I have seen this happen occasionally on other projects too, in large part its caused by multiple builds from different projects queued up on a server and forking multiple jvms to execute unit tests and integration tests and crashing.
I bet it would work now if u trigger a manual build. For a project like Pirk Travis is a good option, I have seen Jenkins being more reliable than Travis for large scale integration tests involving MapReduce, Kafka and the big data friends. But Pirk is plain java for the most part, and shuld be good with Travis. On Thu, Jul 21, 2016 at 11:08 AM, Tim Ellison <[email protected]> wrote: > On 21/07/16 03:56, Apache Jenkins Server wrote: > > See <https://builds.apache.org/job/pirk/8/changes> > > > > Changes: > > > > [eawilliams] [trivial] -- Enhancements to build XML and Paillier class - > closes > > > > [eawilliams] [none] -- Replace multi if-else with switch stmts - closes > > > > ------------------------------------------ > > Started by an SCM change > > [EnvInject] - Loading node environment variables. > > JDK installation skipped: Unknown CPU architecture: ppc64le > > Building remotely on ubuntu-ppc64le (ppc64le) in workspace < > https://builds.apache.org/job/pirk/ws/> > > JDK installation skipped: Unknown CPU architecture: ppc64le > > JDK installation skipped: Unknown CPU architecture: ppc64le > > It looks like we were scheduled onto a poorly configured Jenkins node > causing this build break. > > Either we take it up with the build list, or work around it by tying our > job to a known-good node. Travis seems to be faster, so I haven't been > too fussed by this note. > > Anyone want to tackle this? > Tim >
