Description of problem: Two different environment variables used to control if NPD install : Using ansible playbook: openshift_node_problem_detector_state=present During cluster deployment (on Jenkins): openshift_node_problem_detector_install: true It would be confusing for the users. Version-Release number of selected component (if applicable): [root@qe-weinliu310-2master-etcd-1 ~]# oc version oc v3.10.0-0.30.0 kubernetes v1.10.0+b81c8f8 features: Basic-Auth GSSAPI Kerberos SPNEGO Server https://qe-weinliu310-2master-etcd-1:8443 openshift v3.10.0-0.30.0 kubernetes v1.10.0+b81c8f8 [root@qe-weinliu310-2master-etcd-1 ~]# uname -a Linux qe-weinliu310-2master-etcd-1 3.10.0-862.el7.x86_64 #1 SMP Wed Mar 21 18:14:51 EDT 2018 x86_64 x86_64 x86_64 GNU/Linux [root@qe-weinliu310-2master-etcd-1 ~]# cat /etc/redhat-release Red Hat Enterprise Linux Server release 7.5 (Maipo) How reproducible: always Steps to Reproduce: 1. As description Actual results: Two different environment variables used Expected results: Unique environment variable used to control both of the two install methods, to avoid confusing. Additional info: N/A
https://trello.com/c/OA9z4cMU/701-8-node-deliver-node-problem-detector-for-tech-preview says NPD is tech-preview therefore not a 3.10 blocker here.
(In reply to Scott Dodson from comment #1) > https://trello.com/c/OA9z4cMU/701-8-node-deliver-node-problem-detector-for- > tech-preview says NPD is tech-preview therefore not a 3.10 blocker here. any plans for getting this updated?
openshift_node_problem_detector_install = true should be the only variable that's required. Internal to the role openshift_node_problem_detector_install is used to force removal of the npd and should not need to be set to install.