Jobs and Schedules

The Data Virtuality Server provides a mechanism for getting tasks automatically done. This mechanism consists of two parts: jobs and schedules. A job selects what has to be done, and a schedule defines when and how often it has to be done. To run a job, create and enable a schedule for it. Please remember that schedules are associated with a specific job and cannot be assigned to more than one job. You can also control job execution using job queues managed by job queue handlers.

There are different kinds of jobs:

• optimization jobs
• SQL jobs
• statistic jobs
• system jobs
  • cleanUp jobs
  • performance metrics collection task
  • backup jobs

There are also different types of schedules:

• once
• interval
• cron
• chained

The jobs can have the following execution statuses:

• RUNNING
• SUCCESS
• FAILED
• INTERRUPTED (set when the job execution is cancelled or the underlying query is killed)
• FAILED_AUTORETRY (set when the job has failed and is configured to retry automatically)
• NOT_STARTED - set when
  • all base schedules and/or jobs for an atomic chained schedule are triggered, but some of these jobs finished with unsatisfactory results
  • job is triggered, but disabled
  • if a QUEUED job is disabled
  • job is triggered but hits the limit of parallel runs allowed for it
• QUEUED (set when the job is placed in the queue)

All jobs are stored in the SYSADMIN.ScheduleJobs table and the schedules are kept in the SYSADMIN.Schedule table.

Jobs statuses are logged to server.log at the following levels: