Hi guys, yesterday, I conducted a load test against a LDAP server, using an injector that spawn many threads and run bind/N searches/unbind sequences. Most of the time, it works well, but from time to time, I get some connection errors.
I'm trying to figure out if the pb is on the way we deal with connections closure or if it's the server itself that is so loaded that it refuses a connection. I tend to think that the pb is on the server, not on the API? but I'd like to check if we correctly handle teh connection closure on the API. I'll review this part and give you some feedback. Thanks !
