Hi Chuck,
I was building a clean checkout, with no changes. I did find the
immediate cause of the problem, though - I had a Tomcat instance running
on the default port of 8080, and these tests are creating a
SimpleHTTPServer for the client which apparently requires that port. I
don't know that this is a change from what used to happen before your
SimpleHTTPServer changes - I might not have run the build with Tomcat
active before.
After some digging around, I suspect the actual problem occurs because
org.apache.axis2.transport.http.server.HttpFactory defaults the port to
8080 unless this is overridden in the axis2.xml file. I'd suggest using
a less popular port as the default (or perhaps even trying a random
generated sequence of ports in the >1000 range until a free port is found).
- Dennis
Chuck Williams wrote:
Hi Dennis,
I committed the changes to SimpleHTTPSever and also run linux,
specifically debian etch (testing) with a 2.6.15 kernel. Before
committing, I did a clean build and verified all tests passed with Sun's
1.5.0_05 jvm. I upgraded today to 1.5.0.07_b03, just updated to latest
svn head (with no change since my commit is last revision) and did
another clean build. All tests pass, including the ones you cite below.
Do all tests pass with your changes using a revision before my
check-ins? Do your changes interact with the http server directly?
Please let me know if I can help to isolate your issue.
Chuck
Dennis Sosnoski wrote on 06/10/2006 07:59 PM:
I've been trying to synchronize with the latest code so I can get
changes checked in, but every time I try to build I get test failures
in the integration module
(org.apache.axis2.engine.EchoRawXMLOnTwoChannelsSyncTest FAILED,
org.apache.axis2.engine.EchoRawXMLMultipleTest FAILED,
org.apache.axis2.engine.EchoRawXMLOnTwoChannelsTest FAILED, etc.). I
even did a clean check out of the entire project, with the same
results when I tried to build.
I'm running Mandriva Linux 2006, Sun's 1.5.0-b64 JVM. I know there've
been changes to the SimpleHttpServer recently and I'm wondering if
that's the cause of the failures. Is anyone else building the latest
code on Linux successfully?
- Dennis
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
