nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

As we transition to the new Hyland Community, some post links may be temporarily disrupted as we finalize redirects. We apologize for the inconvenience and are working to resolve this as quickly as possible. Details.

Short-lived persistence

antonylees

Hi all

I want to use activiti for short-lived workflow, similar to the way the HTTP session works in a web app, so I don't really need the persistence. I had the thought that I could use an in-memory database and clear it out at end of the process (perhaps as a listener on end end event), but is there a way to do this already?

Antony

Find more posts tagged with

Comments

ronald_van_kuij

This forum has a search option that probably would provide a result. But there is one http://lmgtfy.com/?q=activiti+memory+database

😉

antonylees

Yeah, that's really helpful thanks :?

There is a quesation about using an in-memory database, but that isn't my question. My question is about not persisting the data - an in-memory database will still keep the data in-memory. I don't want to do that, I only want the data to live for the life of the process. I could write a listener to listen for end tasks and clear the data out myself using JDBC. But does activiti provide this functionality already?

I did try searching first and didn't find the answer. If anyone has any actually helpful answers, please let me know

ronald_van_kuij

Sorry, I did not know you knew that an in-memory db was an option. I misread your question (I'm not a native English speaker)

There is currently not another option to achieve this. Then you'd be rewriting the persistence layer to exchange the in-memory db with an in-memory object list/map/… since the flow needs state. I think the work that needs to be put in there does not outweigh the slight advantage in performance. So using an in-memory db would be my suggestion.

antonylees

Hmmm, ok, shame. It looks like Drools Flow has this option but I do really like Activiti and, as the JVM might be running for a long time, I don't really want to create an ever expanding database/file

I'll have a look into the listener option - it seems like a fairly simple option

antonylees

Actually, re-reading your reply, I think you misunderstood me again

Using an in-memory database is fine, though the option not to (as with Drools Flow) would be nice. What I want is to not store the process data after it has ended

ronald_van_kuij

Take a look at this:

http://www.activiti.org/userguide/index.html#historyConfiguration

'none' will do exactely what you want.

ronald_van_kuij

if I understood you correctly this time 🙂

antonylees

I just tried it and yes, that is exactly what I want

What I hadn't realised is that the current execution ID is removed from the runtime tables and moved to the history tables when the process ends. Ideal

Many thanks

ronald_van_kuij

it's not even a move as far as I know. The history table is always 'up to date' with th runtime tables (if history is enabled)