The 'Number of Simulated Users to Group by' must be a FACTOR of the count of threads (users) in the relevant Thread Group. That's a given, without this you will always get the behaviour you describe - without any kind of timeout feature this is logical and easily solved by ensuring a factor or the user count is used.
But. I think there is an odd bug here where the first grouping of users is sometimes too low. It's proper intermittent, sometimes it happens, sometimes not, so I can only surmise it is related to the timestamp and maybe some sort of rounding error? Try: Thread Group - 12 Users, 16 Sec Ramp Up, 1 Iteration --Dummy Sampler ----Synch Timer - Group By 3 (3 is a factor of 12) About 50% of the time when I run this, the 1st group to get fired off in synch is only composed of 2 users, not 3. This results in one poor orphan user waiting for some friends to come along and the test does not terminate. The rest of the time each synch group is correctly made up of 3 users and the test completes cleanly so it's not a disaster but anyone aiming for more complex control with this Timer might fall prey to this problem. Can't see any other relevant defects on this so I raised https://issues.apache.org/bugzilla/show_bug.cgi?id=52185 Bug 52185 . ----- http://www.http503.com/ -- View this message in context: http://jmeter.512774.n5.nabble.com/Synchronizing-Timer-Issue-tp4990784p4992715.html Sent from the JMeter - User mailing list archive at Nabble.com. --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
