Marianne
The cleanup of logs does not really follow the 12 hour interval as for the normal image cleanup sessions running in the background.
It works more on a day to day basis. Every day at midnight or a few minutes over midnight, bprd makes a call to attempt to cleanup logs. If its just client then you would see something like - "bpclntcmd -clean_old_logs" running on the client machine. One of the hidden switches in bpclntcmd cli to cleanup logs
The parent bprd process forks a child which in turn loops through the server list (and the media_server list). It then makes a call to - " recycle debug logs on server <remote_server>"
The bpcd logs should show messages similar to below
bprd -dontfork -keeplogs <kp_value> -recycle_server <master_server>
The <kp_value> is from the global configuration on the master.
LogRecycle for unified logs is added to enable/disable log recycling for given originator. If enabled, log recycling will occur at the time of rollover.
If number of log files reach NumberOfLogfiles+1, oldest log file will be removed and new log file will be created.
If rollover policy is set to FileSize, logs should never use disk space more than (NumberOfLogFiles+1)*MaxLogFileSizeKB.
At the backend unified logs would be cleared off via the vxlogmgr command.
One of the old technotes in regards to these parameters for reference
https://www.veritas.com/support/en_US/article.100017250
Just noticed that in my previous post while adding it using my phone I did miss out of a couple of -s entries per option.
KirillKrotov The exact command above would be as below
vxlogcfg -a --prodid 51216 --orgid 132 -s RolloverMode=FileSize -s MaxLogFileSizeKB=512000 -s NumberOfLogFiles=15 -s LogRecycle=True