AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:
CFG:
PT:0
S:andy.sharp@onstor.com
RQ:
SSV:mail.onstor.net
NSV:
SSH:
R:<paul.negus@onstor.com>,<dl-LeopardCoreTeam@onstor.com>
MAID:1
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
RMID:#imap/andys@onstor.net@exch1.onstor.net/INBOX	0	102AB4F33EBBDB4C91915B145C8E9FB31370ABBBF5@exch1.onstor.net
X-Sylpheed-End-Special-Headers: 1
Date: Fri, 12 Jun 2009 15:34:10 -0700
From: Andrew Sharp <andy.sharp@onstor.com>
To: Paul Negus <paul.negus@onstor.com>
Cc: dl-Leopard Core Team <dl-LeopardCoreTeam@onstor.com>
Subject: Re: Minutes from 6/11 Leopard HA Core Team Meeting
Message-ID: <20090612153410.2cabf1ae@ripper.onstor.net>
In-Reply-To: <102AB4F33EBBDB4C91915B145C8E9FB31370ABBBF5@exch1.onstor.net>
References: <102AB4F33EBBDB4C91915B145C8E9FB31370ABBBF5@exch1.onstor.net>
Organization: Onstor
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

On Fri, 12 Jun 2009 15:21:35 -0700 Paul Negus <paul.negus@onstor.com>
wrote:

> Hello all!
> 
> If you missed Thursday's meeting, you can see what was presented by
> using the following link: \\Mightydog\program management\Leopard
> Phase 2\Weekly Team
> Meetings\06-2009\06-11-2009<file:///\\Mightydog\program%20management\Leopard%20Phase%202\Weekly%20Team%20Meetings\06-2009\06-11-2009>
> 
> Attendees:  Brian, Andy, Arnaldo, Larry, Eugene, Rich L, Joe,
> Jonathan, Pat, Paul, Sandrine

s/Andy/Andy, Andy/

> 
> Scoping - ONStor & Nexenta
> 
>  1.  Brian reported that an agreement has been reached with Nexenta.
> The remaining two SOW's will be revised to reflect ONStors current
> requirements, particularly for HA.  This will require a small up
> front payment with the balance due on completion of satisfactory
> testing. 2.  Brian reported that we are now expecting the first code
> drop for HA software to be early next week.
> 
> HW Update
> 
>  1.  Brian reported that Dell has announced the EOL for the 2950
> server with the last time buy date set at 7/30/2009.  The new
> hardware should be a direct replacement and engineering have
> identified a couple of configs that should meet our requirements.  At
> this stage it is assumed that the COGS for these new units will be
> higher than our current configs. 2.  Engineering have also identified
> that our current bezel will not work with the new configs and will
> require redesign. This will be looked at and a determination will be
> made whether or not we will be able to continue with a 'lit' logo
> primarily because of cost reasons. 3.  Brian is continuing to look
> for 10GigE options in light of the fact that Melanox will not support
> OpenSolaris on their solution.  This feature is likely to be pushed
> to the end of the phase because of the availability of the required
> hardware.
> 
> QA / Testing Update
> 
>  1.  A full status of current testing is available in Sandrine's
> slides in the folder referenced at the start of this email. 2.  A
> discussion on the status of 'event reporting' was held.  Sandrine
> reported that QE were not is in a position to start testing this as
> it has not been provided by Nexenta. 3.  A discussion was held about
> whether Onstor should push the implementation of Comstar or stay with
> the legacy iSCSI.  Rich reported that we have a workaround/workflow
> that needs more testing to confirm that it works reliably.   Brian
> stated that the legacy should be fully supported but since Nexenta
> will be working towards a full implementation of Comstar then that is
> the direction we should adopt.  It would, however, be better if
> Comstar was fully integrated in to the GUI going forward. 4.  We
> don't have any support contract to upgrade our Solaris10 clients to
> the latest patches to get NFSv4, so Rich H built an Opensolaris
> client. However, he was not able to join it in the open ldap domain
> so we won't be able to do multi-protocol testing with this client. He
> was also working on a RedHat5 client, hopefully this one will work
> better. 5.  Sandrine presented the latest results from the
> performance testing (see results in project folder) which instigated
> a discussion about the systems cache contribution to skewing the
> results.  At some time in the future we will need to run SpecFS but
> this is not part of the phase 2 deliverables.  Brian pointed out that
> the good news was that the performance tests demonstrated that the
> system is not constrained by the processors. 6.  Jumbo Frames are
> still not working in 1.1.8, although Nexenta claimed that they were
> fixed, and it needs to be determined whether this is an OpenSolaris
> issue or Nexenta.  Larry stated that he needs to check the build
> standard of the systems that Sandrine is using as he had noticed that
> sometimes the build does not always pick the correct modules. 7.  20
> new defects were found and reported this week.  Version 1.1.9 should
> fix 6 known defects. 8.  Brian reported that the first JBOD in a
> chain still cannot be seen by the ZFS fault manager.  This needs to
> be fixed asap.  At least to the level where the required software to
> support it is done even if the e-mail event notification is not.
> 
> Support / Documentation Update
> 
>  1.  Rich reported that a resource is still required to complete a
> System Administration Guide but that at present it is on hold pending
> funding.
> 
> Roundtable / Schedule
> 
>  1.  There were no roundtable items.
> 
> Please feel free to "Reply All" if I missed something important.
> 
> 
> 
> Thanks
> 
> Paul
