[Samba] Windows XP SP2 (Official) and Samba 3

[Thomas Bork]

Gerald (Jerry) Carter wrote:

 > | | Looks like MS did not fix the print change notify bug.

is this the bug, which you mean (think not, because this is the client 
part, not the server)?:

test2 # tail -f /var/log/log.smbd
[2004/08/10 23:05:30, 0] smbd/service.c:make_connection(381)
   make_connection: root logged in as admin user (root privileges)
[2004/08/10 23:10:53, 0] smbd/server.c:main(791)
   smbd version 2.2.10 started.
   Copyright Andrew Tridgell and the Samba Team 1992-2002
[2004/08/12 21:45:49, 0] smbd/server.c:main(791)
   smbd version 2.2.10 started.
   Copyright Andrew Tridgell and the Samba Team 1992-2002
[2004/08/12 21:49:43, 0] 
rpc_client/cli_spoolss_notify.c:spoolss_connect_to_clie
nt(91)
   connect_to_client: unable to connect to SMB server on machine R40. 
Error was :
  SUCCESS - 0.

 > Here's a patch.  Please test.
 >
 >
 >
 >
 > cheers, jerry
 > -----BEGIN PGP SIGNATURE-----
 > Version: GnuPG v1.2.4 (GNU/Linux)
 > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
 >
 > iD8DBQFBGqB+IR7qMdg1EfYRAm52AJ41/Dq4SpBfH8i9RK0MrWRKxDENAQCeLDdw
 > g5UutSpgD9NhV+683ow0Qv0=
 > =ux2J
 > -----END PGP SIGNATURE-----
 >
 >
 > ------------------------------------------------------------------------
 >
 > Index: source/rpc_server/srv_spoolss_nt.c
 > ===================================================================
 > --- source/rpc_server/srv_spoolss_nt.c    (revision 1741)
 > +++ source/rpc_server/srv_spoolss_nt.c    (working copy)
 > @@ -2830,6 +2830,12 @@
 >      info->data=NULL;
 >      info->count=0;
 >
 > +    /* a bug in xp sp2 rc2 causes it to send a fnpcn request without 
 > +       sending a ffpcn() request first */
 > +
 > +    if ( !option )
 > +        return WERR_BADFID;
 > +
 >      for (i=0; i<option->count; i++) {
 >          option_type=&(option->ctr.type[i]);
 >
 > @@ -2891,6 +2897,12 @@
 >      info->data=NULL;
 >      info->count=0;
 >
 > +    /* a bug in xp sp2 rc2 causes it to send a fnpcn request without 
 > +       sending a ffpcn() request first */
 > +
 > +    if ( !option )
 > +        return WERR_BADFID;
 > +
 >      get_printer_snum(p, hnd, &snum);
 >
 >      for (i=0; i<option->count; i++) {
 >

I will retest with the patch tonight. Thanks a lot :)

-- 
der tom
[fli4l-/eis-team]