Bug 1516322

Summary: HA VM is not restarted if was was Paused on SPM Host.
Product: [oVirt] ovirt-engine Reporter: Polina <pagranat>
Component: BLL.VirtAssignee: Francesco Romani <fromani>
Status: CLOSED CURRENTRELEASE QA Contact: Polina <pagranat>
Severity: high Docs Contact:
Priority: unspecified    
Version: 4.2.0CC: bugs, michal.skrivanek, tjelinek
Target Milestone: ovirt-4.2.0Flags: rule-engine: ovirt-4.2+
rule-engine: planning_ack+
tjelinek: devel_ack+
mavital: testing_ack+
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-01-12 12:56:35 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: 1516712    
Bug Blocks:    
Attachments:
Description Flags
engine log. the scenario starts at line 1348 none

Description Polina 2017-11-22 13:19:19 UTC
Created attachment 1357507 [details]
engine log. the scenario starts at line 1348

Description of problem:
The HA VM with lease that is started on SPM host could not be restarted after Pause.

Version-Release number of selected component (if applicable):
ovirt-engine-setup-plugin-ovirt-engine-4.2.0-0.0.master.20171119135709.git6d448d3.

How reproducible: 80%

Steps to Reproduce:
1. There are two hosts Normal host1 and SPM host2. Create VM with scsi storage. open HA tab , check HA with nfs lease (Resume Behavior = KILL). Run VM on SPM host2. block the storage on the SPM host2 with  command <iptables -I INPUT -s 10.35.146.129 -j DROP>. As result The VM is "paused due to storage I/O problem".
2. Wait until SPM moves to the host1. AS result we have two healthy hosts - SPM host1 and Normal host2.
3. Unblock the storage with command <iptables -D INPUT 1>.

Actual results:
The VM is not killed and restarted . it is moved to down state forever

Expected results:
the VM must be killed and restarted

Additional info:
engine log attached . please see the scenario starting from 1348 line: VM vm_glance_templ_scsi has been paused due to storage I/O problem.

Comment 1 Tomas Jelinek 2017-11-22 13:39:28 UTC
the reason is that the exit code of the VM killed this way is not "error" so the engine is not trying to restart it.

Comment 2 Francesco Romani 2017-11-28 12:46:47 UTC
this is just a bug. Doesn't deserve a doc_text

Comment 3 Polina 2018-01-02 09:23:37 UTC
put depends on  https://bugzilla.redhat.com/show_bug.cgi?id=1516712 because I want to verify it with kinds of VMs - HA &HA with lease

Comment 4 Polina 2018-01-09 11:53:11 UTC
verified for ovirt-engine-4.2.1-0.2.el7.noarch

Comment 5 Polina 2018-01-09 12:17:58 UTC
verified on ovirt-engine-4.2.1-0.2.el7.noarch