--
You received this message because you are subscribed to the Google Groups "Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/4915c03a-d298-4059-b410-d14e2a9d981e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Is there a good metric to reduce the number of executors by?
To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/9b95be75-2f69-43d8-8fdc-af0e04a7baf3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/9b95be75-2f69-43d8-8fdc-af0e04a7baf3%40googlegroups.com.
No executors on a slave run in the same JVM.
From my experience, when possible and for non trivially might builds I'd recommend to run only one executor per slave. This way you won't have difficult to predict instabilities when X ram-greedy builds will run simultaneously on the same slave.
If you can afford using VM, then just spawn smaller ones but more of them.
My 2 cents
--
You received this message because you are subscribed to the Google Groups "Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/fdffd9ea-085f-405e-a66e-9aaf0ba36645%40googlegroups.com.