Bug 1730179

Summary: The default values for OpenShift metrics(hawkular) on installation should be documented
Product: OpenShift Container Platform Reporter: Masatoshi Hayashi <mhayashi>
Component: DocumentationAssignee: Michael Burke <mburke>
Status: CLOSED CURRENTRELEASE QA Contact: Junqi Zhao <juzhao>
Severity: low Docs Contact: Vikram Goyal <vigoyal>
Priority: low    
Version: 3.11.0CC: aos-bugs, jokerman, juzhao, mmccomas, wsun
Target Milestone: ---   
Target Release: 3.11.z   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2020-04-08 20:50:08 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Masatoshi Hayashi 2019-07-16 05:25:32 UTC 
Document URL: 
https://docs.openshift.com/container-platform/3.11/install_config/cluster_metrics.html#metrics-ansible-variables

Section Number and Name:
Specifying Metrics Ansible Variables

Describe the issue: 
The document doesn't have information about the default value when users don't specify `openshift_metrics_cassandra_nodeselector` / `openshift_metrics_hawkular_nodeselector ` / `openshift_metrics_heapster_nodeselector `.

If users don't have any specific value in their inventory, the nodeselector becomes blank value("") like this:

```
$ oc get project openshift-infra -o yaml
apiVersion: project.openshift.io/v1
kind: Project
metadata:
  annotations:
    openshift.io/node-selector: ""
...
```

It means that the pods in openshift-infra ignore defaultNodeSelector and may be deployed to any kind of node.

I think the behavior is confusing and should be documented.

Suggestions for improvement: 
The documentation about the default behavior can be added to improvement.

Additional information:
Comment 1 Michael Burke 2020-04-02 21:03:52 UTC 
Junqi -- Can you help with this? 

What is the installation behavior if the user does not specify `openshift_metrics_cassandra_nodeselector`, `openshift_metrics_hawkular_nodeselector `, or `openshift_metrics_heapster_nodeselector `?

If the user specifies one of these parameters, does he need to set them all?

Thanks for any help.

Michael
Comment 2 Junqi Zhao 2020-04-03 01:42:41 UTC 
(In reply to Michael Burke from comment #1)
> Junqi -- Can you help with this? 
> 
> What is the installation behavior if the user does not specify
> `openshift_metrics_cassandra_nodeselector`,
> `openshift_metrics_hawkular_nodeselector `, or
> `openshift_metrics_heapster_nodeselector `?
> 
> If the user specifies one of these parameters, does he need to set them all?
> 
> Thanks for any help.
> 
> Michael

If they are set, the components will be deployed to the nodes which match the nodeSelector.
If not set, they can deployed on any schedulable nodes
Comment 3 Michael Burke 2020-04-03 17:08:35 UTC 
Docs PR https://github.com/openshift/openshift-docs/pull/20900
Comment 4 Junqi Zhao 2020-04-07 01:52:23 UTC 
lgtm
Comment 5 Michael Burke 2020-04-08 20:50:08 UTC 
Changes are live
https://docs.openshift.com/container-platform/3.11/install_config/cluster_metrics.html#metrics-ansible-variables
https://access.redhat.com/documentation/en-us/openshift_container_platform/3.11/html-single/configuring_clusters/index#metrics-ansible-variables