DX OI: Elasticsearch and its related Pods are never scheduled after install

Products

• DX Operational Intelligence

Issue/Introduction

Elasticserach and related Pods failed to schedule by following message:
0/n nodes are available : 1 NodeNotReady, n MatchNodeSelector.

Resolution

(1) Check the node labels

# oc get node --show-labels

NAME                          STATUS    ROLES     AGE       VERSION             LABELS

openshift1.local.net   Ready     master    30d       v1.9.1+a0ce1bc657  

beta.kubernetes.io/arch=amd64,

beta.kubernetes.io/os=linux,doi-es-node=master-data-1,

kubernetes.io/hostname=openshift1.local.net,

node-role.kubernetes.io/master=true,

region=infra,

zone=default

openshift2.local.net   Ready     compute   30d       v1.9.1+a0ce1bc657  

beta.kubernetes.io/arch=amd64,

beta.kubernetes.io/os=linux,

kubernetes.io/hostname=openshift2.local.net,

node-role.kubernetes.io/compute=true,

region=primary,

zone=default

openshift3.local.net   Ready     compute   30d       v1.9.1+a0ce1bc657  

beta.kubernetes.io/arch=amd64,

beta.kubernetes.io/os=linux,

kubernetes.io/hostname=openshift3.local.net,

node-role.kubernetes.io/compute=true,

region=primary,

zone=default

(2) Add compute label to master node

# oc label nodes [nodename] node-role.kubernetes.io/compute=true

node "openshift1.local.net" labeled

# oc get nodes

NAME                          STATUS    ROLES            AGE       VERSION

openshift1.local.net   Ready     compute,master   34d       v1.9.1+a0ce1bc657

openshift2.local.net   Ready     compute          34d       v1.9.1+a0ce1bc657

openshift3.local.net   Ready     compute          34d       v1.9.1+a0ce1bc657

Now the waiting pods are all running.