Güvenal · Oct 17, 2017

Ensemble scheduler

We have had incidents with the scheduler in some integrations.
The problem is that the scheduler do not execute at seemingly random occasions.
Has anyone else experienced this? If so how have you handled it?
Is there a smart way to be informed about these non events.
I have an idea to build an integration that goes thru all the other integrations, looks at their schedules and compare it with lastrun information in someway but I am not sure if it is doable.
Another scenario occurs during maintenance windows, if there are "tasks" scheduled to be run during that time.
In my understanding they will not automatically run after the downtime if the time of the schedule has passed, how do you solve this?
Hope I have managed to make myself understood and that someone is willing to share if they have any input.

Best regards,


0 507
Discussion (3)1
Log in or sign up to continue

Take a look at class %SYS.Task.

All the properties you need to be able to assess whether a task has run and when the next run datetime will be, are there. Some of the property names are not that obvious so you might run a query against that class and then compare what you get with what you see in the Ensemble Task Scheduler.

We found the Ensemble Scheduler single threaded, so If you had schedules that would overlap some instances would not be started correctly. One of my colleagues created two simple command line scripts to start and stop interfaces. We are using those scripts along with the cron scheduler on AIX.

Also if You are running purge without keeping message integrity it uses to mess scheduling. But it's easily fixed by restarting production.

So we need to keep message interity on purges on namespaces where scheduling is used on business hosts.

Why would not keep message integrity on purge? Because purging is more efficient that way.

Best Regards