Veritas-bu

[Veritas-bu] NetBackup Disk Staging Performance Questions

2007-03-01 15:26:36
Subject: [Veritas-bu] NetBackup Disk Staging Performance Questions
From: jroyer at digitalmotorworks.com (Joe Royer)
Date: Thu, 1 Mar 2007 14:26:36 -0600 (CST)
I use a STK FLX210 (EOL) SATA array for my DSSU, but I'm using NBU 
5.1MP4 and LTO2.  I have no trouble maxing out LTO2 with a 3+1 RAID-5 
on SATA, although the MAX_STAGING_JOBS and STAGING_JOB_KB_LIMIT 
certainly helped.

As for the failed images not getting cleaned up, I have noticed that 
*most* of them do get cleaned up about 4 hours after the failure.  I'm 
still trying to correlate that 4 hour number to a setting and haven't had 
much luck.  4 hours is the default preprocess interval; I doubt that's it, 
but I haven't actually tried mucking with it yet.


----Original message----
Date: Thu, 01 Mar 2007 07:05:47 -0600
From: Ed Wilts <ewilts at ewilts.org>
Subject: Re: [Veritas-bu] NetBackup Disk Staging Performance Questions

On 2/28/2007 10:34 PM, Anthony Segran wrote:
> /Is SATA a common choice out there for DSSU's?

We use FATA drives - not much difference between them and SATA really.

> //We experienced slow performance during duplication, on both HDS 9585
> arrays and NetApp 3020 over FCP. We're currently talking to NetApp about
> 3020 performance issues, so do you know of other NBU customers using
> NetApp 3020's for DSSU's?/

This appears to be a NetBackup issue and we're working directly with
Symantec engineering on our destaging performance issues.  We believe
we've got the right people at Symantec involved, we've given them a lot
of data, and they've already been out to visit us.  They're actively
working this issue.  There are many factors involved and we don't know
yet if it's a NetBackup bug, design issue, or a customer configuration
issue.  If it's the latter, it's not going to do you much good.  The
logs currently claim that we can't read off of disk fast enough to drive
a single LTO-3 drive but we've demonstrated using non-NBU tools that we
can read at 380+ MB/sec off of disk (essentially saturating the HBAs)
and the LTO-3 isn't that fast (don't we wish!).  If I get something
definitive that I can pass on, I'll post to the list.  You and I are not
the only people that have posted about destaging performance.

> /Another DSSU related issue we're seeing is orphan images left on the
> DSSU from failed/incomplete jobs. We've turned the appropriate logs on
> and will be clearing these out manually and escalate again if/when we
> see further occurrences. Support assure us that this is not a
> normal/reported issue with NBU.

We use DSSUs a *lot*.  We saw this issue regularly in pre-6.0MP4 days
but have not seen it since.  We worked with Symantec extensively during
the MP3 -> MP4 development cycle.  Continue working with support and
escalate the case if you need to.  Obviously orphan images are a bug and
they're supposed to be cleaned up automatically.

         .../Ed


-- 
Joe Royer / SysAdmin / Digital Motorworks / 512-692-1028


This message and any attachments are intended only for the use of the addressee 
and may contain information that is privileged and confidential. If the reader 
of the message is not the intended recipient or an authorized representative of 
the intended recipient, you are hereby notified that any dissemination of this 
communication is strictly prohibited. If you have received this communication 
in error, please notify us immediately by e-mail and delete the message and any 
attachments from your system.