I can indicate when the log gets created but there is clearly no mechanism to control it by size or to keep backups of it.
You need to find this folder, the path can be different depending on what is chosen
/opt/ca/aoPlatform/java/mdo-aggregator/bin
startAgg.sh
This script starts the Aggregator as a nohup process and redirects the output to the mdoaggregatorscrips.out file
nohup ./Aggregator >> $CA_EMM_HOME/logs/mdoaggregatorscript.out 2>&1 &
To loose the file output altogether, you could redirect the output to /dev/null
Or, change the >> to > so that it would get overwritten on every restart of the process - depends how often the process is restarted.
You could also use something like I've noticed in WVCtrl.sh to append a timestamp when the file is created
DATETIME="$(date +'%s')"
nohup ./Aggregator >> $CA_EMM_HOME/logs/mdoaggregatorscript_"$DATETIME".out 2>&1 &
(would have to test this one)
The aggregator should be 'aggregating' the data that comes in so you can see it over certain time periods, like 24 hours, 7 days etc, that would be the function of it.
I'm not sure why the log has got so large, if it hasn't been investigated with support, I would try to get a sample of it to us and raise a case for investigation.
Otherwise, I would suggest to raise an enhancement request that all processes in AXA, of which there are many, have controllable logging ( by size / backup)