W3C home > Mailing lists > Public > www-lib@w3.org > January to March 1999

[www-lib] <none>

From: <Jim_Ravan@avid.com>
Date: Wed, 10 Feb 1999 15:11:03 -0500
To: www-lib@w3.org
Message-ID: <85256714.006EE07D.00@amm02.avid.com>


>
>Could this be connected to the problem pointed out by Raffaele Sena in
>
> http://lists.w3.org/Archives/Public/www-lib/1999JanMar/0190.html
>
>Do you stop/start libwww or the eventloop in between?
>
>    If that's the case, that should be fixed now (Henrik integrated the
last
>few changes
>    yesterday, I guess). I'm happily initializing/terminating the library
>many times, with no more
>    problems (but I never use the POST method, tough).
>
>-- Raffaele

Unfortunately, Raffaele's windows fix did not alter the problem. The socket
is still reporting that it has been shut down. And, no, we did not
stop/start libwww or the eventloop in between.

There is some odd conditional code in HTHost.c/HTHost_Register that says:

#ifdef WWW_WIN_ASYNC
/* Make sure that we are registered for CLOSE on windows */
event = *(host->events+HTEvent_INDEX(HTEvent_CLOSE));
HTEvent_register(HTChannel_socket(host->channel), HTEvent_CLOSE, event);
#endif /* WWW_WIN_ASYNC */

It seems strange because it is not protected by any sort of "if". So every
time any event is registered in the host, a CLOSE event is also
regisitered? Yes? No? I must be missing something. Can anyone explain what
this code is doing?

But it isn't the source of our problems. We removed this code because of
the portion of our log that said:

Event....... Register socket 16, request 00000000 handler 1000D8C0 type
HTEvent_CLOSE at priority 20
Event....... Registering socket for HTEvent_CLOSE
Event....... Register socket 16, request 00E05560 handler 1000D8C0 type
HTEvent_READ at priority 20
Event....... Registering socket for HTEvent_READ
Error....... Add  73    Severity: 1     Parameter: `WinSock reported
error=10058'       Where: `NETREAD'
Host kill... Pipeline due to HTEvent_CLOSE event
Host kill... Terminating net object 00E05140 from pipe line

Notice that a close event is registered before the read event at the same
priority. But the socket still returned Winsock:10058. We're very confused
right now.

So is anyone POSTing under NT with 5.2.1? If so, can you tell us what
you're doing? Or can someone with more experience in WinSock tell us under
what conditions 10058 might be returned?

-jim
Received on Wednesday, 10 February 1999 15:14:55 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 23 April 2007 18:18:28 GMT