Running the openshift-logging.yml playbook throws an OOM in the controller node.
Issue
With a controller node with 2G of mem, 30 managed host, the openshift-logging.yml playbook randomly fails (always during the fact gathering) without clear message. Instead, The controller node shows the following messages:
kernel: ansible-playboo invoked oom-killer: gfp_mask=0x280da, order=0, oom_score_adj=0
kernel: [<ffffffff811863a4>] oom_kill_process+0x254/0x3d0
Environment
OCP 3.7
ansible.noarch 2.4.1.0-1.el7
Controller node with 2G of RAM
Inventory with 30 managed nodes
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
