我使用了OVS网络，并启用了BR。具体命令是
ovs-vsctl add br br ens192
ovs-vsctl add port br ens192 ens192
在仪表盘中创建了两个网络，一个是vxlan网络（内部），一个是flat网络（外部）。
运行在vxlan上的云主机可能很长时间都收不到元数据信息，如果使用curl http://169.254.169.254时间较长，就会被访问。
我可以使用虚拟机来运行 OpenStack 吗？
云主机-cirros-image-logs
在扁平网络中，这是正常的。从vxlan网络的云主机部分日志中，我们可以看到无法访问这个cloud init：
udhcpc: sending discover
udhcpc: sending select for 10.0.0.140
udhcpc: lease of 10.0.0.140 obtained, lease time 86400
route: SIOCADDRT: File exists
WARN: failed: route add -net "0.0.0.0/0" gw "10.0.0.1"
OK
checking http://169.254.169.254/2009-04-04/instance-id
failed 1/20: up 2.31. request failed
failed 2/20: up 51.34. request failed
failed 3/20: up 100.37. request failed
failed 4/20: up 149.40. request failed
failed 5/20: up 198.42. request failed
failed 6/20: up 247.45. request failed
failed 7/20: up 296.48. request failed
failed 8/20: up 345.50. request failed
failed 9/20: up 394.53. request failed
failed 10/20: up 443.55. request failed
failed 11/20: up 492.58. request failed
failed 12/20: up 541.60. request failed
failed 13/20: up 590.62. request failed
failed 14/20: up 639.65. request failed
failed 15/20: up 688.67. request failed
failed 16/20: up 737.70. request failed
failed 17/20: up 786.72. request failed
failed 18/20: up 835.74. request failed
failed 19/20: up 884.77. request failed
failed 20/20: up 933.79. request failed
failed to read iid from metadata. tried 20
failed to get instance-id of datasource