[omniORB] #### You have caught an omniORB2 bug, details are as follows:
Bjoern Wennberg
bjornw@fairplay.no
Thu, 20 Aug 1998 18:17:43 +0200
Sai-Lai Lo wrote:
> Looks like it is a bug.
>
> The place where the message originates (file:
> src/lib/omniORB2/giopServer.cc line: 264) indicates that the ORB runtime
> thinks there is a disagreement in the size of the reply message that has
> been calculated at the beginning of marshalling and the actual number of
> bytes that have been sent.
>
> Could you modify the source to print out the value of WrMessageSpaceLeft()
> (giopServer.cc, line: 264) at the point the exception was thrown? Assuming
> that the application logic does not corrupt any data, the condition when
> this happen depends highly on what the reply values are.
>
> What are the types of the reply arguments?
>
> Sai-Lai
>
Ok, when my brain kicked in I realized that I created the bug :-)
I had defined an interface like:
typedef char stats_t[50 + 1];
struct my_struct {
stats_t stats;
};
and in the server-code I did:
char stats[50 + 1];
my_struct m;
memcpy(m.stats, stats, sizeof(stats));
which obviously isn't the right thing to do....
>
> >>>>> Bjoern Wennberg writes:
>
> > #### You have caught an omniORB2 bug, details are as follows:
> > ../giopServer.cc 264:GIOP_S::ReplyCompleted() reported wrong reply
> > message size.
>
> > I got the above error message from my server-app. What details do you
> > need to fix/reproduce ?
>
> > If you give me some advice I could debug the app here and probide
> > further info.
>
> > My platform is hpux-10.20 with dce-threads (compiled with aCC)
>
> --
> Dr. Sai-Lai Lo | Research Scientist
> |
> E-mail: S.Lo@orl.co.uk | Olivetti & Oracle Research Lab
> | 24a Trumpington Street
> Tel: +44 223 343000 | Cambridge CB2 1QA
> Fax: +44 223 313542 | ENGLAND
--
-------------------------------------------------------
Bjorn Wennberg email: bjornw@fairplay.no
ms: +47 950 82 657
Senior Programmer phone: +47 22405538
FairPlay International AS fax: +47 22405539