[omniORB] Strange problem on HP-UX 11.00

[Srinivasan N.]

Hi,
    I'm not even sure if this is going to make sense but we have a very
curious situation with a product which uses omniORB 2.4 and i'd
appreciate any/all input. We've ported omniORB 2.4 to HP-UX 11.00 with
kernel threads and are using it as part of an application management
product.
    It's been working for quite some time but we've recently hit upon a
curious problem - CORBA calls (inter-process, intra-machine) which used
to take in the order of centi-seconds to complete now take in the order
of seconds. Also, this problem does not occur on all machines though
we're unable to figure out the diff between the machines.
    I was wondering if someone could suggest possible reasons for the
huge latency experienced in making calls on some machines? I could use
those to figure out what could be wrong with the servers.

    TIA.

--
Regards,
ns(N. Srinivasan)
------------------
"Technology is dominated by those who manage what they do not
understand."
--Murphy's Technology Law #3

----------------------------------------------------------------------------

Srinivasan N.                           Email    : srini_n1@verifone.com

Verifone(India)                         PHONE(W) : +91-80-5298151 x5122
Wind Tunnel Road,                       PHONE(H) : +91-80-5268766
Murugeshpalya,
Bangalore 560 017
India.
                                        FAX      : +91-80-5299876
----------------------------------------------------------------------------