CCSS Cracks Down on Long-Running Jobs
April 28, 2009 Alex Woodie
When System i batch jobs run longer than they’re supposed to, there can be serious consequences. Users may not be able to log in. Backups may not run. Industry regulations may not be followed. Last week, CCSS announced a new run time monitor in its i OS systems management tool, QSystem Monitor, that alerts administrators and operators to long-running jobs before they can cause these problems.
According to CCSS, System i shops that use job scheduling software are the most susceptible to rouge jobs that run too long. The complexity of modern job schedulers means that jobs can run at any time, without operators knowing about it. Cross-machine dependencies, where a job on one machine must complete before a second job on a second machine begins, further compounds the problem.
With its new job monitor in QSystem Monitor version 11.5, CCSS is shedding some light into the status of long-running jobs. It’s also adding some automation into the handling of jobs run amok, and enabling System i administrators to manage these jobs by exception.
The new job monitor will alert the administrator to jobs running longer than they should by flashing a red bar in the QSystem Monitor’s PC-based client. In addition to the visual queue, the client shows essential details about the job, such as which subsystem it’s running in and the actual run time of the first job that breached its defined run time parameter. Customers can also configure the software to sound an audio alert or escalate the situation with e-mails and messages sent to pagers and cell phones.
The new feature can save a lot of time, says CCSS product manager Paul Ratchford. “Real-time identification and immediate access to work with a particular job affords managers the most efficient path to resolution,” he says in a press release. “Investigation time is eliminated and that’s very valuable for environments that are expected to perform at optimum levels on a 24/7 basis.”