I've found an issue in the jclouds CLI that makes all compute commands that list things fail: https://issues.apache.org/jira/browse/JCLOUDS-1199
It only affects the CLI (not the jclouds code nor karaf), and only the shell script (the CLI interactive mode works well). The fix I proposed in the issue is pretty straightforward and it would just require a small change to jclouds-karaf (to add a getter) and a small change to the CLI. Do you think it is worth canceling the RC2 and cutting a RC3 with this fix? If so, worth including this fix to the jclouds-labs oneandone provider? https://github.com/jclouds/jclouds-labs/pull/335 I'd pretty much like to include both, but I'd also like to release 2.0 asap (but in a good shape). The initial proposed release date was November 15th, and if we cut a new RC today, we still have the 72 hour (in working days) margin to vote and release, so we should still be on time to release without changing the initial date. Also, all validations made to the moment regarding live test execution, etc, would still be valid, since we would be only changing the unit tests in oneandone, and the jclouds-cli script. Validating RC3 should be just a matter of validating the signatures, and the release artifacts using the verification scripts, and validating the CLI fix, but all testing done for the RC2 should be valid. WDYT? On 8 November 2016 at 18:00, Ignasi Barrera <ignasi.barr...@gmail.com> wrote: > I am unsure. It's mostly the default settings for the TemplateBuilder need >> an update now and then. > > > This is something we have to fix, so we'd better have a JIRA or a PR for > it. It should be pretty straightforward to get the right one >
