Channel freeze on getCloseFuture()deadlock?

이희승 (Trustin Lee) trustin at gmail.com
Tue Aug 25 03:46:50 EDT 2009


Hi Rafael,

IIRC, this dead lock issue is a known issue in mina-statemachine.
Neither Niklas or I had a chance to review the issue yet
unfortunately.  I will definitely take a look, but you might want to
contribute by fixing the bug by yourself just in case I'm late.

Trustin

On Thu, 20 Aug 2009 19:39:17 +0900 (KST)
Rafael Marins <rafael.marins at neociclo.com> wrote:

> 
> Hi Trustin,
> 
> 
> Trustin Lee wrote:
> > 
> > Do you have any full thread dump?  We did not experience such a dead
> > lock in the existing transports.  Perhaps you are synchronizing over
> > some object that Netty uses for synchronization?
> > 
> 
> It's the first time I produce a full Thread dumping. See the dump
> file (ZIP) in attachment, where I see something about a deadlock.
> 
> http://n2.nabble.com/file/n3478227/deadlockDump.zip deadlockDump.zip 
> 
> Seems to be on StateMachine, which is locked on DefaultStateContext
> object. According to my implementation there are 2
> DefaultStateContext objects in place, once my transport is compound
> of Sink + 2 Protocol layer handlers (each handler managed by a
> stateMachine).
> 
> I have no idea yet to resolve such deadlock. Any tips?
> 
> PS: Of course, I intend share my experiences with all netties on
> implementing transport and also making it work with MINA
> StateMachine. :-)
> 
> Best regards,
> 
> -----
> Rafael Marins
> Neociclo - http://www.neociclo.com
> 
> 



More information about the netty-users mailing list