Bug 773689

Summary: Core: maintenance host will appear in host selection when trying to remove domain which renders removed luns unusable and can cause other storage related issues
Product: [Retired] oVirt Reporter: Dafna Ron <dron>
Component: ovirt-engine-coreAssignee: lpeer <lpeer>
Status: CLOSED WONTFIX QA Contact:
Severity: high Docs Contact:
Priority: high    
Version: unspecifiedCC: abaron, acathrow, amureini, iheim, ykaul
Target Milestone: ---   
Target Release: 3.3.4   
Hardware: x86_64   
OS: Linux   
Whiteboard: storage
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-12-12 07:33:32 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: Storage RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
logs none

Description Dafna Ron 2012-01-12 15:44:21 UTC
Created attachment 552435 [details]
logs

Description of problem:

removing storage domain will allow user to select a host which is in maintenance and will cause several random storage issues. 
in example log attached, I added a domain and then removed it using a host which is on maintenance state and although operation succeeded (sometimes it does not) the luns cannot be used again to create a new domain. 
Looking at the host, we connect to the storage once we detect that the storage is not connected and then disconnect but the connections still exist in active host.

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

ovirt-engine-backend-3.0.0_0001-7.fc16.x86_64
ovirt-engine-webadmin-portal-3.0.0_0001-7.fc16.x86_64
vdsm-4.9.2-0.65.gitf945dc2.fc16.x86_64
libvirt-0.9.6-4.fc16.x86_64

How reproducible:

100%

Steps to Reproduce:
1. in two hosts cluster put one host in maintenance and add iscsi domain
2. put one host in maintenance and try to remove the domain using the maintenance host
3.
  
Actual results:

weather you succeed in removing the domain or not, the luns cannot be used again. 

Expected results:

backend results to gui query should filter maintenance host

Additional info: backend, and both vdsm logs are attached


active host after remove: 

[root@blond-vdsg tmp]# iscsiadm -m session
tcp: [174] 10.35.64.10:3260,1 Dafna-Main
tcp: [187] 10.35.64.10:3260,1 Dafna-01
tcp: [188] 10.35.64.10:3260,1 Dafna-02

maintenance host will connect to to remove the storage and then disconnect: 

root@blond-vdsf ~]# iscsiadm -m session
tcp: [13] 10.35.64.10:3260,1 Dafna-01
tcp: [14] 10.35.64.10:3260,1 Dafna-02
[root@blond-vdsf ~]# iscsiadm -m session
iscsiadm: No active sessions.

Comment 1 Itamar Heim 2012-12-12 07:33:32 UTC
Closing old bugs. If this issue is still relevant/important in current version, please re-open the bug.