JiscMail Logo
Email discussion lists for the UK Education and Research communities

Help for LCG-ROLLOUT Archives


LCG-ROLLOUT Archives

LCG-ROLLOUT Archives


LCG-ROLLOUT@JISCMAIL.AC.UK


View:

Message:

[

First

|

Previous

|

Next

|

Last

]

By Topic:

[

First

|

Previous

|

Next

|

Last

]

By Author:

[

First

|

Previous

|

Next

|

Last

]

Font:

Proportional Font

LISTSERV Archives

LISTSERV Archives

LCG-ROLLOUT Home

LCG-ROLLOUT Home

LCG-ROLLOUT  June 2009

LCG-ROLLOUT June 2009

Options

Subscribe or Unsubscribe

Subscribe or Unsubscribe

Log In

Log In

Get Password

Get Password

Subject:

Re: libstdc++-devel.i386 and libstdc++-devel.x86_64

From:

Dimitris Zilaskos <[log in to unmask]>

Reply-To:

LHC Computer Grid - Rollout <[log in to unmask]>

Date:

Fri, 5 Jun 2009 16:57:06 +0300

Content-Type:

multipart/signed

Parts/Attachments:

Parts/Attachments

text/plain (115 lines) , smime.p7s (115 lines)

Hi,

64 bits SL4 WNs always need some manual rpm tweaking here form time to 
time. If we knew it in the beginning we probably would not make them 64 
bit at all. I guess your decision was made due to something like "look 
64 bit WN is out, 64 bits are cool, perform better, let's do it". It is 
a temptation that I also find hard to resist.

But if I recall correctly, at some point our ROC asked us to move to 64 
bit WNs, obviously based on a request coming from somewhere. Most likely 
users. However our own userbase should not care about the glite arch as 
far as I know. There were assurances of backwards compatibility so it 
did not seem a bad idea.

I will agree that the amount of effort involved in debugging, 
discussing, opening tickets, replying to tickets, sending mails, 
escalating tickets, begging for replies, filling reports, waiting for 
solutions, following up, is too big and draws effort that could be 
invested in actual problem solving, including infrastructure problems 
that need quiet some brainstorming and a clear state of mind.

That's why for example I prefer to always play it safe when I know I do 
not have the time to risk for a new unknown update. I learned that the 
hard way. If you find yourself overwhelmed, I would seriously consider 
reinstalling everything with glite arch 32 bit, if you judge that the 
effort involved would save you more effort wasted in a nightmare of 
dependencies.

You do not see for example me complaining because I have complained 
already numerous times for a rather wide range of issues, sometimes for 
the same issues all over and I feel bad to be the one that always 
complains. And producing each time I complain a good report that can be 
productive takes so much effort that I tend to do that only for what is 
critical (like serious breakage) for me forgetting numerous minor 
annoyances. If you add up the usual site workloads, not much time is 
left for anything else.

Cheers,



O/H Arnau Bria έγραψε:
> On Wed, 3 Jun 2009 18:39:59 +0200
> Dario Barberis wrote:
> 
>> Well, let me comment on this one...
> Hi Dario,
>  
> [...]
>>> In resume, I feel in the middle of  experiments & gLite problems,  
>>> and we
>>> lose many time adapting ourselves to them... meanwhile our internal
>>> problems are abandoned.
>>>
>>> Or maybe I'm wrong and we're doing something wrong, cause i see no
>>> complains about problems like ours from other sites... How many
>>> sites are  running 64 btis software?
>>
>> First of all, we (ATLAS) have never asked sites to move to 64-bit  
>> systems. Our existing code does not work in 64-bit mode as it needs  
>> twice as much memory to run as in 32-bit mode. Work is in progress
>> to try to improve, but it will only apply to future releases on
>> SL(C)5 with gcc4.3.
> Ok, for that reason I ask myself why we (site pic) have moved to 4 bits,
> not why glite did move to 64... The migration was our decission, not an
> obligation.
> 
>   
>> CERN has deployed 64-bit SLC4 since a long time and our code works  
>> there. You should ask sites that have (nominally) the same system as  
>> you installed just now, before blaming code that runs perfectly  
>> everywhere else in the world.
> 
> ATLAS needs some 32 bits compatibility, so because of ATLAS's software
> compatibility we have to deal with sl4-x86_64 32bits packages, so the
> problems comes from ATLAS requirements. But I don't blame
> about ATLAS software, I was complaining about how SL4-x86_64 manages 32 bits packages.
> 
> Again, the decission to move to 64 bits was ours. We could ask
> experimets first and then take the correct decission.
> 
> *But, from what I've heard (you'll probably know much than me about
> this) apart from us, other sites (don't know the number) are
> failling new pilot jobs. We were not the only site who hit this issue
> (AFAIK).
> 
>> ATLAS code is running right now in >50k cores around the world on  
>> several flavours of RHEL4 based systems, and also on SL5 systems  
>> (provided SELinux is disabled).
> 
> Well, AFAIK ATLAS requested a test queue for sl5, so I suppose
> supporting ATLAS in SL5 is not trivial at this moment (correct me if
> I'm wrong).
> 
> In my previous post I was explaining my experience after some months in
> 64 bits world. not trying to excuse myself on other people's work
> neither complaining about ATLAS/gLite software. 
> 
>> Cheers
>>           Dario
> Cheers,
> Arnau
> 


-- 
=============================================================================
Dimitris Zilaskos
GridAUTH Operations Centre @ Aristotle University of Thessaloniki , Greece
Tel: +302310998988 Fax: +302310994309
http://www.grid.auth.gr
=============================================================================


Top of Message | Previous Page | Permalink

JiscMail Tools


RSS Feeds and Sharing


Advanced Options


Archives

April 2024
March 2024
November 2023
June 2023
May 2023
April 2023
March 2023
February 2023
September 2022
June 2022
May 2022
April 2022
February 2022
December 2021
November 2021
October 2021
September 2021
July 2021
June 2021
May 2021
February 2021
January 2021
November 2020
September 2020
August 2020
July 2020
June 2020
May 2020
April 2020
March 2020
February 2020
January 2020
November 2019
October 2019
September 2019
August 2019
July 2019
June 2019
May 2019
March 2019
February 2019
January 2019
December 2018
November 2018
October 2018
September 2018
August 2018
July 2018
June 2018
May 2018
April 2018
February 2018
January 2018
November 2017
October 2017
September 2017
July 2017
June 2017
May 2017
March 2017
February 2017
January 2017
December 2016
November 2016
October 2016
September 2016
August 2016
July 2016
June 2016
May 2016
April 2016
March 2016
February 2016
January 2016
December 2015
November 2015
October 2015
September 2015
August 2015
July 2015
June 2015
May 2015
April 2015
March 2015
February 2015
January 2015
December 2014
November 2014
October 2014
September 2014
August 2014
July 2014
June 2014
May 2014
April 2014
March 2014
February 2014
January 2014
December 2013
November 2013
October 2013
September 2013
August 2013
July 2013
June 2013
May 2013
April 2013
March 2013
February 2013
January 2013
December 2012
November 2012
October 2012
September 2012
August 2012
July 2012
June 2012
May 2012
April 2012
March 2012
February 2012
January 2012
December 2011
November 2011
October 2011
September 2011
August 2011
July 2011
June 2011
May 2011
April 2011
March 2011
February 2011
January 2011
December 2010
November 2010
October 2010
September 2010
August 2010
July 2010
June 2010
May 2010
April 2010
March 2010
February 2010
January 2010
December 2009
November 2009
October 2009
September 2009
August 2009
July 2009
June 2009
May 2009
April 2009
March 2009
February 2009
January 2009
December 2008
November 2008
October 2008
September 2008
August 2008
July 2008
June 2008
May 2008
April 2008
March 2008
February 2008
January 2008
December 2007
November 2007
October 2007
September 2007
August 2007
July 2007
June 2007
May 2007
April 2007
March 2007
February 2007
January 2007
2006
2005
2004
2003


JiscMail is a Jisc service.

View our service policies at https://www.jiscmail.ac.uk/policyandsecurity/ and Jisc's privacy policy at https://www.jisc.ac.uk/website/privacy-notice

For help and support help@jisc.ac.uk

Secured by F-Secure Anti-Virus CataList Email List Search Powered by the LISTSERV Email List Manager