Bug 1121185 - When using rabbitmq AMQP traffic should be load balanced accross all nodes
Summary: When using rabbitmq AMQP traffic should be load balanced accross all nodes
Alias: None
Product: Red Hat OpenStack
Classification: Red Hat
Component: openstack-foreman-installer
Version: 5.0 (RHEL 7)
Hardware: Unspecified
OS: Unspecified
Target Milestone: ga
: Installer
Assignee: John Eckersberg
QA Contact: Leonid Natapov
Depends On:
TreeView+ depends on / blocked
Reported: 2014-07-18 15:07 UTC by Steve Reichard
Modified: 2016-04-26 22:41 UTC (History)
7 users (show)

Fixed In Version: openstack-foreman-installer-2.0.17-1.el6ost
Doc Type: Bug Fix
Doc Text:
Clone Of:
Last Closed: 2014-08-21 18:05:45 UTC

Attachments (Terms of Use)

System ID Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2014:1090 normal SHIPPED_LIVE Red Hat Enterprise Linux OpenStack Platform Enhancement Advisory 2014-08-22 15:28:08 UTC

Description Steve Reichard 2014-07-18 15:07:16 UTC
Description of problem:

Using O-F-I, Iwas able to install a Nove Net HA config taht used rabbitmq for AMQP.

Looking at the haproxy.conf, I noticed:

listen amqp
  mode  tcp
  option  tcplog
  stick  on dst
  stick-table  type ip size 2
  timeout  client 120s
  timeout  server 120s
  server ospha1  check
  server ospha2  check
  server ospha3  check

Which directs traffic to only one node.  I beleive for rabbit all nodes should be used.

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

[root@ospha-inst config]# yum list installed | grep -e openstack -e foreman -e puppet
This system is not registered to Red Hat Subscription Management. You can use subscription-manager to register.
foreman.noarch              @RHEL-6-Server-OS-Foreman
foreman-installer.noarch            1:1.5.0-0.4.RC2.el6ost
foreman-mysql2.noarch       @RHEL-6-Server-OS-Foreman
foreman-proxy.noarch         @RHEL-6-Server-OS-Foreman
foreman-selinux.noarch              1.6.0-2.el6sat     @RHEL-6-Server-OS-Foreman
openstack-foreman-installer.noarch  2.0.15-1.el6ost    @RHEL-6-Server-OS-Foreman
openstack-puppet-modules.noarch     2014.1-19.1.el6ost @RHEL-6-Server-OS-Foreman
puppet.noarch                       3.6.2-1.1.el6      @RHEL-6-Server-OS-Foreman
puppet-server.noarch                3.6.2-1.1.el6      @RHEL-6-Server-OS-Foreman
rubygem-foreman_api.noarch          0.1.11-4.el6sat    @RHEL-6-Server-OS-Foreman
rubygem-hammer_cli_foreman.noarch   0.1.0-6.el6sat     @RHEL-6-Server-OS-Foreman
[root@ospha-inst config]# 

How reproducible:

Steps to Reproduce:

Actual results:

Expected results:

Additional info:

Comment 2 John Eckersberg 2014-07-18 16:17:59 UTC
Steve, sanity check me since I am an haproxy newbie.  I should just need to remove these two lines to achieve the expected behavior, correct?

  stick  on dst
  stick-table  type ip size 2

Comment 3 Steve Reichard 2014-07-18 17:29:12 UTC
That was my understanding but more importantly this agrees with what Ryan O'Hara e-mailed me:

Just take out the 'stick on dst' and the 'stick-table type ip size
2' and you're good.


Comment 4 John Eckersberg 2014-07-18 19:23:03 UTC

Comment 8 Leonid Natapov 2014-08-18 11:22:17 UTC

listen amqp
  mode  tcp
  option  tcplog
  timeout  client 120s
  timeout  server 120s
  server mac047d7b627d5a.example.com  check inter 1s
  server mac047d7b685a04.example.com  check inter 1s
  server mac047d7b61754e.example.com  check inter 1s

stick  on dst
stick-table  type ip size 2 were removed.

Comment 9 errata-xmlrpc 2014-08-21 18:05:45 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.


Note You need to log in before you can comment on or make changes to this bug.