Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1674066

Summary:	_predeploy_verify_capabilities compares the current node count from ironic with count from the heat stack before it's updated with the new values in the templates
Product:	Red Hat OpenStack	Reporter:	David Vallee Delisle <dvd>
Component:	python-tripleoclient	Assignee:	Emilien Macchi <emacchi>
Status:	CLOSED WORKSFORME	QA Contact:	Gurenko Alex <agurenko>
Severity:	high	Docs Contact:
Priority:	high
Version:	10.0 (Newton)	CC:	agurenko, dvd, emacchi, hbrock, jslagle, mburns, ramishra, rhos-maint
Target Milestone:	---	Keywords:	Triaged, ZStream
Target Release:	10.0 (Newton)
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:	1674054	Environment:
Last Closed:	2019-06-19 19:14:34 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	1674054
Bug Blocks:

Description David Vallee Delisle 2019-02-08 22:40:35 UTC

Description of problem:
We ran into this interesting issue. We have 2 compute roles: Compute and ComputeV2.

On this environment, we have 128 Compute and 0 ComputeV2. 

Someone made a mistake and added 96 ComputeV2, in addition to the 128 Compute. 

This didn't triggered the validation in _predeploy_verify_capabilities, instead, heat went ahead and created all stacks and resources and failed later with "not enough host".

When we tried to redeploy with the adjusted templates, we kept seeing "Not enough nodes - available: 128, requested: 224". We had to --validation-errors-nonfatal to bypass this.

I'm not sure if we should update the heat stack before we validate, or validate against the templates.

Version-Release number of selected component (if applicable):
python-tripleoclient-5.4.6-1.el7ost.noarch                  Mon Feb  4 18:51:14 2019


How reproducible:
All the time

Steps to Reproduce:
1. Have too many ComputeCount in templates
2. Deploy & wait for failure
3. Fix templates
4. Redeploy