On 1/3/2012 9:19 PM, Rainer Jung wrote: > > I tried to reproduce with various AcceptFilter setting inclusing "https none" > using MSIE, > FF and Chrome. I always get the response on the first request. > > Steffen, Gregg et. al.: Can you reproduce on a test system? Did you already > reproduce once > with increased log level, e.g. > > LogLevel info ssl_module:trace8 mpm_winnt_module:trace8 > > or maybe, since it seems you can reproduce with a single request just use > > LogLevel trace8 > > and post one example for the working case and one for the broken case.
I'm not sure if this was lost in translation. This information would be very, very helpful, I haven't seen it posted yet. This appears to be an artifact of systems with the inability to recycle sockets. Unfortunately my network drivers have no problem recycling sockets, and nor do Ranier's. I believe the problem is that the disconnected-and-now-broken socket is used for AcceptEx. That AcceptEx'ed connection has no socket, and the connection is thrown away as AcceptEx reports the error. We'll need to 'preview' if the socket structure is alive. Or throw out this idea of recycling sockets altogether. If someone can provide the detailed error logging, we might work out the solution before 2.4.0 is tagged... but the request was hanging out there for a couple weeks so I'm not expecting to make much progress. And I don't want a 2.4.0 to wait on this issue.
