csa/devops/ansible-patches/openshift.git/roles/openshift_logging_elasticsearch/tasks, branch ands

Whenever we create a new es node ignore health checks, changing prometheus pw gen for increased secret idempotency

2018-02-13T20:26:33+00:00

Fix indentation to make yamllint happy

2018-02-10T11:41:14+00:00

Signed-off-by: Vadim Rutkovsky

Verify that requested services have schedulable nodes matching the selectors

2018-02-10T09:41:45+00:00

Signed-off-by: Vadim Rutkovsky

changed oc to {{ openshift_client_binary }}

2018-02-08T11:27:35+00:00

Adding support for ES 5.x tech preview opt in

2018-02-05T19:44:16+00:00

bug 1537857. Fix retrieving prometheus metrics

2018-01-26T21:46:13+00:00

Only automatically restart if cluster is in yellow or green state

2018-01-24T22:27:31+00:00

Ensure we are running oc execs against running pods

2018-01-16T16:45:39+00:00

Merge pull request #6602 from ewolinetz/es_full_cluster_restart

2018-01-10T09:36:20+00:00

Automatic merge from submit-queue. Adding logic to do a full cluster restart if we are incrementing our … …major versions of ES This will help with the upgrade from 2.x to 5.x for ES, it also fixes something I came across with the handler on 3.7 where it checks the prior deployed version of the ES pod rather than the new one.

Fix: change import_role to include_role

2018-01-08T21:29:27+00:00

It appears that when one role dynamically imports another, usage of import_role inside the dynamically included role is not possible. If something is included with include_role (dynamic), all tasks therein must also use include_role (dynamic).