Hi Ralph,
On 10/16/2012 05:00 PM, Ralph Mueller-Pfefferkorn wrote:
Dear
all,
as todays skype call was cancelled due to low participation here
my comments per mail:
Site:
The second sentence is confusing me: "This property should contain
a descriptive name of the group of resources which are accounted
for in the record."
It sounds like it means a group of resources at a site and not the
site itself.
we meant that a group of resources is in fact a site. Maybe we could
rephrase it like that: "This property should contain a descriptive
name of the logical group of resources (that toghether define the
site) which are accounted for in the record"
Status/ExitStatus:
Beside the pure status description I see another use case: Was
this a useful/valid job for/of the user and thus be accountable?
But this would need that - in case of failure - it states if the
problem was caused by the application or the resource.
The ExitStatus is an undefined numerical value - it could be
everything and thus hard to interpret except by the resource owner
and the user.
Status could be used to intercept status related to the
infrastructure because it gives information "from the batch queuing
system’s accounting record". ExitStatus depends on the
implementation and might have a meaning only for the VO/user. The
site could filter the records using those two values to discard jobs
which failed because of problems on the infrastructure.
HostType:
Its a nice to know but I don't see a real use case. In the storage
area we have already a description of the storage with
StorageMedia, StorageClass.
In the compute area what should it state? The job ran on a
cluster, an SMP machine ...?
This description is meant for the description of the service. We are
suggesting to use the GLUE values that are proposed here:
https://docs.google.com/spreadsheet/ccc?key=0An1CLx-Ye4OFdDRkV180clZQampoRGwyWVpsdmRicmc#gid=0
for example org.glite.ce.CREAM, org.dcache.storage, etc.
Cheers,
Andrea and Jon
Andrea Cristofori wrote on 11.10.2012 16:44:
Dear all,
A new updated version of the document can be found at this
address:
https://redmine.ogf.org/dmsf_files/167?download=
What changed compared to yesterday email are the attributes and
blocks
description in RecordIdentidyBlock, SubjectIdentityBlock and
ComputeUsageBlock.
Cheers,
Andrea
On 10/10/2012 03:19 PM, Andrea Cristofori wrote:
Dear all,
Last phone meeting almost no one was available and was
cancelled. Next
phone meeting is fixed for next Tuesday, the 16th of October
at 15:00
CEST.
Jon and I discussed some of the attribute present on the
attribute
matrix and we updated the table. The relevant attribute are
those
marked with a question mark.
We just copied the description of the Status and ExitStatus
from EMI
CAR 1.0.
You can find the updated document here:
https://redmine.ogf.org/dmsf_files/161?download=
In case of further changes before the phone meeting we will
send a
notification.
Cheers,
Andrea
--
ur-wg mailing list
ur-wg@ogf.org
https://www.ogf.org/mailman/listinfo/ur-wg
--
Andrea Cristofori
INFN-CNAF
Viale Berti Pichat 6/2
40127 Bologna
Italy
Tel. : +39-051-6092920
Skype: andrea-cnaf