W3C home > Mailing lists > Public > www-lib@w3.org > July to September 2000

more https segmentation fault

From: Francois Nicot <fnicot@silicom.fr>
Date: Mon, 10 Jul 2000 17:50:19 +0200
Message-ID: <3969F0BB.901DBA1B@silicom.fr>
To: "www-lib@w3.org" <www-lib@w3.org>
Francois Nicot wrote:

> Hi all,
>
> I have a segmentation fault  (see traces below) when I insert  in my
> Robot the following code  extracted from libwww 5.3.0 /Examples/wwwssl.c
> :
>
> in "main" file
>     /* Register SSL stuff for handling ssl access */
>     HTSSLhttps_init(YES);
>
> ......
>
> in my terminate Handler :
>
>     /* Close down SSL */
>     HTSSLhttps_terminate();
>
> but I do not go  very far ! I get SegFault  at first HTLoadAnchor !
>
> Is it enough to use  https protocole or do I need to declare more things
> ?
>
> Of course,  I now use w3c-libwww-pre-5.3.0.
>
> Thanks for help.
>
> Francois Nicot
>
> *******************************************************
> traces  begin here:
>
> $4 = {
>   mainLink = {
>     dest = 0x0,
>     type = 0x0,
>     method = METHOD_INVALID,
>     result = HT_LINK_NONE
>   },
>   links = 0x0,
>   parent = 0x289e98,
>   children = 0x0,
>   sources = 0x0,
>   document = 0x289f28,
>   physical = 0x0,
>   address = 0x28c690 "https://www.alapage.com/",
>   isIndex = 0 '\000',
>   headers = 0x0,
>   header_parsed = 0 '\000',
>   variants = 0x0,
>   title = 0x0,
>   allow = METHOD_INVALID,
>   content_type = 0x278008,
>   type_parameters = 0x0,
>   meta_tags = 0x0,
>   content_base = 0x0,
>   content_encoding = 0x0,
>   content_language = 0x0,
>   content_length = -1,
>   content_location = 0x0,
>   content_md5 = 0x0,
>   cte = 0x0,
>   date = -1,
>   expires = -1,
>   last_modified = -1,
>   age = -1,
>   etag = 0x0,
>   derived_from = 0x0,
>   version = 0x0
> }
> (xxgdb) cont
> Program received signal SIGSEGV, Segmentation fault.
> 0xff152a54 in HTChannel_setSemaphore (channel=0x19d98, semaphore=0) at
> HTChannl.c:402
>
> (xxgdb) where
> #0  0xff152a54 in HTChannel_setSemaphore (channel=0x19d98, semaphore=0)
> at HTChannl.c:402
> #1  0xff157bf0 in HTHost_free (host=0x294088, status=-1) at
> HTHost.c:1119
> #2  0xff157c88 in HTHost_deleteNet (host=0x294088, net=0x289f78,
> status=-1) at HTHost.c:1133
> #3  0xff15c52c in HTNet_delete (net=0x289f78, status=-1) at HTNet.c:910
> #4  0xff2b9f0c in HTTPCleanup (req=0x293dc8, status=-1) at HTTP.c:168
> #5  0xff2bc27c in HTTPEvent (soc=0, pVoid=0x28c6b8, type=HTEvent_READ)
> at HTTP.c:1277
> #6  0xff2bb810 in HTLoadHTTP (soc=-1, request=0x293dc8) at HTTP.c:962
> #7  0xff15c2b0 in HTNet_newClient (request=0x293dc8) at HTNet.c:807
> #8  0xff160520 in HTLoad (me=0x293dc8, recursive=0 '\000') at
> HTReqMan.c:1643
> #9  0xff358e6c in launch_request (request=0x293dc8, recursive=0 '\000')
> at HTAccess.c:77
> #10 0xff359194 in HTLoadAnchor (anchor=0x0, request=0x293dc8) at
> HTAccess.c:208
> #11 0x5b0d0 in main (argc=5, argv=0xffbee844) at main.c:573
> (xxgdb)

I wanted to add to the previous mail that  the HTLoadAnchor  parameter
"anchor=0x0" displayed in the trace does not  match reality.
If you look at the HTLoadAnchor code you can see that it is not possible to
execute "launch_request "  if either anchor or request is null.
But I do execute HTLoadAnchor ...

Thanks for help.

Francois Nicot.
Received on Monday, 10 July 2000 11:46:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 23 April 2007 18:18:37 GMT