Re: o Re: sequence of octets bug in Orbit when running server as root?t

[Claude Beazley <cbeazley hgmp mrc ac uk>]

When printing out the length of the octet-sequence that the client tries to 
pass to the server, the maximum sequence length that the server as root 
accepts (on my PC. anyway) appears to be  262044 

>On Monday 23 Sep 2002 10:27 am, Claude Beazley wrote:
> I've tried different sizes of octet sequence and it works for smaller
> sequences under root, but falls down on the larger ones. So it appears that
> when running as root, Orbit2 places an upper limit on the size of an octet
> sequence. The server running as normal doesn't seem to have this upper
> limit. Is this a feature or bug?
>
> >On Thursday 19 Sep 2002 1:46 pm, Claude Beazley wrote:
> > Here are the traces:-
> >
> > -------------------------------------------------------------------------
> >-- ----------------------
> > -------------------------------------------------------------------------
> >-- ---------------------- Running the server as a normal user was
> > successful and gives the trace:- p18972 :
> > ([0x80500b8])->testoctetsequenceupload (seq[6665311]={ 0x2d,... ...,0x2d
> > }, )
> >
> > ** (process:18972): WARNING **: CORBA_ORB_destroy: ORB still has 2 refs.
> >
> > ** (process:18972): WARNING **: ORB: a total of 7 refs to 4 ORB objects
> > were leaked
> > g_thread supported!
> > Written the file embcas.ior with the IOR of this server.
> >  Now waiting for requests...
> > Entering impl_test_testoctetsequenceupload
> >
> > -------------------------------------------------------------------------
> >-- ---------------------
> > -------------------------------------------------------------------------
> >-- --------------------- running server as root fails and gives:-
> >
> > ** (process:18921): CRITICAL **: file linc-protocols.c: line 1050
> > (linc_protocol_destroy_cnx): assertion `proto != NULL' failed
> >
> > ** (process:18921): WARNING **: CORBA_ORB_destroy: ORB still has 2 refs.
> >
> > ** (process:18921): WARNING **: ORB: a total of 7 refs to 4 ORB objects
> > were leaked
> > g_thread supported!
> > Written the file test.ior with the IOR of this server.
> >  Now waiting for requests...
> >
> > -------------------------------------------------------------------------
> >-- ----------------------
> > -------------------------------------------------------------------------
> >-- ----------------------
> >
> > >On Wednesday 18 Sep 2002 9:29 am, Claude Beazley wrote:
> > > >On Tuesday 17 Sep 2002 5:48 pm, you wrote:
> > > > Hi Claude,
> > > >
> > > > On Tue, 2002-09-17 at 10:50, Claude Beazley wrote:
> > > > > I'm trying to run a local corba server as root.
> > > >
> > > > 	That's intrinsically quite a risky thing to do you know ? :-)
> > > > especially so in older ORB versions than ~2.4.2
> > >
> > > Yeah, I do, but I need root for certain tasks, so I automatically drop
> > > down seteuid to a user for everything else and when root is required I
> > > can temporarily  seteuid back up again. I have just updated to the
> > > latest version I can find. "Hyper unusual..." no longer occurs but
> > > client still fails as before. I will generate the debug output sometime
> > > tomorrow. thanx
> > >
> > > claude
> > >
> > > > > The server works fine in  accepting ints, strings,sequences of
> > > > > strings and other variables  from the client, however it falls
> > > > > apart when trying to accept a sequence of octets:-
> > > >
> > > > 	Most odd;
> > > >
> > > > > The server never enters testoctetsequenceupload but
> > > > > prints out the glib warning:-
> > > > > ** (process:28210): WARNING **: Hyper unusual code path of little
> > > > > testing
> > > >
> > > > 	Ok - so that code isn't there anymore; upgrade the ORB and see what
> > > > happens. Also configure it with --enable-debug, then export
> > > > ORBIT2_DEBUG=traces, and post what you see printed out here.
> > > >
> > > > > ORBit2-2.4.0.0.200206240452-0.snap.ximian.1
> > > > > ORBit2-devel-2.4.0.0.200206240452-0.snap.ximian.1
> > > >
> > > > 	Grab a new snapshot, that's 3 months old :-)
> > > >
> > > > 	HTH,
> > > >
> > > > 		Michael.
> > >
> > > -------------------------------------------------------
> > >
> > > _______________________________________________
> > > orbit-list mailing list
> > > orbit-list@gnome.org
> > > http://mail.gnome.org/mailman/listinfo/orbit-list
> >
> > _______________________________________________
> > orbit-list mailing list
> > orbit-list@gnome.org
> > http://mail.gnome.org/mailman/listinfo/orbit-list
>
> _______________________________________________
> orbit-list mailing list
> orbit-list@gnome.org
> http://mail.gnome.org/mailman/listinfo/orbit-list