Last week we had an issue in production that caused it to freeze up for an hour.
Basically a custom Java application called an RFC that I wrote multiple times (20 or more) within a couple seconds and we ran out of dialog processes. Everything hung until the dialog processes timed out and then everything was back to normal.
What safety machanism can be put in place to prevent this? I know you can load balance and increase the number of dialog work process slots, but is there any way to gaurantee this won't happen again?
This is on R/3 version 4.7.
Thanks very much.