Problem description
When using rdo for multi-node deployment and testing, the models of several dilapidated machines are different (some hp PCs and some Dell PCs ), the following error occurs:
Applying 192.168.40.107_neutron.pp192.168.40.107_neutron.pp: [ DONE ] 192.168.40.248_neutron.pp: [ DONE ] 192.168.40.105_neutron.pp: [ ERROR ] Applying Puppet manifests [ ERROR ]ERROR : Error appeared during Puppet run: 192.168.40.105_neutron.ppError: Local ip for ovs agent must be set when tunneling is enabled at /var/tmp/packstack/b8647a3c31b84caa8b43227bda752bc3/modules/neutron/manifests/agents/ml2/ovs.pp:107 on node computenode.example.org
Problem Analysis
Analysis Error shows that the reason is that the following parameters are configured in the packstack file:
CONFIG_NEUTRON_OVS_TUNNEL_IF=eth0
This configuration specifies that the tunnel communication between the computing node and the network node is implemented through the EM1 port, but 192.168.40.107 is an HP machine (the Network Interface Name Is ETH *), 192.168.40.105 is a Dell machine (the network interface name is em *). The above error occurs because port eth0 does not exist in port 105. To solve this problem, you must unify the port names of port 105 and port 107. Therefore, you must modify the port names of one of the machines.
Solution
Modify the network port name of the machine 105 as eth0. The specific operation is as follows:
(1) Modify name = "Em1" to name = "eth0" in the/etc/udev/rules. d/70-persistent-net.rules File"
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="70:f3:95:00:cd:b9", ATTR{type}=="1", KERNEL=="eth*", NAME="em1"
(2) Rename the configuration file
mv /etc/sysconfig/network-scripts/ifcfg-em1 /etc/sysconfig/network-scripts/ifcfg-eth0
(3) change the device in the eth0 file to eth0.
(4) restart the server
Openstack deployment Summary: "error: local IP for ovs agent must be set when tunneling is enabled" during deployment