Bug 1034260

Summary: [Tracker] HA Apache webfarm deployment scenario
Product: Red Hat Enterprise Linux 7 Reporter: Fabio Massimo Di Nitto <fdinitto>
Component: pacemakerAssignee: Andrew Beekhof <abeekhof>
Status: CLOSED CURRENTRELEASE QA Contact: cluster-qe <cluster-qe>
Severity: high Docs Contact:
Priority: high    
Version: 7.0CC: cluster-maint, fdinitto, jharriga, kgaillot, mnovacek
Target Milestone: rcKeywords: TestOnly, Tracking
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
: 1080150 (view as bug list) Environment:
Last Closed: 2017-05-24 16:04:45 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:
Bug Depends On:    
Bug Blocks: 1111381, 1080150    

Description Fabio Massimo Di Nitto 2013-11-25 14:02:45 UTC 
Description of problem:

This bug is intended to track all bugs encountered in HA Apache deployment scenario testing for RHEL 7.

Testplan: TBA

Draft of HA Apache webfarm deployment scenario: 
http://rhel-ha.etherpad.corp.redhat.com/PRD-HA-webfarm
Comment 1 Andrew Beekhof 2015-01-23 00:42:05 UTC 
Did this happen?  Can it be closed or do we need to bump to 7.2?
Comment 2 michal novacek 2015-01-28 16:11:34 UTC 
I have tested that it is possible to create and run apache in active/active
mode as described in comment 1. Pages served by apache are being kept on shared
gfs2 filesystem. I have used haproxy to loadbalance queries to both apache
instances equally.

Manually I have done the following functional testing:

    * Create and start apache, clusterfs and haproxy according to comment 1

    * Downloaded several hunderds of pages from ip address where haproxy
        listens and verified that both nodes served about the same amount of
        queries.

    * While the download was running I hard rebooted one of the apache nodes
        and checked that there were no disturbance in pages downloading and 
        once the rebooted node joined back the cluster it started replying
        to the queries again.


I do not have the tests automated yet and did not run it over longer period of
time.
Comment 4 Ken Gaillot 2016-06-23 22:19:21 UTC 
Is the testing in Comment 2 considered sufficient to close this bz?