On Sun, Apr 25, 2010 at 06:48, Amila Suriarachchi <[email protected]> wrote: > > > On Sat, Apr 24, 2010 at 12:21 PM, Andreas Veithen > <[email protected]> wrote: >> >> All, >> >> There are currently multiple issues with the build from trunk: >> >> 1. An out of memory condition as reported by Katherine Sanders in >> AXIS2-4691. I was able to reproduce this on a Windows virtual machine, >> but also to work around this by increasing the heap size allocated to >> Maven (which seems to contradict Katherine's findings). Probably this >> issue is not related to a specific commit. > > I could build axis2 properly in my linux machine with java 1.6. But anyway > by default I have increase the > memory 2048M. I think increasing the memory is not a work around isn't it?
No, but we should try to make sure that the build works out of the box. Do you know if it is possible to specify the heap size in the POMs, or to configure the POMs such that the build fails immediately with a clear error message when the heap size is insufficient? >> >> 2. r937071 causes test failures in jaxws-integration, as detected by >> Hudson. >> >> 3. Even after increasing the heap size and locally reverting r937071, >> I still get a build failure in the integration module: > > is this in window or your machine? I could build successfully with r937743. That's normal because in the meantime I fixed the build: I corrected r937071 to work with Java 1.5 and I rolled back another commit (see AXIS2-4687). > thanks, > Amila. >> >> >> ------------------------------------------------------------------------------- >> Test set: org.apache.axis2.deployment.WSDL11ToAxisServiceBuilderTest >> >> ------------------------------------------------------------------------------- >> Tests run: 5, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 1.625 >> sec <<< FAILURE! >> testVersion(org.apache.axis2.deployment.WSDL11ToAxisServiceBuilderTest) >> Time elapsed: 1.11 sec <<< FAILURE! >> junit.framework.AssertionFailedError: org.custommonkey.xmlunit.Diff >> [different] Expected attribute value 'getVersion' but was 'parameters' >> - comparing <mime:content part="getVersion"...> at >> /definitions[1]/binding[3]/operation[1]/input[1]/content[1]/@part to >> <mime:content part="parameters"...> at >> /definitions[1]/binding[3]/operation[1]/input[1]/content[1]/@part >> >> at junit.framework.Assert.fail(Assert.java:47) >> at >> org.custommonkey.xmlunit.XMLAssert.assertXMLEqual(XMLAssert.java:125) >> at >> org.custommonkey.xmlunit.XMLAssert.assertXMLEqual(XMLAssert.java:255) >> at >> org.custommonkey.xmlunit.XMLAssert.assertXMLEqual(XMLAssert.java:200) >> at >> org.custommonkey.xmlunit.XMLTestCase.assertXMLEqual(XMLTestCase.java:235) >> at >> org.apache.axis2.deployment.WSDL11ToAxisServiceBuilderTest.testVersion(WSDL11ToAxisServiceBuilderTest.java:57) >> >> This is probably caused by r937436 (to be confirmed). >> >> >> I will temporarily revert the commits that cause build failures (and >> reopen the corresponding JIRAs where applicable), so that we can get >> back to a stable baseline. >> >> Some reminders: >> * You are supposed to run the build locally before checking changes >> into SVN. Of course, sometimes problems are environment specific and >> not caught at this stage. >> * We have continuous integration (Hudson) in place for Axis2. See [1]. >> Unfortunately notifications to the mailing list don't work yet, but >> you might want to check manually. >> * It is possible to cleanly rollback a commit using "svn merge -c >> -XXXXXX . ." where XXXXXX is the revision number of the commit. It can >> later be reapplied using "svn merge -c XXXXXX . ." or "svn merge -c >> -YYYYYY . ." where YYYYYY is the revision number of the commit where >> the rollback happened. >> >> Andreas >> >> [1] http://hudson.zones.apache.org/hudson/view/Axis2/job/Axis2/ >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] >> > > > > -- > Amila Suriarachchi > WSO2 Inc. > blog: http://amilachinthaka.blogspot.com/ > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
