Bug 1113165

Summary:	[RFE] can't set HA-Flag for VMs in a local storage DC
Product:	[Retired] oVirt	Reporter:	Sven Kieske <s.kieske>
Component:	ovirt-engine-core	Assignee:	bugs <bugs>
Status:	CLOSED NOTABUG	QA Contact:	Pavel Stehlik <pstehlik>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	3.3	CC:	acathrow, amureini, dfediuck, gklein, iheim, yeylon
Target Milestone:	---
Target Release:	---
Hardware:	x86_64
OS:	Linux
Whiteboard:	sla
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2014-06-26 07:01:19 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 Sven Kieske 2014-06-25 14:28:28 UTC

Description of problem:
you can't set the HA-Flag for VMs in ovirt 3.3.3
in order to make them boot automatically when a fatal host shutdown
occured (host refers to the host the vm runs on).
At least, this is not possible via gui.

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


How reproducible:
always

Steps to Reproduce:
1. Try to configure HA for any vm on a local storage system, find the option greyed out.
2. There's a tooltip, which says: "Host cannot be set highly available when 'Do not allow migration' or 'Allow manual migration' is selected"
3. You can not select any migration option, as there is just one host in the cluster

Actual results:
you can't configure vms to be automatic started in case of a power failure
or other disaster events, like described here:
https://access.redhat.com/site/documentation/en-US/Red_Hat_Enterprise_Virtualization/3.3/html/Administration_Guide/sect-Improving_Uptime_with_Virtual_Machine_High_Availability.html

specifically, some of the mentioned conditions described in [1] are not necessary:

https://access.redhat.com/site/documentation/en-US/Red_Hat_Enterprise_Virtualization/3.3/html/Administration_Guide/High_availability_considerations.html

Expected results:
be able to reduce downtime in a local storage dc by configuring HA for vms, so
they get rebooted if an outage occurs.

I know this is not "real" HA like in a shared storage scenario, but this would
improve uptime a lot in this use case.

Additional info:

Comment 1 Doron Fediuck 2014-06-26 07:01:19 UTC

Hi Sven,
local storage is defined as the server's internal storage.
In the event of a hardware failure, the system will have no way
to migrate / resume the VM as the local storage becomes unavailable and
possibly even completely lost (SDD failure).

So for the above reason there's no point of tainting the engine to do
something it cannot do. HA should remain HA as defined. Reducing down
time in such a case can be achieved in other ways such as a script that
is run by cron and monitors your local storage VMs, and once a failure
is detected the script may try to run the VMs. You can also use monitoring
tools such as nagios.
If you need to do this inside the engine, you can write your own load-balancing
module which will be used by the ovirt scheduler to monitor and take actions
in specific clusters.