Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 881004

Summary: WSRP clustering - configuration of producer isn't properly replicated to second node after failover
Product: [JBoss] JBoss Enterprise Portal Platform 6 Reporter: Michal Vanco <mvanco>
Component: PortalAssignee: claprun <claprun>
Status: CLOSED CURRENTRELEASE QA Contact: Michal Vanco <mvanco>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 6.0.0CC: bdawidow, epp-bugs, mvecera, theute
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-04-16 08:55:55 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:
Attachments:
Description Flags
NPE after direct access to WSRP portlet after failover none

Description Michal Vanco 2012-11-28 11:47:10 UTC 
Description of problem:
WSRP producer configuration which is done on loadbalancer (active node) isn't present at loabalancer (second node) when first node was stopped.

Version-Release number of selected component (if applicable):
JPP 6.0.0 ER03

How reproducible:
Always

Steps to Reproduce:
1. setup producer as 2 node cluster + loadbalancer & start
2. access loadbalancer (verify active node)
3. go to WSRP admin, producer configuration
4. do some changes (for example uncheck first checkbox + add registration property), Save
5. stop active node, access loadbalancer again (direct access to WSRP admin or access to other page + WSRP admin or access some page, logout, login, access WSRP page) -> saved configuration isn't present (wasn't replicated from first node)
  
Actual results:
Producer configuration isn't replicated

Expected results:
Producer configuration is replicated, what was saved before is now present
Comment 1 claprun@redhat.com 2012-11-28 15:22:15 UTC 
Can't replicate this one either with latest WSRP… :(
Comment 2 Boleslaw Dawidowicz 2012-11-28 15:49:57 UTC 
Closing for now. Assuming it'll be fixed in next JPP release at the end of next week
Comment 3 Michal Vanco 2012-11-29 11:04:10 UTC 
I've just reproduced the issue with gatein master :(
I would recommend to use two separate machines for replication verification instead of single machine with port offsets.
I've also hit another failure when I accessed wsrp admin portlet directly after failover (log attached). Ctrl+f5 or logout solved the issue and portlet was displayed - but not with saved configuration.
Comment 4 Michal Vanco 2012-11-29 11:04:57 UTC 
Created attachment 654188 [details]
NPE after direct access to WSRP portlet after failover
Comment 5 Thomas Heute 2013-01-09 10:47:21 UTC 
So the failover node is not accessed at all during steps 1-4 ?

I wonder how the failover node can access old data, unless the active node didn't flush changes before it got killed ?
Comment 6 claprun@redhat.com 2013-01-11 09:40:58 UTC 
I believe this issue is fixed in GateIn WSRP master now.
Comment 7 Michal Vanco 2013-01-24 12:03:21 UTC 
This issue was verified at JPP 6.0.0 ER05 and is no longer present. Configuration is replicated properly now!