Dear All,
I had an action from the ops meeting yesterday to check on EGI understanding with regards to YAIM in EMI-3.
I asked this question at today's EGI operations meeting and the response was that it is expected that YAIM core will continue to be supported for the immediate future by most product teams. Only two products appear to have dropped YAIM in the first EMI-3 release: APEL and VOMS. APEL has not been pushed in staged rollout due to the dropping of YAIM and the lack of documentation for the configuration (especially as there is new functionality) - a ticket to improve the documentation has been created and is being followed-up.
The expectation is that product teams will continue to give priority to YAIM as it is simple for them to update, however going forwards there is no overall coordination at present and product teams are likely to only support their preferred tool (unclear at present but Puppet is widely used). Looking through the slides linked by Stephen there is a HEPiX working group looking at templates extracting the most relevant survey slide result:
"•Likely, there isn’t a configuration tool good for all the grid sites
•But… Puppet is taking the lead:
–Growing user community within EGI
•Already a big user community worldwide (true also for other tools)
–Big site (CERN) working on a set of templates (master/masterless)
•Lot of templates, but not all the products are covered
–Coordination work starting within an Hepix working group"
I will follow up on this topic more during the EGI CF next week, but clearly as Alessandra noted there is an expectation that the community needs to be increasingly involved in this area. Perhaps those attending HEPiX can bring back information on the working group status and we can then discuss how we can most usefully contribute.
Jeremy
On 3 Apr 2013, at 12:23, Stephen Burke wrote:
> Testbed Support for GridPP member institutes [mailto:TB-
>> [log in to unmask]] On Behalf Of Stephen Jones said:
>> It feels like being a passenger in a moving vehicle after the driver has jumped
>> out.
>
> The driver jumped out in about 2007, but it's only now that we're hitting some junctions :)
>
> http://www.sciencehumor.org/2007/quantum-junction-sign/
>
> Stephen
>
> --
> Scanned by iCritical.
|