On Jul 9, 2010, at 9:17 AM, Stefan Bodewig wrote: > On 2010-07-09, Jeremy Boynes wrote: > >> On Jul 9, 2010, at 8:57 AM, Stefan Bodewig wrote: > >>> I see. I'll switch Gump to use the package goal and add a new >>> definition running the install goal so we can track (wihout nagging for >>> now) if/how the Cactus problem can be resolved. > >> Maven's sequence is package, install, ..., integration-test so you >> might be able to use install directly (assuming Gump normally runs >> integration-test). > > No, install is what Gump has been using when the build failed. I think > it fails while building the war which is before it even tries to run the > tests (Gump also sets skipTests to true right now, not sure whether > integration-test honors that property).
My bad. I had Maven's lifecycle wrong. Integration-test runs between package and install. The cactifywar goal is attached to package so that may not help. I've reattached it to pre-integration-test as that seems more logical (assuming it runs before jetty run-war). > >> Is there a way to have Gump run with a profile (e.g. "-P gump") so >> that we could disable the tests there but still have them run in a >> normal build? > > Not yet, but I may be able to make it do that over the weekend. OK. Thanks Jeremy --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
