Fedora Account System
Red Hat Associate
Red Hat Customer
Description of problem: Create scalable app and add the customized haproxy_ctld.rb to the app git repo and push to server. The app will use the customized haproxy_ctld.rb for auto scaling. But after user remove the file from .openshift/action_hook and push the change to server, the gear will keep using the customized haproxy_ctld.rb, restart app cannot fix the issue. The process of the custom haproxy_ctld will not be killed in gear. Version-Release number of selected component (if applicable): devenv_4951 How reproducible: always Steps to Reproduce: 1. Create scalable app 2. Add a custom haproxy_ctld.rb to the .openshift/action_hooks and git push 3. Remove the custom haproxy_ctld.rb and push it again Actual results: The app will still using the custom haproxy_ctld.rb and the process still leaves there even restart the app. Expected results: Should use the default haproxy_ctld.rb when user removed the custom one. Additional info:
https://github.com/openshift/origin-server/pull/5586
Checked on devenv_4955, the issue has been fixed, after remove the action hook, it will fall back to use the cartridge default haproxy_ctld.rb for auto scaling. But the output is a little confuse from my side, it said the action hook removed but restarted after that. remote: HAProxy already running remote: Action hook haproxy_ctld.rb removed remote: Action hook haproxy_ctld.rb restarted remote: HAProxy instance is started @bparees Not sure if this is a typo, seems you meant to say restart w/ the normal cart haproxy_ctld.rb?
Yes that's a typo in the output message. master: https://github.com/openshift/origin-server/pull/5588 stage: https://github.com/openshift/origin-server/pull/5590
remote: HAProxy already running remote: Action hook haproxy_ctld.rb removed remote: Cartridge haproxy_ctld.rb started remote: HAProxy instance is started Checked on current STG (devenv-stage_907), issue has been fixed.