[Klone-users] Session bugs

[Stefano Barbato]

Hi Mickael,

given the nature of uClinux the bug may be everywhere :) but you said  
that you've seen the same problem on Linux right?  In that case could  
you please provide me a minimal webapp/ directory that fires the bug?  
I'm not able to reproduce the bug myself.

Also, if you have not tried already, you can run the KLone daemon  
under valgrind on Linux (valgrind.org) and see if it reports any error.

Let me know,
stefano


On 06/mag/08, at 21:37, Mickael Auger wrote:

> Hi,
>
> (excuse me for multiple mails)
>
> A different back trace and logs of syslog :
>
> backtrace :
> =================================
> Program received signal SIGILL, Illegal instruction.
> 0x8 in ?? ()
> (gdb) bt
> #0  0x8 in ?? ()
> #1  0xaba34e in u_log_write_ex (fac=0, lev=7, flags=1, err=0,  
> file=0x9b9aa2 "", line=240,
>     func=0x9b9b9d "", fmt=0x9b9a95 "") at toolbox/log.c:135
> #2  0xab9b56 in u_path_snprintf (buf=0x9bc2e0 "s/%s", sz=4351,  
> sep=47, fmt=0x984cb8 "%s/%s")
>     at toolbox/misc.c:240
> #3  0xa892ac in http_alias_resolv (h=0xd98ad0, rq=0xbf5200,  
> dst=0x9bc2e0 "s/%s", uri=0x9bbae0 "",
>     sz=4351) at http.c:212
> #4  0xa894a6 in http_is_valid_uri (rq=0xbf5200, buf=0x9bd3f8 "/",  
> len=0) at http.c:235
> #5  0xa9a6d2 in request_set_uri (rq=0xbf5200, uri=0x9be43a "/",
>     is_valid_uri=0xa89304 <http_is_valid_uri>, arg=0xbf5200) at  
> request.c:642
> #6  0xa9e25c in request_parse_header (rq=0xbf5200,  
> is_valid_uri=0xa89304 <http_is_valid_uri>,
>     arg=0xbf5200) at request.c:1609
> #7  0xa8afde in http_serve (h=0xd98ad0, fd=4) at http.c:565
> #8  0xa8c36c in http_backend_serve (be=0xd98a78, fd=4) at http.c:849
> #9  0xaa630e in backend_serve (be=0xd98a78, fd=4) at backend.c:129
> #10 0xa82cb2 in server_be_serve (s=0xd98008, be=0xd98a78, ad=4) at  
> server.c:653
> #11 0xa83508 in server_dispatch (s=0xd98008, fd=3) at server.c:811
> #12 0xa83eb0 in server_loop (s=0xd98008) at server.c:979
> #13 0xa85f34 in app_run () at main.c:142
> #14 0xa86436 in main (argc=2, argv=0x9bfefc) at entry.c:408
> #15 0xabbd6e in __uClibc_start_main ()
> #16 0xabbd92 in __uClibc_main ()
> #17 0xa80054 in __exit ()
> #18 0xda96ae in ?? ()
> #19 0x48d7001c in ?? ()
> Ignoring packet error, continuing...
> Reply contains invalid hex digit 116
> (gdb)
> =================================
>
> logs of syslog :
> =================================
> <135>May  7 21:40:59 2008 syslog: [dbg][235:toolbox/hmap.c: 
> 1046:u_hmap_opts_dbg] [hmap options]
> <135>May  7 21:40:59 2008 syslog: [dbg][235:toolbox/hmap.c: 
> 1047:u_hmap_opts_dbg] size: 512
> <135>May  7 21:40:59 2008 syslog: [dbg][235:toolbox/hmap.c: 
> 1048:u_hmap_opts_dbg] max: 512
> <135>May  7 21:40:59 2008 syslog: [dbg][235:toolbox/hmap.c: 
> 1049:u_hmap_opts_dbg] policy: none
> <135>May  7 21:40:59 2008 syslog: [dbg][235:toolbox/hmap.c: 
> 1052:u_hmap_opts_dbg] ownsdata: 0, &f_free: 9bbce8
> <135>May  7 21:40:59 2008 syslog: [dbg][235:toolbox/hmap.c: 
> 1053:u_hmap_opts_dbg] no_overwrite: 0
> <135>May  7 21:40:59 2008 syslog: [dbg][235:toolbox/hmap.c: 
> 282:u_hmap_new] [hmap]
> <135>May  7 21:40:59 2008 syslog: [dbg][235:toolbox/hmap.c: 
> 283:u_hmap_new] threshold: 449
> <135>May  7 21:40:59 2008 syslog: [dbg][235:emb.c:31:emb_init]  
> registering embedded resources
> ...
> <135>May  7 21:41:01 2008 syslog: [dbg][235:toolbox/config.c: 
> 277:cs_getline] p == ((void *)0)
> <135>May  7 21:41:01 2008 syslog: [dbg][235:klog.c: 
> 367:klog_threshold] threshold == ((void *)0)
> <135>May  7 21:41:01 2008 syslog: [dbg][235:klog.c:386:klog_logopt]  
> options == ((void *)0)
> <132>May  7 21:41:01 2008 syslog: [wrn][235:klog.c: 
> 461:klog_args_check] threshold unspecified: assuming lowest possible  
> (DEBUG)
> <135>May  7 21:41:01 2008 : [dbg][235:server.c:1301:server_create]  
> configuring backend: app_http
> <132>May  7 21:41:01 2008 : [wrn][235:broker.c:107:broker_create]  
> CGI support disabled, use --enable_cgi to enable it
> <132>May  7 21:41:02 2008 : [wrn][235:broker.c:114:broker_create]  
> File system support disabled, use --enable_fs to enable it
> =================================
>
> Hope that help us ;-)
>
> Thanks.
>
> -- 
> Mickaël AUGER
> mickael.auger at gmail.com  
> _______________________________________________
> Klone-users mailing list
> Klone-users at koanlogic.com
> http://koanlogic.com/cgi-bin/mailman/listinfo/klone-users