Bug 1313009

Summary: [scale] Failed Disk Migration may lead to disks/vm in unregistered state
Product: [oVirt] ovirt-engine Reporter: mlehrer
Component: BLL.StorageAssignee: Daniel Erez <derez>
Status: CLOSED WORKSFORME QA Contact: eberman
Severity: high Docs Contact:
Priority: unspecified    
Version: 3.6.3.2CC: amureini, bugs, tnisan
Target Milestone: ovirt-4.2.0Flags: tnisan: ovirt-4.2?
rule-engine: planning_ack?
rule-engine: devel_ack?
rule-engine: testing_ack?
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2017-03-01 08:32:43 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: Storage RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description mlehrer 2016-02-29 17:24:33 UTC 
Description of problem:

Attempted disk migration of 6 disks, during this migration 4 completed and 2 failed.  An error was reported that disk deletion was not successful.
After the failed disk migration the disks and vm were no longer listed in ovirt web UI but are listed via rest api:

   -/ovirt-engine/api/storagedomains/9e60a7d8-f94f-4dfe-a012-3b8ccc216178/vms;unregistered   ==> shows 6 disks listed
   -/ovirt-engine/api/storagedomains/9e60a7d8-f94f-4dfe-a012-3b8ccc216178/disks;unregistered ==> shows 1 VM


Env details
2 hosts
50 iSCSI domains of which all are 20g and 2 are 1TB each


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

vdsm-python-4.17.21-0.el7ev.noarch
vdsm-jsonrpc-4.17.21-0.el7ev.noarch
vdsm-4.17.21-0.el7ev.noarch
vdsm-yajsonrpc-4.17.21-0.el7ev.noarch
vdsm-hook-vmfex-dev-4.17.21-0.el7ev.noarch
vdsm-xmlrpc-4.17.21-0.el7ev.noarch
vdsm-cli-4.17.21-0.el7ev.noarch
vdsm-infra-4.17.21-0.el7ev.noarch



How reproducible:
unclear/not sure

Steps to Reproduce:
1. Migrate 6 disks between iSCSI domains
2. VDSM reports host unreachable during migration 
3. Host gets marked unresponsive and then recovers, resulting in migration partially failing in that 2 disks of 6 attempted are unsuccessful.

Actual results:

Migration fails, disks belonging to the domain, and connected VM are left in unregistered state.

Expected results:

When migration fails, disks still shown, or message lists disks in unregistered state.

Additional info:

sos log info collected and accessible in private comment

Per dev request we opened this bug, but exact reproduction is unclear, as its _assumed_ that failed deletion during cold disk migration led to disks being left in unregistered state.
Comment 2 Red Hat Bugzilla Rules Engine 2016-03-02 11:21:31 UTC 
Bug tickets must have version flags set prior to targeting them to a release. Please ask maintainer to set the correct version flags and only then set the target milestone.
Comment 3 Daniel Erez 2017-03-01 08:32:43 UTC 
The live migration flow has been re-written since 3.6. The issue doesn't seem relevant any more. Closing, please re-open if reproduced.