Back to bug 1036532

Who When What Removed Added
Fabio Massimo Di Nitto 2013-12-02 09:13:53 UTC Priority unspecified urgent
Mark McLoughlin 2013-12-02 09:42:48 UTC CC markmc
Summary neutron-openvswitch-agent traceback neutron-openvswitch-agent - MultipleAgentFoundByTypeHost: Multiple agents with agent_type=Open vSwitch agent and host=rhos4-node6.int.fabbione.net found
RHEL Program Management 2013-12-02 12:16:59 UTC Target Release --- 4.0
lpeer 2013-12-02 18:21:34 UTC Status NEW ASSIGNED
CC lpeer
Assignee rhos-maint jlibosva
Jakub Libosvar 2013-12-03 13:39:56 UTC CC fdinitto
Flags needinfo?(fdinitto)
Fabio Massimo Di Nitto 2013-12-03 13:44:26 UTC Flags needinfo?(fdinitto)
Fabio Massimo Di Nitto 2013-12-04 12:26:31 UTC CC jlibosva
Flags needinfo?(jlibosva)
Jakub Libosvar 2013-12-04 12:45:21 UTC Flags needinfo?(jlibosva)
Perry Myers 2013-12-05 14:48:09 UTC Target Release 4.0 5.0
Target Milestone --- Upstream M2
Maru Newby 2013-12-08 13:21:59 UTC CC mnewby
Link ID Launchpad 1254246
John Skeoch 2014-01-13 01:08:21 UTC CC hateya
Perry Myers 2014-02-16 22:39:54 UTC Flags needinfo?(mnewby)
Maru Newby 2014-02-17 18:51:41 UTC CC mangelajo
Flags needinfo?(mnewby) needinfo?(mangelajo)
Miguel Angel Ajo 2014-02-18 06:04:07 UTC CC majopela
Flags needinfo?(mangelajo)
Jakub Libosvar 2014-02-18 11:46:50 UTC Status ASSIGNED ON_QA
Link ID Launchpad 1254246 Launchpad 1257293
Jakub Libosvar 2014-02-18 11:47:13 UTC Link ID OpenStack gerrit 60525
Jakub Libosvar 2014-02-18 11:51:15 UTC Fixed In Version openstack-neutron-2013.2.1-1.el6ost
RHEL Program Management 2014-02-18 12:19:24 UTC Target Release 5.0 4.0
RHEL Program Management 2014-02-18 12:19:26 UTC Keywords ZStream
Perry Myers 2014-02-18 19:17:46 UTC Blocks 1066642
Perry Myers 2014-03-09 14:51:52 UTC Target Milestone Upstream M2 A3
Jakub Libosvar 2014-03-10 07:57:14 UTC Doc Text Cause:
RPC code wasn't able to work in cluster mode.

Consequence:
In case of clustering neutron server agent's heartbeat was sent to both neutron servers which lead in attempt to write two same piece of information into database.

Fix:
New code was taken from oslo-incubator that has capabilities for running in cluster mode.

Result:
Only one neutron server gets heartbeat and writes into database.
Jakub Libosvar 2014-03-10 08:15:58 UTC Doc Text Cause:
RPC code wasn't able to work in cluster mode.

Consequence:
In case of clustering neutron server agent's heartbeat was sent to both neutron servers which lead in attempt to write two same piece of information into database.

Fix:
New code was taken from oslo-incubator that has capabilities for running in cluster mode.

Result:
Only one neutron server gets heartbeat and writes into database.
Cause:
RPC code wasn't able to work in cluster mode.

Consequence:
In case of clustering neutron server agent's heartbeat was sent to both neutron servers which led in attempt to write two same values into database.

Fix:
New code was synced from oslo-incubator that has capabilities for running in cluster mode.

Result:
Only one neutron server obtaines message about heartbeat and writes it into database.
Ofer Blaut 2014-03-11 20:32:48 UTC Keywords OtherQA
CC oblaut
Scott Lewis 2014-03-13 17:52:25 UTC Status ON_QA MODIFIED
CC sclewis
errata-xmlrpc 2014-03-13 18:01:12 UTC Status MODIFIED ON_QA
Ofer Blaut 2014-03-18 12:01:12 UTC QA Contact oblaut nmagnezi
Nir Magnezi 2014-03-19 15:46:46 UTC Status ON_QA VERIFIED
Martin Lopes 2014-03-24 01:14:15 UTC CC mlopes
Doc Text Cause:
RPC code wasn't able to work in cluster mode.

Consequence:
In case of clustering neutron server agent's heartbeat was sent to both neutron servers which led in attempt to write two same values into database.

Fix:
New code was synced from oslo-incubator that has capabilities for running in cluster mode.

Result:
Only one neutron server obtaines message about heartbeat and writes it into database.
Prior to this update, RPC code did not function as expected in cluster mode. Consequently, clustered deployment of Networking resulted in the agent heartbeat being sent to both Networking servers, which then attempted to write duplicate values into the database.
This update syncs code from 'oslo-incubator' with added capabilities for running in cluster mode. As a result, only a single Networking server obtains heartbeat messages and writes the value into the database.
errata-xmlrpc 2014-03-25 08:01:03 UTC Status VERIFIED RELEASE_PENDING
Nir Yechiel 2014-03-25 17:12:37 UTC Blocks 1080561
errata-xmlrpc 2014-03-25 19:23:20 UTC Status RELEASE_PENDING CLOSED
Resolution --- ERRATA
Last Closed 2014-03-25 15:23:20 UTC
Perry Myers 2016-04-26 15:04:29 UTC CC pmyers

Back to bug 1036532