[omniORB] help

[bruno patin]

Hi,

I am at a loss to understand clearly what occurs and I could use some
help. We created a distributed appli where there are some corba objects
on these different servers. Randomly for one of these servers including
our special object, there is a crash of the program. It is false if we
put all these objects on only one server only '(same user app different
mapping). It is difficult to describe in more details but Have you some
ideas on how to search for this type of pb ?

thanks

[Duncan Grisby]

Post by bruno patin
I am at a loss to understand clearly what occurs and I could use some
help. We created a distributed appli where there are some corba objects
on these different servers. Randomly for one of these servers including
our special object, there is a crash of the program. It is false if we
put all these objects on only one server only '(same user app different
mapping). It is difficult to describe in more details but Have you some
ideas on how to search for this type of pb ?

Nobody can possibly help much unless you give us some kind of
information about the crash and the circumstances surrounding it. What
do you get, a segfault? An exception? What? What triggers it?

You could try running everything with -ORBtraceLevel 25
-ORBtraceInvocations 1 -ORBtraceThreadId 1 to see what's happening in
omniORB at the time of the crash.

Cheers,

Duncan.

[bruno patin]

Thanks for your answer on so few hints. Ok it seems it is correlated to
the load on the software. We are trying to reduce its volume in a
proportion that lets the error occurs and records what you asked for.
For the time being what we have as infos (using the debug level you
asked for) do not show really anything evident.

So thank you again and informations to come soon
B.Patin

Post by Duncan Grisby

Post by bruno patin
I am at a loss to understand clearly what occurs and I could use some
help. We created a distributed appli where there are some corba objects
on these different servers. Randomly for one of these servers including
our special object, there is a crash of the program. It is false if we
put all these objects on only one server only '(same user app different
mapping). It is difficult to describe in more details but Have you some
ideas on how to search for this type of pb ?

Nobody can possibly help much unless you give us some kind of
information about the crash and the circumstances surrounding it. What
do you get, a segfault? An exception? What? What triggers it?
You could try running everything with -ORBtraceLevel 25
-ORBtraceInvocations 1 -ORBtraceThreadId 1 to see what's happening in
omniORB at the time of the crash.
Cheers,
Duncan.