[omniORB] bug report: data alignment error in giopStream.cc

[Bastiaan Bakker]

This is a multi-part message in MIME format.

------_=_NextPart_001_01C1AFBF.F5049489
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi,

Attached is a small patch fixing the mentioned alignment bug. A quick =
scan of the code in giopStream and giopImpl* did not reveal similar =
cases: line 737 seems to be the only place where an integer is extracted =
before the GIOP message has been moved to the beginning of the data =
buffer.


> -----Original Message-----
> From: Duncan Grisby [mailto:dgrisby@uk.research.att.com]
> Sent: Wednesday, February 06, 2002 6:04 PM
> To: Bastiaan Bakker
> Cc: omniORB-list (E-mail) (E-mail)
> Subject: Re: [omniORB] bug report: data alignment error in=20
> giopStream.cc
>=20
>=20
>=20
> On Wednesday 6 February, "Bastiaan Bakker" wrote:
>=20
> > I ran into yet another bug which can crash omniORB4 corba=20
> servers. It's =3D
> > a data alignment problem in ensureSaneHeader() in=20
> giopStream.cc. Line =3D
>=20
> Thanks for your continued bug reports. I'm still away, but I'll look
> at them when I get back next week.
>=20

OK, I'll keep sending in reports for the bugs I stumble on :-)
=20
> > Does any of the omniORB developers know whether in other=20
> places similar =3D
> > flawed assumptions about alignment may have been made?
>=20
> There may well be other similar bugs. All of the transport code in
> omniORB 4 is new, so it's bound to have a few bugs. It has been
> extensively tested, including for interoperability with other ORBs,
> but there are, as you are finding, still some problems. That's why
> it's still in pre-release rather than even beta.
>=20

Which platforms do you use for testing?=20
In my experience it has been much easier to trigger race conditions on =
our X1 SPARC solaris box than on a dual Pentium III Linux box. =20

Regards,

Bastiaan Bakker
LifeLine Networks bv

> Cheers,
>=20
> Duncan.
>=20
> --=20
>  -- Duncan Grisby  \  Research Engineer  --
>   -- AT&T Laboratories Cambridge          --
>    -- http://www.uk.research.att.com/~dpg1 --
>=20

------_=_NextPart_001_01C1AFBF.F5049489
Content-Type: application/octet-stream;
	name="omniORB4.giopStream_alignment.patch"
Content-Transfer-Encoding: base64
Content-Description: omniORB4.giopStream_alignment.patch
Content-Disposition: attachment;
	filename="omniORB4.giopStream_alignment.patch"
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=

------_=_NextPart_001_01C1AFBF.F5049489--