Fedora Account System
Red Hat Associate
Red Hat Customer
+++ This bug was initially created as a clone of Bug #1151244 +++ Description of problem: On a node, any files placed in `/var/lib/openshift/.cartridge_repository` that are not cartridge directories will cause mcollective to fail semi-silently. There is just a warning message reported, but the mcollective process continues without error. However, no messages are serviced or sent. Version-Release number of selected component (if applicable): 2.1.6 How reproducible: Always Steps to Reproduce: 1. Place a file in `/var/lib/openshift/.cartridge_repository` $ touch /var/lib/openshift/.cartridge_repository/test_file 2. Restart mcollective 3. Actual results: The following warning is seen in the ruby193-mcollective.log: WARN -- : agents.rb:95:in `rescue in activate_agent?' Agent activation check for openshift failed: Errno::ENOTDIR: Not a directory - /var/lib/openshift/.cartridge_repository/test_file No messages are serviced. oo-mco ping still works, however actions (such as removing/adding ssh keys to applications) always fail. Expected results: Warning message reported, mcollective continues un-phased. Additional info: --- Additional comment from RHEL Product and Program Management on 2014-10-09 15:35:59 MST --- Since this issue was entered in bugzilla, the release flag has been set to ? to ensure that it is properly evaluated for this release.
Fixed in https://github.com/openshift/origin-server/pull/5864
Commit pushed to master at https://github.com/openshift/origin-server https://github.com/openshift/origin-server/commit/430cfb25036ce566aa38266190c965efdc8e853a Bug 1151648 - Skip any files in .cartridge_repository directory
Checked on devenv_5227, issue has been fixed. With random file touched under /var/lib/openshift/.cartridge_repository Restart ruby193-mcollective service will not report error in /var/log/ruby193-mcollective.log and app actions will not fail. Move bug to verified.