[Openswan Users] RE: Pluto core dumps FC4 + 2.3.1

[Paul Wouters]

On Wed, 22 Jun 2005, Andrew Baumhauer wrote:

> Subject: [Openswan Users] RE: Pluto core dumps FC4 + 2.3.1
> 
> Does anyone have any ideas on this?  How about the compiling problem with
> the latest CVS code on Fedora Core 4?

It is not clear to me whether your coredump is from the FC4 suplpied openswan rpm.
or our own rpms.

Compilation of CVS HEAD on fc4 is giving many warnings (treated as errors) due to
gcc-4.0.0 that comes with FC4. Try:

USE_LWRES=false make programs

to build openswan.

> I'm getting a segmentation fault when starting the connection with:
> /usr/sbin/ipsec auto -up SOMENETWORK

> Program terminated with signal 11, Segmentation fault.
> warning: svr4_current_sos: Can't read pathname for load map: Input/output
> error
> Loaded symbols for /usr/libexec/ipsec/pluto
> Reading symbols from /usr/lib/sse2/libgmp.so.3...done.
> Loaded symbols for /usr/lib/sse2/libgmp.so.3
> Reading symbols from /lib/libresolv.so.2...done.
> Loaded symbols for /lib/libresolv.so.2
> Reading symbols from /lib/libc.so.6...done.
> Loaded symbols for /lib/libc.so.6
> Reading symbols from /lib/ld-linux.so.2...done.
> Loaded symbols for /lib/ld-linux.so.2
> #0  informational (md=0x990e4b8) at demux.c:1047
> 1047    demux.c: No such file or directory.
>        in demux.c
> (gdb) bt
> #0  informational (md=0x990e4b8) at demux.c:1047
> #1  0x0019fcd6 in process_packet (mdp=0x21560c) at demux.c:2247
> #2  0x001a0b39 in comm_handle (ifp=0x990cf20) at demux.c:1167
> #3  0x00183907 in call_server () at server.c:1124
> #4  0x00180b85 in main (argc=9, argv=0xbfef5554) at plutomain.c:747

It seems gdb did not find all information needed to show the exact line
causing your segfault.

I think however, that this is a problem fixed in CVS, since around line 1047
there was a fix for some stray packets during startup:

             return STF_IGNORE;

         default:
#ifdef DEBUG
             if(st!=NULL
                && st->st_connection->extra_debugging & IMPAIR_DIE_ONINFO) {
                 loglog(RC_LOG_SERIOUS, "received and failed on unknown informational message");
                 return STF_FATAL;
             }
#endif
             if (pbs_left(n_pbs) >= sizeof(disp_buf)-1)
                 disp_len = sizeof(disp_buf)-1;
             else
                 disp_len = pbs_left(n_pbs);
             memcpy(disp_buf, n_pbs->cur, disp_len);

I believe the :
 	if(st!=NULL


was added in cvs to fix this error.

Paul