Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1430012

Summary: Specify nodeSelector for PV-recycler pod worked in 3.1 but broke since then
Product: OpenShift Container Platform Reporter: Eric Jones <erjones>
Component: StorageAssignee: Matthew Wong <mawong>
Status: CLOSED NOTABUG QA Contact: Liang Xia <lxia>
Severity: urgent Docs Contact:
Priority: low    
Version: 3.4.1CC: aos-bugs, bchilds, eparis, erjones, hchen, knakayam, mawong
Target Milestone: ---Keywords: NeedsTestCase, Reopened
Target Release: ---   
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: 2017-03-31 15:18:26 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 Eric Jones 2017-03-07 16:18:30 UTC
- What is the nature and description of the request? 
As an admin I require the ability to limit the pv-recycler pod to only run on certain nodes.

- Why does the customer need this? (List the business requirements here) 
The pod sometimes experiences problems on certain nodes but others work fine, so we need to be able to limit the pv-recycler to run on the good nodes

- How would the customer like to achieve this? (List the functional requirements here) 
Include a template that can be referenced so a nodeSelector can be included. Or document how to create a template properly (as there is already a way to specify a pv-recycler-template: kubernetesMasterConfig.controllerArguments.pv-recycler-pod-template-filepath-nfs)

- For each functional requirement listed, specify how Red Hat and the customer can test to confirm the requirement is successfully implemented.
Deploy a cluster with multiple nodes and label them all differently. Then test the pv-recycler a few times to make sure it goes to any of the nodes (based on the normal scheduler) and then add the tempalte with the nodeSelector and watch it only go to the one (or two) nodes, based on the selector)

- Is there already an existing RFE upstream or in Red Hat Bugzilla?
[0] is the closest I could find, though I know this used to work in OpenShift 3.1 [1], but I am not certain if it is the platform's fault or the openshift recycler image's fault.

[0] https://github.com/openshift/origin/pull/6884
[1] https://access.redhat.com/solutions/2301941

Comment 17 Bradley Childs 2017-03-24 20:55:52 UTC
@eric jones - matthew tried to recreate it per your setup directions and it looks like its an issue with the service account setup you're using.

Comment 19 Bradley Childs 2017-03-31 15:18:26 UTC
Closing this per Matthews update.