Send Event & Event Monitor parameters

[Anon Anon]

Hello,

It’s possible, that you can help me with one more problem? Or point me any document, where i can find it?

How I can configure two blocks in PAM – Event Monitor and Send Event?

I mean:

Send Event:
What value I have to put in:
- event name, type, source, destination
AND
- event parameters and expire after (sec.)

Event Monitor:
What value I have to put in:
- event name, type, source, destination, expression?

Also - how i can get and put values to these operatos??
What condition i have to put in EventMonitor to equal values with SendMonitor?


Thanks for help.

Regards,
Jack

[Legacy User]

I'm also looking to use PAM Send Event & Monitor operators but am having trouble finding any documentation behind it.

[Pier Olivier Tremblay]

Use case :

 

Workflow running for a long period of time with several approbations.

Event : Someone cancel the request

 

You have a process that trigger when someone cancel a request.

 

In it you can use : Send Event :

 

Event name : "Request Cancelled"

Event Type : "Request Cancelled"

source : Process.requestID

 

 

In the process with the approbation. you create another "corridor" (not sure if it's called like this in english version).

In it you add a "Monitor Event" operator, you link it at the beginning of your process, and link it to the Stop operator (Green sign). (Stop operator must be configured to "ignore running task : immediate stop")

You configure the "Monitor Event" operator to look at "Request Cancelled" event, with the source being the request related to the current process.

 

The Monitor Event operator will monitor as long as your process is running, when it finds an event with the parameters you entered, it will end the process going to the Stop operator.

 

After re-reading myself i found my explanations not crystal clear so here's a simple solution to test it :

 

Event monitoring process :

 

Process that sends an event

[Legacy User]

Thanks POSSq!

Starting to realize how much power this thing has. What are the limitations of the Event propagation? is it across the entire orchestrator for all workflows?

 

I'm envisioning the possibility for a master kill switch type of situation for orphaned processes.

[Pier Olivier Tremblay]

I have no idea, perhaps Michael or Andrew have the answer.

 

Pretty sure it's for all workflows, otherwise, it's kinda useless.

[Michael Niebuhr]

The event operator monitors for any event and matches it against its conditions, so in theory you could have one process instance issue an event that all other active instances would catch that event.   I suppose you could have a dedicated path that breaks from the start of the process and does nothing but listen for an event.

 

I have not played with this, but if it worked this could be a method for doing mass suspends and resumes, though as the total number of active instances climbs this could put excessive stress on the environment.