Networker

[Networker] NetWorker host: Solaris vs. Windows vs. Linux

2009-03-13 04:50:50
Subject: [Networker] NetWorker host: Solaris vs. Windows vs. Linux
From: tkimball <networker-forum AT BACKUPCENTRAL DOT COM>
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Date: Fri, 13 Mar 2009 04:42:25 -0400
[Going off-topic for a bit, as this may be important for others on T2000 et al 
servers.]

Sorry, I don't have documentation, only test results.

Unfortunately, the solution to our problem is "install/re-jump from 10/08 and 
install latest cluster / firmware.'  Which is what I'm doing with our 
replacement NW Server as I type this.  Argh.

Patching a 5/08 install with latest cluster did not work; I don't know about 
using latest cluster on 8/07, since we don't have any T2K on it that I know of. 
 We have one on 6/06, patched to 5/08, but he's experiencing different issues 
(which I think is with an app, not Solaris).

The reason for the problem is Sun changing the 'core' driver on T* server 
networking ports from nge to e1000g.  The initial implementation on 5/08 was 
generating a bunch of internal interrupts / retransmit requests (apparently for 
no reason) and throttling the speed.

Note that we also had cases of x4600 and x4100 servers with the same problem on 
net2-3 (which were nge on the M2 versions).  T1000 does not have the problem, 
since it does not use nge.

--TSK




Rachel Polanskis wrote:
> On Tue, 10 Mar 2009, tkimball wrote:
> 
> 
> > Solaris 10 is mostly rock solid.  Just don't install 5/08 or patch to it 
> > (in particular if a T2000 or similar is involved).  A network change (for 
> > consistent network drivers) was horribly bugged, and throughput went into 
> > the gutter on the T2K.  The 10/08 release fixed it enough to make me happy 
> > to continue.
> > 
> 
> Hi, 
> can you please shed more light on the T2000 issue?
> 
> We are using T2000's in various ways and are experiencing the 
> horrible throughput issues you describe without resolution.
> 
> We run a T5220 with Solaris 10 8/07 for Legato and so far it's doing fine 
> but our 20 or so T2000's are not all working very well at all 
> throughput wise, with Zones dropping off, sluggish response, 
> java performance etc....   and they are all patched towards 5/08.....
> 
> If you could steer me to more documented info I'd be very grateful!
> 
> rachel


+----------------------------------------------------------------------
|This was sent by t.s.kimball AT gmail DOT com via Backup Central.
|Forward SPAM to abuse AT backupcentral DOT com.
+----------------------------------------------------------------------

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