AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:
CFG:
PT:0
S:andy.sharp@lsi.com
RQ:
SSV:mhbs.lsil.com
NSV:
SSH:
R:<brian.stark@lsi.com>
MAID:2
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
X-Sylpheed-End-Special-Headers: 1
Date: Tue, 9 Mar 2010 14:44:59 -0800
From: Andrew Sharp <andy.sharp@lsi.com>
To: Brian Stark <brian.stark@lsi.com>
Subject: how to fix NDMP and other cases
Message-ID: <20100309144459.247bf019@ripper.onstor.net>
Organization: LSI
X-Mailer: Sylpheed-Claws 2.6.0 (GTK+ 2.8.20; x86_64-pc-linux-gnu)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

Here's how I would attack these affairs:

1.  We need to have an engineer on site debugging these issues/cases.
We have to let go of our SOP of "it has to be reproduced here" to work
on it.  Or depend on some tool to gather data from their system and
send it to us.

2.  Analyze what differences/fixes are in dev branch that will address
some of these bugs that can be integrated into r403rel branch.  This
could take a week of someone's time.

3.  A senior engineer sits down with Svati and review all the NDMP code.
This will likely take close to a week.  Jobi or Max or Chris.  Mods
have to come out of that and go before a design review.

4.  Danjing and Bill have to put together a patch that we can put on a
machine that we think might experience the Linux mgmtbus crash.  That
patch will log/gather information that we can use to diagnose the
problem.

