Then just delete the line from the LDIF. That way you make no delcaration; therefore there is no reason for a job to try doing anything specific. I think that this is better than no data.
("." aught to work, but no publishing is even more aught to work).
On 1 May 2012, at 11:28, Daniela Bauer wrote:
> I still think publishing "." is going to cause more trouble that it's worth....
>
> On 1 May 2012 10:39, Stuart Purdie <[log in to unmask]> wrote:
>> If you need to hack GlueSubClusterWNTmpDir, it's set on the CE's in
>>
>> /opt/glite/var/tmp/gip/glite-info-static-cluster.conf
>>
>> (As always, if you have to edit it, then watch out for whitespace at the end of the line.)
>>
>> It appears to default to /tmp, based on /opt/glite/yaim/functions/config_gip ... so I don't think that YAIM lets you configure it.
>>
>>
>> [Reminder: This Glue Attribute is (although apparently unused) the location where jobs should write files when on the worker node. So if /tmp is not a place suitable for that, it would be better to edit the publishing. I doubt that anyone will use that published value - but it would be rather bad form if a user were to do bad things, when they are trying to do the right thing in following published data]
>
>
>
> --
> -----------------------------------------------------------
> [log in to unmask]
> HEP Group/Physics Dep
> Imperial College
> Tel: +44-(0)20-75947810
> http://www.hep.ph.ic.ac.uk/~dbauer/
|