| Re: [ecf-dev] builder failures |
|
Thanks Markus.
Sorry to take advantage of your diligence, but do you have any insight why the L-HEAD-sdk-tests.feature is failing? [1]. It appears to be failing to setup the test vm: [TemplateProject] Starting builders from: Template-Test [workspace] $ /opt/hudson/tools/hudson.tasks.Ant_AntInstallation/apache-ant-1.8.1/bin/ant "-DTEST_DEBUG_RUN=-Xrunjdwp:transport=dt_socket,address=${TEST_DEBUG_PORT},server=y,suspend=n -Xdebug" -DUPSTREAM_BUILD=C-HEAD-sdk.feature Error occurred during initialization of VM Could not reserve enough space for code cache I haven't messed with the config, but I don't believe this has worked as expected since the changeover to java8/new Jenkins/Buckminster, etc. Thanksinadvance for any info. Scott [1] https://build.ecf-project.org/jenkins/job/L-HEAD-sdk.tests.feature/2849/console On 3/28/2016 1:39 PM, Markus Alexander Kuppe wrote: On 28.03.2016 19:11, Scott Lewis wrote:Over the past few days a number of build projects have been failing very early [1]. There have been no code changes, but it might have corresponded with a ddos attack at osu osl on the same subnet as our build machine [2]. It looks to me from the console logs as if the git task to retrieve projects into the workspace is failing somehow...perhaps because of what was done for the ddos by osu osl people? Do we need to notify the osu osl admin people...or does this look like something else?Hi, the DDos is unrelated. Instead, it seems as if a recent Jenkins plug-in upgrade has changed the git behaviour to clone directly into the workspace/ directory. It used to checkout into workspace/org.eclipse.ecf/. I changed the configs to explicitly clone into workspace/org.eclipse.ecf/ and the builds are back to green. M. _______________________________________________ ecf-dev mailing list ecf-dev@xxxxxxxxxxx To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/ecf-dev |