RB/WMS Monitoring

Metrics

• Number of jobs in the state Running in the condor queue.
• Number of jobs in the states Running or Idle in the condor queue.
• Number of jobs treated in the previous day.
• Average load of the machine during the past 15 minutes.
• Number of entries in file /var/edgwl/workload_manager/input.fl (Workload Manager).
• Number of entries in file /var/edgwl/jobcontrol/queue.fl (Job controler).
• Number of dg20logd_* files in directory /var/tmp.
• Size (in KB) of file /var/{edgwl,glite}/workload_manager/input.fl.
• Size (in KB) of file /opt/condor/var/condor/spool/job_queue.log.
• Number of file descriptors opened by process edg-wl-log_monitor.
• Occupancy (in %) of directory /var/{edgwl,glite}/SandboxDir.

Web server: lxb2007.cern.ch

You can access to the informations published by the RBs and WMS nodes at the following url:

http://lxb2007.cern.ch/monitoring/monitoring.html

The informations are updated every 30 minutes.

TODO list

• Run this service as a daemon.
• Add new metrics / sensors
• Write Lemon sensors.