X-MimeOLE: Produced By Microsoft Exchange V6.5
Received: by onstor-exch02.onstor.net 
	id <01C8B5EF.20276E18@onstor-exch02.onstor.net>; Wed, 14 May 2008 11:20:02 -0700
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Content-class: urn:content-classes:message
Subject: RE: sub#22 prolly needs respin
Date: Wed, 14 May 2008 11:20:01 -0700
Message-ID: <BB375AF679D4A34E9CA8DFA650E2B04E09EE8322@onstor-exch02.onstor.net>
In-Reply-To: <20080514104526.3188944a@ripper.onstor.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: sub#22 prolly needs respin
Thread-Index: Aci16kr2YHGtGsliQVmPZiXXIi5DvgABAnHA
From: "Tim Gardner" <tim.gardner@onstor.com>
To: "Andy Sharp" <andy.sharp@onstor.com>
Cc: "Larry Scheer" <larry.scheer@onstor.com>


Ok, I have gotten the rest of the story.
Change 29211 fixes a problem that was introduced in the fix to defect
23389 which went into sub 22.
I will ask Max to integrate this change to cg_beta and then we will need
to
respin sub22.


> -----Original Message-----
> From: Andy Sharp
> Sent: Wednesday, May 14, 2008 10:45 AM
> To: Tim Gardner
> Cc: Larry Scheer
> Subject: Re: sub#22 prolly needs respin
>=20
> Perhaps I didn't make myself clear.  Currently sub#22 is unuseable
> unless you're doing an upgrade, ie., you already have a cluster.conf
> file and a clusterdb.
>=20
> On Wed, 14 May 2008 10:25:39 -0700 "Tim Gardner"
> <tim.gardner@onstor.com> wrote:
>=20
> > No, I want to understand this change first.
> > The change does not even have a defect associated with it.
> > It also looks to be related only to failure handling.
> >
> > > -----Original Message-----
> > > From: Larry Scheer
> > > Sent: Wednesday, May 14, 2008 9:57 AM
> > > To: Andy Sharp
> > > Cc: Tim Gardner
> > > Subject: RE: sub#22 prolly needs respin
> > >
> > > Tim,
> > >   What do you think? I need to re-spin sub 22 anyway. I can pick
up
> > this
> > > change if approved.
> > >
> > > I will wait to start the rebuild.
> > >
> > > Larry
> > >
> > >
> > > -----Original Message-----
> > > From: Andy Sharp
> > > Sent: Wed 5/14/2008 9:40 AM
> > > To: Larry Scheer
> > > Cc: Tim Gardner
> > > Subject: Re: sub#22 prolly needs respin
> > >
> > > Nothing to do with dmalloc.  As I mentioned, integrating that fix
> > > from max solved the problem.  There is a bug in
cluster-controller,
> > > I believe, that causes it to crash over and over when you are in
an
> > > unconfigured state, ie., after a system config reset, or after
> > > doing a flash_install without copying the config files.
> > >
> > > On Wed, 14 May 2008 09:28:18 -0700 "Larry Scheer"
> > > <larry.scheer@onstor.com> wrote:
> > >
> > > > Andy,
> > > >    Let's talk about what you were seeing when I get in. The
sub22
> > > > beta build has dmalloc enabled and I saw a couple of crashes,
one
> > > > immediately on reboot the other was with cluster controller when
I
> > > > tried to run a snapshot. When you say messed up what is the
nature
> > of
> > > > the messyness? Check the cores in /var/run and see if they are
all
> > > > due to dmalloc.
> > > >
> > > > After a reboot I didn't see the first crash. Because of this
> > behavior
> > > > I wonder if Max's change really was the fix? When you did your
own
> > > > personal build dmalloc would have been turned off. Perhaps it is
> > > > dmalloc that is ruining your buzz.
> > > >
> > > > I will be respinning the cougar builds with dmalloc turned off
by
> > > > commenting out the DMALLOC variable in /etc/defaults/onstor.
> > > >
> > > > Larry
> > > >
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: Andy Sharp
> > > > Sent: Wed 5/14/2008 12:58 AM
> > > > To: Andy Sharp
> > > > Cc: Larry Scheer; Tim Gardner
> > > > Subject: Re: sub#22 prolly needs respin
> > > >
> > > > On Wed, 14 May 2008 00:21:56 -0700 Andrew Sharp
> > > > <andy.sharp@onstor.com> wrote:
> > > >
> > > > > Just an FYI,
> > > > >
> > > > > I'm trying to run sub#22 beta equivalent on my cougar, and
it's
> > > > > quite a mess.  I think we need to integrate max's change #
29211
> > > > > into beta branch.  I'm going to try that now to see if it
works.
> > > >
> > > >
> > > > Yup, that fixed it.
> > > >
> >
