Job monitoring over the Grid with GridICE infrastructure.

In a wide-area distributed and heterogeneous grid environment, monitoring plays an important and crucial role. It includes system status checking, performance tuning, bottlenecks detecting, troubleshooting, fault notifying. In particular a good monitoring infrastructure have to provide all the information needed to track the status of each job submitted to the GRID in order to be able to detect problems as soon as they come up. In this respect a good interaction between the monitoring system and the other Grid services is required. The current LCG testbed integrates GridICE as monitoring system. It already measures and publishes the status of each grid resource as a function of time. In this paper is presented the effort to integrate additional information in the current GridICE infrastructure, in order to track the status of the jobs executed, running or queued on the Grid, starting from the job name, the Virtual Organization (VO) to which the job belongs, the real or mapped user who submitted the job, the effective CPU time consumed and the job exit status