Merci, Emmanuel. Is there any workaround available? Or the workaround is switching to MINA 2.0?
Regards. 2010/3/26 Emmanuel Lecharny <[email protected]> > On 3/26/10 3:35 PM, Cleber Cassol wrote: > >> Hello again. >> >> Thank you in advance for your great support so far. As you may well >> remember, I was having a memory leak issue with MINA 1.1.7, but managed to >> solve it thanks to your insights. Now, I have a differente problem to deal >> with: sometimes, under not so heavy load (about 2 requests per second), >> the >> server CPU goes to 50%, 60% and sometimes to 100% usage. It stays there >> for >> some minutes, and then it goes down again for the normal level of 5-10%. >> >> > It may be caused by the selector() bug : sometime, under some specific > condition, the selector() methods returns immediately as if some sockect is > ready to be read, but without any socket ready. The problem is that in this > case, we renter into the selector() method immediately, and you can see that > it sucks 100% CPU very quickly. > > This problem has been fixed in 2.0.0-RC1. > > -- > Regards, > Cordialement, > Emmanuel Lécharny > www.nextury.com > > > -- Abraços!
