Discussion:
[omniORB] Omniorb - working through firewall question
Jayaraman, Thirupurasundari
2012-03-03 15:30:13 UTC
Permalink
Hi All

We are using omniorb 4.0.7 for a client / server application and we have
used endpoints with ports, so that it can help communicate across
FIREWALL.

We would like to restrict all CORBA ports to a range of 2800 - 2900.

Recently this solution failed to work in a particular customer site
because they blocked all source and destination ports outside the range.

We configured a network firewall, blocked all destination ports except
2800-2900.

We see that CORBA is using RANDOM Source ports.

Given below is the output of the FIREWALL Log.

Date Time Action Source Destination Duration Bytes
Sent Bytes Received Application
2/27/2012 6:28:32 Permit 192.168.1.11:50368 172.16.1.11:2809
1799 sec 950 791 TCP Port 2809
2/27/2012 5:51:04 Permit 192.168.1.11:50983 172.16.1.11:2809
121 sec 1090 861 TCP Port 2809
2/27/2012 5:48:48 Permit 192.168.1.11:50979 172.16.1.11:2809
85 sec 1090 861 TCP Port 2809

We are trying to reach a destination 2809 and we are using the PORT as
part of endpoint, but we see internally, source ports > 50000 are being
used.

Is there a way we can control source ports?


Regards
Sundari.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.omniorb-support.com/pipermail/omniorb-list/attachments/20120303/eea946a6/attachment.htm
Loading...