Is it possible to trigger an alert via optimize rule when the wait time of a process step exceeds a fixed value like 30 mins ? When i look at the intrinsic process step metrics the step wait time is defined as fact with “average” calculation which i cannot use in my case. I need more like an event/threshold rule. Looks like the only way to accomplish this is to add a timer to the corresponding step in the model and transition it to the alert step.
Holger, As i already mentioned that is the last option I have.
I was wondering if there is a way to monitor this outside the model and raise alert without any code/model change. We may get asked to monitor such “wait time” SLA of any step in the process and i don’t want to keep such timeout in all the steps for such cases.
If i understand correctly, you can achieve desired feature by creating Instant Event Rule which you can create on below mentioned default Process Intrinsic Data Metrics.
Step Timeout -The number of times the step timed out before completing successfully
Wait Time by Step - The amount of time that the step spent waiting to run, calculated from the end of the previous step to the start of the current step.
Open an of aforementioned kpi detail pages and top right you can see Create Rule button, using this you can create a rule which send alert to based on “Execution On” value. Ensure setting this as Event-Every Occurrence (real time) rather than KPI-Aggregation Interval.
The rule seem to get fired only when that step completes processing which is not what i want. i need an alert if the step wait time exceeds 30 minutes even if the step waits and completes execution after an hour. Seems like the wait time is calculated only after the step completes followed by alert and not like on the fly to trigger any alert.
have you come up with a viable solution? I’d just create a service that would periodically retrieve information about processes and steps using the services from WmMonitor and then analyze it. And send a message if something is wrong. It’s a bit of work, but then you can have a solution that perfectly fits your needs.
Actually, we’ve done something similar and it works well. The objective was a bit different, but the approach was as I described.