[Go-essp-tech] Outcomes of the IS-ENES September Coding Sprint

[stephen.pascoe at stfc.ac.uk]

Thanks Neil,

I didn't look too closely at why Globus failed and I've now had to move on to other matters but when I return to it I'll let you know what went wrong.

We filed a bug on esgf.org about OpenSSL 1.0, however I think it was the reason for only one of the failures (I was definitely using openssl 0.9.x).  I don't think we should rush into OpenSSL 1.0 as it seems a major change, although I defer to what ever you and Gavin reckon is the best route.

Other than Globus my installation experience was very encouraging.  I can see the script is improving all the time.  However, I did have to switch to the devel version which isn't ideal for maintaining production systems.  

Stephen.

---
Stephen Pascoe  +44 (0)1235 445980
Centre of Environmental Data Archival
STFC Rutherford Appleton Laboratory, Harwell Oxford, Didcot OX11 0QX, UK


-----Original Message-----
From: Neill Miller [mailto:neillm at mcs.anl.gov] 
Sent: 23 September 2011 15:49
To: Pascoe, Stephen (STFC,RAL,RALSP)
Cc: go-essp-tech at ucar.edu; esg-gateway-dev at earthsystemgrid.org; esg-node-dev at lists.llnl.gov; is-enes-sa2-jra4 at lists.enes.org
Subject: Re: [Go-essp-tech] Outcomes of the IS-ENES September Coding Sprint

Hello Stephen,

I see a lot of Globus installation failures here, and wanted to note something that's come up lately.

The version of Globus we're basing things off of for ESG (i.e. 4.2.x) is API incompatible with OpenSSL 1.0+, which came out after and broke everything API wise (even the certificate hashes, which we ran into a while back).  We need a version that is less than OpenSSL 1.0 for the node build, such as 0.9.8, for example.

Gavin has been advised to remove OpenSSL 1.x, install a version prior to 1.x and build the node, and he has confirmed that worked. I am not aware of any s/w in the ESG stack that needs OpenSSL 1.x, so removing that seems to be most expedient way of resolving this.

Looks like Gavin is trying to build with both OpenSSL options on the node to see if we can manage with both, but it is causing things to break at this point.

If any of you see a need to resolve this will OpenSSL 1.x on the node, please let us know, otherwise the advice would be to remove OpenSSL 1.x for the ESG node.

thanks,
-Neill.

----- Original Message -----
From: "stephen pascoe" <stephen.pascoe at stfc.ac.uk>
To: "stephen pascoe" <stephen.pascoe at stfc.ac.uk>, is-enes-sa2-jra4 at lists.enes.org
Cc: go-essp-tech at ucar.edu, esg-gateway-dev at earthsystemgrid.org, esg-node-dev at lists.llnl.gov
Sent: Friday, September 23, 2011 7:59:01 AM
Subject: Re: [Go-essp-tech] Outcomes of the IS-ENES September Coding Sprint





See attached an amendment to our outcomes to include our investigation of CORDEX support in Gateway 2.0. 



Stephen. 




--- 

Stephen Pascoe +44 (0)1235 445980 

Centre of Environmental Data Archival 

STFC Rutherford Appleton Laboratory, Harwell Oxford, Didcot OX11 0QX, UK 





From: Pascoe, Stephen (STFC,RAL,RALSP) 
Sent: 23 September 2011 13:20 
To: is-enes-sa2-jra4 at lists.enes.org 
Cc: go-essp-tech at ucar.edu; esg-gateway-dev at earthsystemgrid.org; esg-node-dev at lists.llnl.gov 
Subject: Outcomes of the IS-ENES September Coding Sprint 



Dear all, 



Developers from the EU IS-ENES project have been meeting this week to work intensively on installation of the latest datanode system and the Gateway 2.0 beta release. It has been a very productive meeting which brings us much closer to deploying both Gateway 2 and the ESGF P2P system. 



Please find attached a summary of outcomes which includes feedback on the datanode install system and Gateway 2.0. 



Thanks, 

Stephen. 



--- 

Stephen Pascoe +44 (0)1235 445980 

Centre of Environmental Data Archival 

STFC Rutherford Appleton Laboratory, Harwell Oxford, Didcot OX11 0QX, UK 




-- 
Scanned by iCritical. 

_______________________________________________
GO-ESSP-TECH mailing list
GO-ESSP-TECH at ucar.edu
http://mailman.ucar.edu/mailman/listinfo/go-essp-tech