LCG Computing Element (lcg CE)

LCG CE is a native computing resource access service with Globus Gatekeeper. LCG has modified some of its component to improve its performance like job manager.

• Functional description
• Daemons running
• Init scripts and options (start|stop|restart|...)
• Configuration files location with example or template
  • /opt/globus/etc/globus-gass-cache-marshal.conf, /opt/globus/etc/globus-job-manager-marshal.conf
    • logf (srting) - location of the log file (default is relative to GLOBUS_LOCATION)
    • maxproc (numeric) - maximum number of parallel requests [this is the most useful variable for tuning] (5 by default)
    • tick (numeric) - hung child processes are killed every this number of seconds (if no other events are happening) (300 by default).
    • reqtout (numeric) - client should send a complete request in this number of seconds after connection (10 by default)
    • proctout (numeric) - each request (child process) is allowed to run this number of seconds (600 by default)
    • reqlimit (numeric) - maximum size of a request in bytes (16384 by default). One should increase this limit if environment is very large.
    • window (numeric) - data block for recv/send in bytes, probably should never be changed (default value is 4096 (x86 page size)).
    • debug (numeric) - debug level. There are three of them: 0 - only warnings (default), 1 - all messages, 2 - stderr is being redirected to the log file (bad for log parsers, but good for catching problems in perl jobmanagers)
    • All parameters (except debug 1 -> 2) could be changed online (modify config file and sent a sighup), both daemons create pidfiles in /var/run/ but their location is not configurable.

• Logfile locations (and management) and other useful audit information
• Possible unit test of the service
• Where is service state held (and can it be rebuilt)
• Cron jobs
• Utility scripts
• Location of reference documentation for users
• Location of reference documentation for administrators

-- DiQing - 28 Apr 2008