On Tue, 23 Oct 2018 at 18:18, Robbie Gemmell <[email protected]> wrote: > > On Mon, 22 Oct 2018 at 19:03, Gordon Sim <[email protected]> wrote: > > > > On 22/10/18 13:47, Justin Ross wrote: > > > Hi, everyone. Please test this release candidate and vote based on what > > > you find. > > > > > > Source archive and signatures: > > > https://dist.apache.org/repos/dist/dev/qpid/cpp/1.39.0-rc1/ > > > > I am seeing failures in the 'unit_test' and 'sasl_tests' (running on > > fedora 27)... is it just me or are others seeing the same? > > > > I see some 'sasl_tests' and 'python_tests' failures in F27 using > qpid-python 1.37.0. All the tests passed on F26.
+1 Checked the signatures, checksums, verified LICENCE+NOTICE present, built and ran the tests, all good in one env, 1 seemingly env-specific failure in the other. The 'sasl_tests' failure I saw on F27 was because I didnt have nss-devel installed and the broker failed to start due to not understanding an option. I went to raise a JIRA but found there is one already, https://issues.apache.org/jira/browse/QPID-5449 I'm not sure about the 'python_tests' failure on F27 which I still see, but it seems to be env specific and perhaps relates to behaviour around the "badhostname" the tesst use. The F27 env is using IPv6 while the other is using IPv4 which could be an area for the different behaviour. In the F27 env the qpid.tests.messaging.endpoints.TimeoutTests.testConnectionOpen test minutes to run (vs a few seconds) but does eventually say it passed. The qpid.tests.messaging.endpoints.TimeoutTests.testSenderClose test later says it failed. However, if I run that test by itself it passes, and if I run all the TimeoutTests except testConnectionOpen then it also passes. I checked this with the 1.38.0 release as well and it does the same so whatever is doing on isnt a regression, would probably seem to be a python client or test issue if anything, and does seem env specific, so proceeding seem fine here. Robbie --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
