetBackup system of the same
version, then upgrade that new system to the new release of NetBackup

o upgrade the current NetBackup version in place, backup the upgraded
catalog and recover it to a system running the new release of NetBackup

You probably already know that.  Can you clarify what process you used
to get to 6.5.1?

> We send all of our backups to 2 Data Domain disk storage
> devices which then replicate to our DR site.
> 
> Since the 6.5.1 upgrade, huge swathes of our backups have
> been sitting queued until their backup window closes before
> returning a 196.  We don't really want to start playing
> around with the schedules as they were working just fine
> prior to the upgrade.  We are looking into the Open Storage
> option (OST) but this is more from the point of view of
> adding additional functionality rather than bringing 6.5.1
> up to par with 6.0 when using disk storage !!
> 
> Does anyone have any ideas on what might have changed
> regarding policy execution, job scheduling or NBU's handling
> of disk Storage that could be causing this new and very
> annoying issue ?  

_______________________________________________
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>