[torquedev] gssapi branch update & bug in chk_file_sec.c

[Dave Jackson]

Hans,

  This should already be fixed in the latest TORQUE 2.2.0 trunk.
Perhaps we just need to do a merge/refresh to sync up the changes with
the GSSAPI branch.

  Garrick, can you do the merge?

Dave

On Fri, 2007-08-17 at 19:05 +0200, Hans Grobler wrote:
> I have an earlier version of the gssapi branch deployed and it's working
> reasonably well (after a lot of hacking). I'm now trying to update the
> deployment to the latest version. The first problem encountered is
> segfaults in pbs_mom:
> 
> Core was generated by `/usr/sbin/pbs_mom'.
> Program terminated with signal 11, Segmentation fault.
> #0  0xb7d4a731 in vsnprintf () from /lib/i686/cmov/libc.so.6
> (gdb) bt
> #0  0xb7d4a731 in vsnprintf () from /lib/i686/cmov/libc.so.6
> #1  0xb7d305a5 in snprintf () from /lib/i686/cmov/libc.so.6
> #2  0xb7f2874c in chk_file_sec (path=0x8104af8
> "/var/spool/torque/undelivered/", isdir=1, sticky=1,
>     disallow=2, fullpath=0, EMsg=0x0) at ../Liblog/chk_file_sec.c:312
> #3  0x0805e23f in main (argc=1, argv=0xbfb65524) at mom_main.c:6717
> 
> The recent change to chk_file_sec() is puzzling: it states that EMsg is
> supposed to be optional, however if there is an error it does a
> snprintf, causing the segfault. 
> 
> Comments?
> 
> Regards,
> -- Hans
> 
> _______________________________________________
> torquedev mailing list
> torquedev at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torquedev