Bug 1384301

Summary:	[RFE] Director support for compute node replacement with same hostname
Product:	Red Hat OpenStack	Reporter:	Bryan Yount <byount>
Component:	rhosp-director	Assignee:	Angus Thomas <athomas>
Status:	CLOSED WORKSFORME	QA Contact:	Omri Hochman <ohochman>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	8.0 (Liberty)	CC:	aguetta, aschultz, asoni, athomas, bnemec, dabrown, dbecker, jcoufal, mburns, morazi, rhel-osp-director-maint
Target Milestone:	---	Keywords:	FutureFeature, Triaged
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:	2017-09-06 15:59:12 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:

Description Bryan Yount 2016-10-13 04:10:22 UTC

Description of problem:
Similar to Bug 1351753 which requests the ability for director to redeploy an existing controller node, my customer would like the ability to redeploy an existing compute node.

Version-Release number of selected component (if applicable):
OSP 8 director

How reproducible:
Very

Steps to Reproduce:
1. Deploy HA overcloud with predictable hostnames and ips
2. Attempt to replace one of the controllers with another controller using the same hostname and ips.

Additional info:
The onsite consultant has tried various scenarios of nova delete, ironic delete, service delete to get the node to rebuild with the same name for a compute node. The only way he got it to rebuild a node was to completely remove it, and deploy with the next number in sequence, and leave a skip for the old node name. This is not the behavior that the customer would like.

Comment 2 Bryan Yount 2016-10-20 17:30:45 UTC

The customer has asked if there are any suggestions from engineering on how to achieve this with a workaround while we wait for implementation of this RFE?

Comment 3 Angela Soni 2017-02-10 18:21:59 UTC

Following up to see if there are any updates on this for the workaround while this RFE is implemented.

Comment 5 Angela Soni 2017-04-28 23:29:39 UTC

From Customer recent comments, the automation Red Hat has developed (based on Director) doesn’t have provision for building a compute node with specific hostname – it only counts up starting at 0.  So if we have 10 compute nodes and compute node 7 fails we can build node 11 but not node 7 if that makes sense.  Is there another procedure we could employ to rebuild or otherwise replace a specific compute host?