Multi Instance Trigger Capability (Magic xpi 4.6)
You can use multiple instances of some Magic xpi triggers in your projects. Since Magic xpi 4.1, the distribution of triggers over the project’s engines is configurable at deployment time. This not only allows you to control where each trigger will run, but also to set the same trigger (for example, a WebSphere MQ trigger defined in Flow 4 of BP1 at design time) to run on more than one engine. This lets you scale the trigger processing capability.
The following table explains which triggers can be used in this way.
|
|
|
Directory Scanner
|
Yes
|
|
Email
|
No (conditional)
|
Possible duplicate results if the Leave On Server parameters are set to Yes. There is no performance gain.
|
Exchange
|
No
|
Possible duplicate results, and the persistency file may be overwritten. There is no performance gain.
|
HL7
|
No
|
Opens TCP port. Only one instance per machine.
|
IBM i
|
Yes
| |
JMS
|
Yes
|
|
MSMQ
|
Yes
|
|
Salesforce
|
No
|
Possible duplicate results, and the persistency file may be overwritten. There is no performance gain.
|
SAP R/3
|
Yes
|
|
SAP A1
|
Yes
|
|
SAP Business One
|
No
|
Possible duplicate results. There is no performance gain.
|
Sugar
|
No
|
Possible duplicate results, and the persistency file may be overwritten. There is no performance gain.
|
TCP Listener
|
No
|
Opens TCP port. Only one instance per machine.
|
WebSphere MQ
|
Yes
|
|
SDK Trigger (external)
|
Yes
|
|
SDK Trigger (polling)
|
Depends
|
Dependent on the implementation.
|