Justin Erenkrantz <[EMAIL PROTECTED]> writes:
> On Fri, Jul 20, 2001 at 04:39:26PM -0400, Jeff Trawick wrote:
> > I just started looking at APR to see what to change but I don't see a
> > problem.
> >
> > I just played on Solaris 8... I don't see a problem there.
>
> Aha!
>
> I'm seeing ret being equivalent to 20507 which looks to be
> APR_ETIMEUP. Oh dear, don't we need to check if the return
> value is APR_ETIMEUP? -- justin
I sure didn't see that
[Fri Jul 20 16:35:14 2001] [emerg] (11)Resource temporarily
unavailable: n 0
I think we shouldn't use socket read functions for
files^H^H^H^H^Hpipes (whatever)
how can you get APR_ETIMEUP?
listensocks[0] is built by apr_socket_from_file() which doesn't
initialize the timeout field (so it remains zero)
apr_recv() thinks that sock->timeout==0 means that there is no timeout
so I don't see how you would get to the code that knows to return
APR_ETIMEOUT (i.e., wait_for_io_or_timeout()).
are you sure you're not running with other patches?
gotta go paint (really!)
I'll try to catch up with you when I can but that may not be until
tomorrow...
--
Jeff Trawick | [EMAIL PROTECTED] | PGP public key at web site:
http://www.geocities.com/SiliconValley/Park/9289/
Born in Roswell... married an alien...
