On Wed, Sep 05, 2012 at 01:13:50PM +0300, Paschalis Korosoglou wrote:
> Hi all,
>
> we have recently setup a new cream CE (cream.afroditi.hellasgrid.gr) on HG-03-AUTH (UMD-2, Centos6). A site bdii is also running on this node. We have registered this instance on gocdb as non production so that it is not affecting reliability stats.
>
> Our problem at this point is that the WMSes cannot find this instance yet (it does not show up in the glite-wms-job-list-match command). We can however submit jobs via WMSes using the -r <resource_id> option of the glite-wms-job-submit command.
>
> Does anyone have any idea on where we should be looking for the error in the information system? i.e. What it there that should be published but is not?
As Stephen wrote there is no subcluster definition. From wms02.athena
logs with temporarily increased WorkloadManager LogLevel:
05 Sep, 13:52:58 -D: [Debug] fetch_bdii_ce_info(/home/condor/execute/dir_26089/userdir/emi.wms.wms-ism/build/../src/purchaser/ldap-utils.cpp:510): Skipping cluster cream.afroditi.hellasgrid.gr due to empty subcluster definition
05 Sep, 13:52:58 -D: [Debug] fetch_bdii_ce_info(/home/condor/execute/dir_26089/userdir/emi.wms.wms-ism/build/../src/purchaser/ldap-utils.cpp:518): Skipping ce cream.afroditi.hellasgrid.gr:8443/cream-pbs-ops belonging to cluster cream.afroditi.hellasgrid.gr
05 Sep, 13:52:58 -D: [Debug] fetch_bdii_ce_info(/home/condor/execute/dir_26089/userdir/emi.wms.wms-ism/build/../src/purchaser/ldap-utils.cpp:518): Skipping ce cream.afroditi.hellasgrid.gr:8443/cream-pbs-dteam belonging to cluster cream.afroditi.hellasgrid.gr
Kyr.
>
> Thank you in advance for your help,
> Paschalis
>
> Scientific Computing Support Unit
>
> --
>
> Scientific Computing Center, Aristotle University of Thessaloniki
> University Campus, GR 541 24, School of Sciences, Building 22A
> Tel. +302310998988, Fax. +302310994309, http://www.grid.auth.gr
>
|