 |
 |
Hi Phil/All When spotting jobs like this running under dteam or ops, it is worth checking with the user why they are using that VO. I'm sure most are monitoring and testing related jobs but there may be some who "accidentally" use this VO when they should be using a user based one. Thanks, Jeremy -----Original Message----- From: Testbed Support for GridPP member institutes [mailto:[log in to unmask]] On Behalf Of Phil Roffe Sent: 08 August 2008 10:27 To: [log in to unmask] Subject: Inefficient Jobs I noticed this morning that Durham have 76 very inefficient jobs, all coming from the same dteam user. From looking at Ganglia and Monami it seems Glasgow may have too! A quick check on the node to see what the problem is show the following problem... dteam019 5194 0.0 0.0 6216 700 ? S 01:25 0:00 sh -c ./Ecorad_fondo.sh 11162 > Ecorad_fondo_11162-out.txt 2>Ecorad_fondo_11162-err.txt dteam019 5200 0.0 0.0 5732 540 ? S 01:25 0:00 sleep 669720 The sleep 669720 seconds is 7.75 days! I have ticketed the user! Does any other site have this problem? Thanks, Phil