The cluster settings aren't added to the resume feature so when a resume state is saved it won't know the offset or cluster size should a work load be restarted. So if the cluster settings are changed some attempts may be skipped. For example; if --cluster 1:4 is used the first time and then it's later started with --cluster 2:4 the offset will be different by one but the save point from 5 minute intervals may be at different places for different systems so some small misses may occur.
This seems to be only a very minor issue and there's no need for concern if the same command line parameters are given on the same systems.
The cluster settings aren't added to the resume feature so when a resume state is saved it won't know the offset or cluster size should a work load be restarted. So if the cluster settings are changed some attempts may be skipped. For example; if
--cluster 1:4
is used the first time and then it's later started with--cluster 2:4
the offset will be different by one but the save point from 5 minute intervals may be at different places for different systems so some small misses may occur.This seems to be only a very minor issue and there's no need for concern if the same command line parameters are given on the same systems.