I'm looking into this issue. The tests use an embedded naming server. We can
the tests independently. It seems that orb.destroy() doesn't really release
the port and it causes issues for the next test.
Thanks,
Raymond
--------------------------------------------------
From: "Wojtek Janiszewski" <[EMAIL PROTECTED]>
Sent: Tuesday, July 01, 2008 5:22 PM
To: <[email protected]>
Subject: Re: Fwd: [continuum] BUILD FAILURE: Tuscany - Apache Tuscany SCA
Implementation Project - Build Def:
It looks strange - it seems that even now (when I assume that build is not
working) port 11100 is being open.
In earlier version of tests for CORBA binding procedure was to spawn
separate OS process - tnameserv. I suppose that maybe something went wrong
and tnameserv wasn't closed by test. Can we try to locate and kill this
tnameserv?
If my guess is right next step I recommend is applying latest patch from
JIRA-2357. Why? Because tests for binding-corba-runtime contained ugly
bug, which was causing tests hanging - consequence is frozen build
process.
Thanks,
Wojtek
Luciano Resende wrote:
Looks like the build is still failing in Continuum with Corba
initialization failure due to address already in use.
---------- Forwarded message ----------
From: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Date: Tue, Jul 1, 2008 at 2:52 PM
Subject: [continuum] BUILD FAILURE: Tuscany - Apache Tuscany SCA
Implementation Project - Build Def:
To: [EMAIL PROTECTED]
Online report :
http://vmbuild.apache.org/continuum/buildResult.action?buildId=99264&projectId=277
