Bug 1340345

Summary:	Cluster level can be changed while there are running VMs
Product:	[oVirt] ovirt-engine	Reporter:	Michal Skrivanek <michal.skrivanek>
Component:	BLL.Virt	Assignee:	Marek Libra <mlibra>
Status:	CLOSED CURRENTRELEASE	QA Contact:	sefi litmanovich <slitmano>
Severity:	high	Docs Contact:
Priority:	high
Version:	4.0.0	CC:	adevolder, agkesos, bugs, emahoney, germano, gklein, gveitmic, inetkach, lsurette, mavital, mgoldboi, michal.skrivanek, mlibra, pstehlik, rbalakri, Rhev-m-bugs, slitmano, srevivo, stirabos, tjelinek, ycui, ykaul
Target Milestone:	ovirt-3.6.7	Keywords:	Reopened
Target Release:	3.6.7.1	Flags:	rule-engine: ovirt-3.6.z+ mgoldboi: planning_ack+ michal.skrivanek: devel_ack+ rule-engine: testing_ack+
Hardware:	x86_64
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:	Cause: Cluster level was able to be changed with VMs running. Consequence: The running VMs cause non-deterministic issues while underlying devices/services has changed. Fix: Enforce no VM is running when changing the cluster level. Result: All VMs in a cluster must be down before changing cluster level.	Story Points:	---
Clone Of:	1336527	Environment:
Last Closed:	2016-07-04 12:33:20 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	Virt	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:
Bug Blocks:	1336527, 1341023

Description Michal Skrivanek 2016-05-27 06:27:10 UTC

+++ This bug was initially created as a clone of Bug #1336527 +++

Description of problem: After migration from 3.4 to 3.6, the .vv spec file for VM is not updated to the new specs


Version-Release number of selected component (if applicable): 3.4 DC to 3.6.5


How reproducible: Have not created a reproducer, but will repeat on customers hardware/environment.


Steps to Reproduce:
1. Create 3.4 DC with rhevm managment/display network. 
2. Create VM on the 3.4 envi
3. Create a new 3.6.5 cluster with ovirtmgmt management/display network
4. Add rhevm network to the 3.6.5 for the migration
5. Migrate the VM from 3.4 to 3.6.5
6. After the migration, the VM still has the specs from the original .vv that say that the management/display comes from rhevm instead of from ovirtmgmt

Actual results: .vv file for test VM points to rhevm network


Expected results: .vv file for test VM points to ovirtmgmt


Additional info:

--- Additional comment from Michal Skrivanek on 2016-05-17 16:55:02 CEST ---

after further investigation we found out that editing cluster level is not blocked when VMs are running. Several things are checked by bug 1298487 but this basic case is not applied. We need to check any non-Down state and block on it

Comment 1 Michal Skrivanek 2016-06-22 09:45:57 UTC

see bug 1348907 for possible removal of this restriction

Comment 2 sefi litmanovich 2016-06-22 17:02:04 UTC

Verified that restrictions are implemented on rhevm-3.6.7.4-0.1.el6.noarch.

1. Created a 3.4 DC.
2. Created a 3.4 cluster.
3. Added rhel 6.6 host with latest 3.4 vdsm.
4. Created a vm and started it on the cluster.
5. Tried to change comparability version of the cluster to 3.5 - cannot change, got the message suggesting to shutdown vm.

Did the same with 3.5 to 3.6.