bug #94 news

Hi,

I have some news on Bugzilla bug #94:

In the nt branch I have applied the Ribeiro's patch, but it seem's don't
stop the problem. So I have added the following lines to http.c:

Index: http.c
===================================================================
RCS file: /cvsroot/squid/squid/src/http.c,v
retrieving revision 1.20.6.5
retrieving revision 1.20.6.6
diff -w -u -r1.20.6.5 -r1.20.6.6
--- http.c 24 Sep 2002 17:45:31 -0000 1.20.6.5
+++ http.c 30 Sep 2002 17:18:30 -0000 1.20.6.6
@@ -947,8 +947,11 @@
         cfd = -1;
      else if (entry->mem_obj == NULL)
         cfd = -1;
- else
+ else {
         cfd = entry->mem_obj->fd;
+ if (FD_SOCKET != fd_table[cfd].type)
+ fatalf("httpSendRequest(): FD type not FD_SOCKET - FD#: %i,
type:%u, desc: %s\n", cfd, fd_table[cfd].type, fd_table[cfd].desc);
+ }
      assert(-1 == cfd || FD_SOCKET == fd_table[cfd].type);
      if (p != NULL)
         httpState->flags.proxying = 1;

After some testing with datacomm-1 test of Web Polygraph, I have discovered
some interesting things about.

- on nt branch, it seems to be systematic when Squid is compiled with
--enable-icmp, with the following fatal info:
FATAL: httpSendRequest(): FD type not FD_SOCKET - FD#: -1, type: 7242316,
desc:
So, please someone can run a datacomm-1 with 10 req/sec on HEAD ? I need to
know if it is a nt or a HEAD problem.

- When ending the datacomm-1 sometime I get the error, whit the following
fatal info:
FATAL: httpSendRequest(): FD type not FD_SOCKET - FD#: 15, type: 0, desc:
May be an incomplete request ?

Regards

Guido

-
=======================================================
Serassio Guido
Via Albenga, 11/4 10134 - Torino - ITALY
E-mail: guido.serassio@serassio.it
WWW: http://www.serassio.it
Received on Mon Sep 30 2002 - 13:45:12 MDT