Spotfire Ideas Portal
A watcher type process that monitors Webstatistica jobs that fail, and automatically restarts the failed job from on the windows scheduler. We often see a job that fails, for sometimes innocuous reasons like a DB connectivity timeout, but once the job is either re-run manually in Statistica or resubmitted via the Windows Scheduler it runs fine.
As a lot of our jobs are sequence dependent, i.e. job A must run before job B otherwise a warehouse table doesn't update etc etc, an auto restart of single failed jobs would remove the need to re-run a number of sequence driven jobs, and also the need for human intervention in the process.
An example of a crashed job from webstatistica Batch Jobs screen is shown in the attachment, along with some more information about how the process could work.
This would be a great tool to have. Since we are not able to resolve most of random failures, it would be really helpful to atleast rerun jobs. This will ensure that the reports are updated as expected when decision makers resume their office in morning.