I probably simply do not know the correct search terms for that issue.
At the moment, on my on-premise installation, I can see completed processes for around 48h, but we would love to move that up to 168h. Can I do that via the Admin Console? Or do I have to adjust some config file? In any case: Where do I do that?
Discussion posts and replies are publicly visible
I think what you're looking for is the Data Management available in the process model: docs.appian.com/.../process-model-object.html
The "system default" value is set in some system config file which I would've thought would be one of the standard ones (like custom.properties) though I haven't touched these in forever. Any chance your question is answered here? https://docs.appian.com/suite/help/21.4/Managing_Process_Archives.html
Edit to add: this link appears to contain the exact answer: https://docs.appian.com/suite/help/21.4/Data_Maintenance.html
Also just to note, each process model can be set independently, though sadly each process model's value must be hardcoded for some reason.
It sounded to me as if he was asking where you set the "system default auto-archive processes after __ days" value, though I could just be projecting there.
Ah yeah I think you're correct . You'll also have to confirm that the processes are actually using the default archive policy though.
To summarize, I have to proceed as follows, right?
I'm out of date on how to stop / restart the engines for a local install, but assuming your steps 1-4 and 6-9 are the correct steps and ordering, then this sounds good to go ;-)
All of the other responses address your actual question, whereas I want to challenge the underlying thinking here. Why do you want to increase the amount of time a successfully complete process instance remains in memory? Given that an application server has a limited amount of memory the objective should be to try and remove completed instances from memory as soon as possible. Processes that are still in flight or have problems are not candidates to be removed and it is right and proper these remain until they are successfully completed.
If your objective is to provide some means of introspection as to what has happened then a database-based audit trail will be a better option long-term as the data can be kept for a much longer period without affecting the performance or scalability of your application.
What is your use case for wanting to extend the period of time completed instances reside in memory?