IBM® i was one of the first platforms on which IBM MQ was released and it was originally developed with a code base that depended on many IBM i specific features.  The MQ code was reworked in MQ V5.1 so that the IBM i implementation shared much more in common with other MQ platforms, but the team in Hursley kept some features which make MQ on IBM i unique - one of them being MQ's use of IBM i's journal objects for its transaction and recovery log. 

Journals are specifically designed to protect and recover changes to IBM i objects so they are an ideal choice for MQ's transaction log.  The concept of journal receivers closely mirrors MQ Multiplatforms' concept of log extents, but there is a key difference in that IBM i journal receivers cannot be changed once they have been detached.  This means that, unlike MQ log extents, journal receivers cannot be re-used. This is why MQ on IBM i supports linear logging but not circular logging, and why the enhancements which automate log management in MQ 9.0.2 by re-using log extents do not apply to IBM i.

As a consequence some work is necessary to manage MQ's journal receivers and to ensure that they do not fill up the disk.  Some time ago IBM published an article and sample program which helped to automate this process but the code became unavailable after a web site reorganisation.  We have had requests to re-release the code, so I am pleased to say that it is now available as an open source repository on the ibm-messaging community on GitHub:

https://github.com/ibm-messaging/mq-ibm-i-journal-sample/

Please give it a try and feel free to contribute to the repo.