From morrone2 at llnl.gov  Fri Nov 16 23:06:08 2012
From: morrone2 at llnl.gov (Christopher J. Morrone)
Date: Fri, 16 Nov 2012 15:06:08 -0800
Subject: [lustre-devel] Remove libsysio?
Message-ID: <50A6C6E0.80105@llnl.gov>

Will anyone still be using the libsysio subpackage with lustre as of 
Lustre 2.4?  My vague recollection is that it is only needed for systems 
that use catamount, and is no longer needed.

Does anyone object to me removing it from the lustre tree?

Chris

From andreas.dilger at intel.com  Sat Nov 17 06:00:05 2012
From: andreas.dilger at intel.com (Dilger, Andreas)
Date: Sat, 17 Nov 2012 06:00:05 +0000
Subject: [lustre-devel] Remove libsysio?
In-Reply-To: <50A6C6E0.80105@llnl.gov>
Message-ID: <CCCC7581.2C32%andreas.dilger@intel.com>

On 11/16/12 4:06 PM, "Christopher J. Morrone" <morrone2 at llnl.gov> wrote:

>Will anyone still be using the libsysio subpackage with lustre as of
>Lustre 2.4?  My vague recollection is that it is only needed for systems
>that use catamount, and is no longer needed.
>
>Does anyone object to me removing it from the lustre tree?

I believe libsysio is needed in order for userspace liblustre to be
usable.  It provides the "POSIX/VFS" layer that applications link against,
and then liblustre provides the glue between libsysio and the Lustre
client interfaces (LOV/OSC/MDC).

There isn't a huge amount of interest in liblustre these days I think.

Cheers, Andreas

From alexey_lyashkov at xyratex.com  Mon Nov 19 10:08:13 2012
From: alexey_lyashkov at xyratex.com (Alexey Lyahkov)
Date: Mon, 19 Nov 2012 14:08:13 +0400
Subject: [lustre-devel] [wc-discuss] Remove libsysio?
In-Reply-To: <50A6C6E0.80105@llnl.gov>
References: <50A6C6E0.80105@llnl.gov>
Message-ID: <95C675E4-5671-4ADB-A5B2-1CEF85A2CA88@xyratex.com>

that is also used for FUSE clients.

On Nov 17, 2012, at 03:06, Christopher J. Morrone wrote:

> Will anyone still be using the libsysio subpackage with lustre as of Lustre 2.4?  My vague recollection is that it is only needed for systems that use catamount, and is no longer needed.
> 
> Does anyone object to me removing it from the lustre tree?
> 
> Chris

----------------------------------------------
Alexey Lyahkov
alexey_lyashkov at xyratex.com


From uja at ornl.gov  Mon Nov 19 13:35:47 2012
From: uja at ornl.gov (James A Simmons)
Date: Mon, 19 Nov 2012 08:35:47 -0500
Subject: [lustre-devel] Remove libsysio?
In-Reply-To: <CCCC7581.2C32%andreas.dilger@intel.com>
References: <CCCC7581.2C32%andreas.dilger@intel.com>
Message-ID: <1353332147.25490.39.camel@bohr.ornl.gov>

On Sat, 2012-11-17 at 01:00 -0500, Dilger, Andreas wrote:
> On 11/16/12 4:06 PM, "Christopher J. Morrone" <morrone2 at llnl.gov> wrote:
> 
> >Will anyone still be using the libsysio subpackage with lustre as of
> >Lustre 2.4?  My vague recollection is that it is only needed for systems
> >that use catamount, and is no longer needed.
> >
> >Does anyone object to me removing it from the lustre tree?
> 
> I believe libsysio is needed in order for userspace liblustre to be
> usable.  It provides the "POSIX/VFS" layer that applications link against,
> and then liblustre provides the glue between libsysio and the Lustre
> client interfaces (LOV/OSC/MDC).
> 
> There isn't a huge amount of interest in liblustre these days I think.

I had this discussion some time ago with Oleg. Libsysio actually exist
outside of the tree at http://sourceforge.net/projects/libsysio/. We
could have it be pulled automatically when some one wants to build
liblustre instead of it sitting in the tree. In order to do this some
one would have to maintain libsysio at sf. Currently their are fixes in
the lustre tree that are not upstream.

From seguranca at itauseg.com.br  Mon Nov 19 13:56:43 2012
From: seguranca at itauseg.com.br (Banco Itau)
Date: Mon, 19 Nov 2012 14:56:43 +0100 (CET)
Subject: Sua senha eletr�nica foi expirada pelo sistema.
Message-ID: <20121119135644.10761824061@h1402715.stratoserver.net>

An HTML attachment was scrubbed...
URL: <http://lists.opensfs.org/pipermail/lustre-devel_lists.opensfs.org/attachments/20121119/80b8c2fd/attachment.html>

From chris.gearing at intel.com  Wed Nov 21 13:00:18 2012
From: chris.gearing at intel.com (Gearing, Chris)
Date: Wed, 21 Nov 2012 13:00:18 +0000
Subject: [lustre-devel] New Test Framework Development Group Call
Message-ID: <CCD13DB2.33705%chris.gearing@intel.com>

Hi All,

Nathan, Roman and I have been having discussions about the new test framework, many of you will have read Nathan's requirements specification. In order to start making some serious progress with this we have decided to instigate a weekly technical call.

The first objective is to agree a common goal and once that is in place we will manage the process of executing the development of the new test framework. Jodi Levi who many of you know will be providing her PM skills to this development.

The call is open to all and has the following time and bridge information.

Every Tuesday starting on the 27th November
01:00 CST, 08:00 PST, 17:00 GMT, 21:00 MSK

Bridge Info
+1 916-356-2663, Bridge: 1, Passcode: 1146033

Note: We have a whole bunch of world wide local numbers so if your not US based (and I'm not) please email me your location and I'll reply with with your local number.

Many thanks

Chris



---------------------------------------------------------------------
Intel Corporation (UK) Limited
Registered No. 1134945 (England)
Registered Office: Pipers Way, Swindon SN3 1RJ
VAT No: 860 2173 47

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.

From chris.gearing at intel.com  Thu Nov 22 17:47:04 2012
From: chris.gearing at intel.com (Gearing, Chris)
Date: Thu, 22 Nov 2012 17:47:04 +0000
Subject: [lustre-devel] New Test Framework Development Group Call -
	Correction
In-Reply-To: <CCD13DB2.33705%chris.gearing@intel.com>
Message-ID: <CCD413C6.33970%chris.gearing@intel.com>

Hi All,

I sent out a mail about the new test framework development group call and as Bruce spotted didn't get the timezones right!

Again call is open to all and has the following time and bridge information.

The correct times are

Every Tuesday starting on the 27th November:
00:00 CST, 08:00 PST, 16:00 GMT, 20:00 MSK

Bridge Info:
+1 916-356-2663, Bridge: 1, Passcode: 1146033

Note: We have a whole bunch of world wide local numbers so if your not US based (and I'm not) please email me your location and I'll reply with with your local number.

Many thanks

Chris



---------------------------------------------------------------------
Intel Corporation (UK) Limited
Registered No. 1134945 (England)
Registered Office: Pipers Way, Swindon SN3 1RJ
VAT No: 860 2173 47

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.

From spitzcor at cray.com  Mon Nov 26 19:56:52 2012
From: spitzcor at cray.com (Cory Spitz)
Date: Mon, 26 Nov 2012 13:56:52 -0600
Subject: [lustre-devel] [Lustre-discuss] New Test Framework Development
 Group Call - Correction
In-Reply-To: <CCD413C6.33970%chris.gearing@intel.com>
References: <CCD413C6.33970%chris.gearing@intel.com>
Message-ID: <50B3C984.9040500@cray.com>

Hi, Chris.

I'm still confused about the time.

Every Tuesday starting on the 27th November
<< 01:00 CST, 08:00 PST, 17:00 GMT, 21:00 MSK
 >> 00:00 CST, 08:00 PST, 16:00 GMT, 20:00 MSK

Both of those listings are inaccurate.  Neither 01:00 or 00:00 CST == 
08:00 PST.

Since 8 PST == 16 GMT == 20 MSK, I'm guessing that you mean:
10:00 CST, 08:00 PST, 16:00 GMT, 20:00 MSK

Is that right?

Thanks,
-Cory

On 11/22/2012 11:47 AM, Gearing, Chris wrote:
> Hi All,
>
> I sent out a mail about the new test framework development group call and as Bruce spotted didn't get the timezones right!
>
> Again call is open to all and has the following time and bridge information.
>
> The correct times are
>
> Every Tuesday starting on the 27th November:
> 00:00 CST, 08:00 PST, 16:00 GMT, 20:00 MSK
>
> Bridge Info:
> +1 916-356-2663, Bridge: 1, Passcode: 1146033
>
> Note: We have a whole bunch of world wide local numbers so if your not US based (and I'm not) please email me your location and I'll reply with with your local number.
>
> Many thanks
>
> Chris
>
>
>
> ---------------------------------------------------------------------
> Intel Corporation (UK) Limited
> Registered No. 1134945 (England)
> Registered Office: Pipers Way, Swindon SN3 1RJ
> VAT No: 860 2173 47
>
> This e-mail and any attachments may contain confidential material for
> the sole use of the intended recipient(s). Any review or distribution
> by others is strictly prohibited. If you are not the intended
> recipient, please contact the sender and delete all copies.
>
> _______________________________________________
> Lustre-discuss mailing list
> Lustre-discuss at lists.lustre.org
> http://lists.lustre.org/mailman/listinfo/lustre-discuss

From bruce_korb at xyratex.com  Mon Nov 26 20:53:55 2012
From: bruce_korb at xyratex.com (Bruce Korb)
Date: Mon, 26 Nov 2012 12:53:55 -0800
Subject: [lustre-devel] [Lustre-discuss] New Test Framework Development
 Group Call - Correction
In-Reply-To: <50B3C984.9040500@cray.com>
References: <CCD413C6.33970%chris.gearing@intel.com>
	<50B3C984.9040500@cray.com>
Message-ID: <CADE-nV_BxA6Nu1NBvODo2j4Sx3miPzwF79q6CLLArikRSwMf0Q@mail.gmail.com>

On Mon, Nov 26, 2012 at 11:56 AM, Cory Spitz <spitzcor at cray.com> wrote:
> Hi, Chris.
>
> I'm still confused about the time.
>
> Every Tuesday starting on the 27th November
> << 01:00 CST, 08:00 PST, 17:00 GMT, 21:00 MSK
>>> 00:00 CST, 08:00 PST, 16:00 GMT, 20:00 MSK
>
> Both of those listings are inaccurate.  Neither 01:00 or 00:00 CST == 08:00
> PST.
>
> Since 8 PST == 16 GMT == 20 MSK, I'm guessing that you mean:
> 10:00 CST, 08:00 PST, 16:00 GMT, 20:00 MSK
>
> Is that right?

For myself, I'd as soon see strictly UTC for international times (aka
"GMT" or "Zulu").
I know my offset and, obviously, it is often confusing to get the time
zone correct
for other people in other time zones.  I had also presumed 16:00 Zulu
was correct. :)

Cheers - Bruce

From chris.gearing at intel.com  Tue Nov 27 10:22:10 2012
From: chris.gearing at intel.com (Gearing, Chris)
Date: Tue, 27 Nov 2012 10:22:10 +0000
Subject: [lustre-devel] [wc-discuss] Re: [Lustre-discuss] New Test
 Framework Development Group Call - Correction
In-Reply-To: <CADE-nV_BxA6Nu1NBvODo2j4Sx3miPzwF79q6CLLArikRSwMf0Q@mail.gmail.com>
Message-ID: <CCDA407F.33CE0%chris.gearing@intel.com>

Bruce is right just give the canonical form which is 16:00 UTC.

Also CST is China Standard Time not to be confused with the other CST
being Central Standard Time
(Is that really true, I can't find any other designator for the China
timezone)

Thanks for the patience.

Chris

On 26/11/2012 20:53, "Bruce Korb" <bruce_korb at xyratex.com> wrote:

>On Mon, Nov 26, 2012 at 11:56 AM, Cory Spitz <spitzcor at cray.com> wrote:
>> Hi, Chris.
>>
>> I'm still confused about the time.
>>
>> Every Tuesday starting on the 27th November
>> << 01:00 CST, 08:00 PST, 17:00 GMT, 21:00 MSK
>>>> 00:00 CST, 08:00 PST, 16:00 GMT, 20:00 MSK
>>
>> Both of those listings are inaccurate.  Neither 01:00 or 00:00 CST ==
>>08:00
>> PST.
>>
>> Since 8 PST == 16 GMT == 20 MSK, I'm guessing that you mean:
>> 10:00 CST, 08:00 PST, 16:00 GMT, 20:00 MSK
>>
>> Is that right?
>
>For myself, I'd as soon see strictly UTC for international times (aka
>"GMT" or "Zulu").
>I know my offset and, obviously, it is often confusing to get the time
>zone correct
>for other people in other time zones.  I had also presumed 16:00 Zulu
>was correct. :)
>
>Cheers - Bruce

---------------------------------------------------------------------
Intel Corporation (UK) Limited
Registered No. 1134945 (England)
Registered Office: Pipers Way, Swindon SN3 1RJ
VAT No: 860 2173 47

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.

From bruce_korb at xyratex.com  Tue Nov 27 15:06:57 2012
From: bruce_korb at xyratex.com (Bruce Korb)
Date: Tue, 27 Nov 2012 07:06:57 -0800
Subject: [lustre-devel] [wc-discuss] Re: [Lustre-discuss] New Test
 Framework Development Group Call - Correction
In-Reply-To: <CCDA407F.33CE0%chris.gearing@intel.com>
References: <CADE-nV_BxA6Nu1NBvODo2j4Sx3miPzwF79q6CLLArikRSwMf0Q@mail.gmail.com>
	<CCDA407F.33CE0%chris.gearing@intel.com>
Message-ID: <CADE-nV8rXVsaTdjCQy=yj1Ei-G_fuXeLzKMSB0-G1XCzh0s3-g@mail.gmail.com>

http://www.worldtimezone.com/index24.php

On Tue, Nov 27, 2012 at 2:22 AM, Gearing, Chris <chris.gearing at intel.com> wrote:
> Also CST is China Standard Time not to be confused with the other CST

The time zone TLA's are local conventions.  CST is meaningful only in
the US (or China).
That's why the phonetic alphabet is used for the canonical names, though
the referenced map didn't spell out Mary and Yankee. (Time zones with offsets
of +/- 12 from Zulu.)

Speaking of TLA's, I'm sure this is TMI.  :)

From chris.gearing at intel.com  Wed Nov 28 14:41:15 2012
From: chris.gearing at intel.com (Gearing, Chris)
Date: Wed, 28 Nov 2012 14:41:15 +0000
Subject: [lustre-devel] New Test Framework Development Group Call - Minutes
In-Reply-To: <CCD413C6.33970%chris.gearing@intel.com>
Message-ID: <CCDBD176.33E6A%chris.gearing@intel.com>

Hi,

Here are the brief notes from the meeting we held on Tuesday 27th November .

Attendees: Nathan Rutman, Roman Grigoryev Cory Spitz, Bruce Korb, Denis Kondratenko, Ashley Pittman, Jeff Garlough and at least one other I could not identify.

The major discussion was about evolution vs revolution, the outcome was that evolution would be nice but that we should not compromise the new framework to enable it.

To enable this it as decided that we should define the complete requirements/design for the new framework and then assess if a evolution is possible, if possible then this is an added benefit if not then old and new will have to exist in parallel.

The actions are;

1. Create a wiki entry on the OpenSFS wiki where everybody can add their wish list and requirements. (Chris)
2 .Send a note to the usual lists to explain the request for ideas. (Chris)

This group will meet again in 2 weeks to review and prioritise the list, and agree how a usable document of some sort can be created from the scribble.

Thanks

Chris

The meeting info is, feel free to attend at will but please send me a note so I can keep track of the attendees;

Bridge Info
916-356-2663, Bridge: 1, Passcode: 1146033

.........................................................................................................................................
Join online meeting<https://meet.intel.com/chris.gearing/M6HD8CYF>
https://meet.intel.com/chris.gearing/M6HD8CYF

First online meeting?<http://r.office.microsoft.com/r/rlidOC10?clid=1033&p1=4&p2=1041&pc=oc&ver=4&subver=0&bld=7185&bldver=0>
.........................................................................................................................................


---------------------------------------------------------------------
Intel Corporation (UK) Limited
Registered No. 1134945 (England)
Registered Office: Pipers Way, Swindon SN3 1RJ
VAT No: 860 2173 47

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.