Page tree
Skip to end of metadata
Go to start of metadata


Microblog automatically creates drafts of microposts which haven't been sent. Sometimes it's advised to clean up the draft cache memory.

What are Microblog drafts

The Microblog creates draft versions of microposts.

A draft is created and stored when a user enters the text editor but then reloads the tab without publishing their micropost.


Usually, drafts that are older than 6 hours are cleaned up by a scheduled cron job. The Microblog tries to delete up to 1000 drafts (which are older than 6 hours) with every scheduled job.




How to clean-up Microblog drafts

Microblog drafts should be automatically cleaned-up by the Microblog every hour.

However, prior to Microblog 5.1.4, there was a bug which prevented the scheduled job from running, therefore possibly leading to large amounts of stored drafts.

We fixed this bug in Microblog 5.1.4, but in case you are still having some issues with the drafts, let's see how to clean them up manually.


Navigate to Confluence administration → Administration → Scheduled Jobs.

Once there, search for "Microblog - Draft CleanUp".

Next to the scheduled job's name, click on the Run button.

If you suspect that there are more than 1000 drafts stored in your system, let the job run several times, just to be sure.


Please note that this action might take some time, depending on the amount of drafts that have to be cleaned up.




Your feedback

We would love to hear your feedback on this matter. Right now, the scheduled job runs hourly, and deletes up to 1000 drafts which are older than 6 hours.

When that's not enough for you, because, for example, users in your system create more than 1000 drafts every hour, please let us know.

Perhaps we can tweak the settings so they fit more use cases. Currently we don't know of any use case where this would be needed.


You can contact us via the intercom, the contact forms on our website or you can visit the canny board for the Microblog and create new feature requests or vote on feature requests sent in by the community.