[Gluster-devel] netbsd regression update : cdc.t

Krishnan Parthasarathi kparthas at redhat.com
Fri May 8 09:14:43 UTC 2015


There are no sleeps between polls in the event poll thread, ref: event_dispatch_poll(...).
I am not sure if we are referring to the same 'poll'. I haven't gotten a chance to look into
this. I will try adding logs when I get back to this.

~kp

----- Original Message -----
> Krishnan Parthasarathi <kparthas at redhat.com> wrote:
> 
> > I do. I see 2 threads 'waiting' on posix_health_check_thread_proc, 3
> > threads 'blocked' on select(2) in the changelog xlator and the remaining
> > in lwp_park. I still don't see thread 'waiting' on poll(2), listening for
> > events on socket fds registered in the 'poll set'.
> 
> There are sleeps between polls, right? I suspect you see the sleeps.
> You could log before and after poll() calls to make sure.
> 
> --
> Emmanuel Dreyfus
> http://hcpnet.free.fr/pubz
> manu at netbsd.org
> 


More information about the Gluster-devel mailing list