Bug 1162654 - HA | L3 and DHCP neutron services are running simultaneously on several cluster nodes.
Summary: HA | L3 and DHCP neutron services are running simultaneously on several clust...
Keywords:
Status: CLOSED DUPLICATE of bug 1175869
Alias: None
Product: Red Hat OpenStack
Classification: Red Hat
Component: openstack-foreman-installer
Version: 5.0 (RHEL 7)
Hardware: Unspecified
OS: Unspecified
urgent
urgent
Target Milestone: ga
: Installer
Assignee: Jason Guiditta
QA Contact: Ofer Blaut
URL:
Whiteboard:
Depends On:
Blocks: 1177026
TreeView+ depends on / blocked
 
Reported: 2014-11-11 12:44 UTC by Leonid Natapov
Modified: 2015-01-05 19:18 UTC (History)
7 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2015-01-05 19:18:23 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)
logs (5.45 MB, application/octet-stream)
2014-11-11 12:44 UTC, Leonid Natapov
no flags Details

Description Leonid Natapov 2014-11-11 12:44:12 UTC
Created attachment 956270 [details]
logs

Description of problem:
 
L3 and DHCP neutron services are running simultaneously on several cluster nodes.
I have an HA + Neutron VXLAN deployment using latest puddle (A2). At some point I've checked the neutron services and saw that pcs status shows that all neutron services are running on one node but when I check the nodes ,runnig "ps -aux" on each  neutron service, I see that neutron-dhcp-agent  and neutron-l3-agent are running on at least one two nodes which is wrong.

openstack-neutron-openvswitch-2014.1.3-7.el7ost.noarch
python-neutron-2014.1.3-7.el7ost.noarch
openstack-neutron-2014.1.3-7.el7ost.noarch
openstack-neutron-ml2-2014.1.3-7.el7ost.noarch
python-neutronclient-2.3.4-3.el7ost.noarch

rhel-osp-installer-0.4.7-1.el6ost.noarch
openstack-puppet-modules-2014.1-24.el6ost.noarch
ruby193-rubygem-foreman_openstack_simplify-0.0.6-8.el6ost.noarch
openstack-foreman-installer-2.0.32-1.el6ost.noarch

system logs,neutron logs and crm_report are attached.

Comment 1 Jason Guiditta 2014-11-17 16:57:00 UTC
Leonid, a few questions:

1. When you observed the A/P services running on more than one node, had you:
  - rebooted any of the nodes, and if so, was the extra running services on one that was rebooted, or one that was NOT rebooted?
  - run puppet a second time, or only once?

2. On the nodes that had a service running where you did not expect it, was the output of service <name> status correct, or incorrect?

3. Did you try to manually stop the services on the nodes where that service should not be running?  If so, did it stay down or attempt to start itself again?

4. does the service show as disabled on boot, or enabled?

Comment 2 Leonid Natapov 2014-11-18 09:11:27 UTC
1.I believe that at some point I did reboot to the nodes. I can't tell  exactly it it was the node with the extra running services. I didn't run puppet manually. The services status was correct. I didn't try to stop services manually. Services are disabled on boot.

Comment 3 Asaf Hirshberg 2015-01-05 13:12:57 UTC
repredoced at my setup today(5.1). tried to create a new network and some instances under them. the instances didn't got any ip address and when i checked 
the name-spaces i saw that the network ar not exist. after that i checked the other server and saw that the neutron services are running also there and i saw my network there.



version:
rhel-osp-installer-client-0.5.4-1.el7ost.noarch
libreport-rhel-2.1.11-10.el7.x86_64
rhel-osp-installer-0.5.4-1.el7ost.noarch

Comment 4 Mike Burns 2015-01-05 19:18:23 UTC
When bug 1175869 is fixed, this becomes notabug, so closing it as a duplicate since it will be resolved.

That bug makes these services A/A so they can run on multiple hosts.

*** This bug has been marked as a duplicate of bug 1175869 ***


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