there is a regress on `testing/master` caught by our SI tests... this is an issue I was able to track back a few weeks ago (found comments in slack from jun 23 timefrawe): https://mesosphere.slack.com/archives/C0PQL46QJ/p1499708240150092
Jesus says it will be fixed in the next CLI release.. however all of our shakedown SI tests are broken until we resolve this. The issue manifests with the following error:
This issue is shakedown 1.4.3 uses 0.4.16 and it doesn't work with this version of DCOS. I'm not sure another the release of new CLI is the right fix.