After a period of testing Docker work is hard but after collecting and viewing the log to finish the work
The next thing I want to do is to record the biggest network problem that was bothering us. First describe the deployment environment
1, installed on the esxi5.5 operating system Linux 6.3 ESXi binding dual network card and the VM is just a piece of network card in the early days in accordance with their own want to send to try to debug the network br0 bridging the way the result is the Docker container network when the interruption is difficult to understand the premise is ESXi NIC to accept promiscuous mode, Finally, the tcpdump catch packet found that some time the IP MAC address of the continuous change is not the only MAC address
So when the time between the containers when the Br0 learning Mac to shut down the network is OK brctl setageing br0 0
2, some people say that this is esxi5.5 bug but I think that can be solved by the order is not a bug in the two days under the need to enter the OpenStack has been basically able to run the VM but the current problem needs to be considered
1. If you use shared storage later how to expand the local storage has not been tested
2. Configuration and trend of virtual routing
3, how to create a VM can directly implement the Vnet mount point Br0 and identify the vnet
4, the expansion and distribution of the network segment can only be allocated at the beginning of a segment how to implement the post-add network segment
5, the host's network card redundancy
Interested students can leave a message to discuss common progress
This article is from the "Let Me Men Grow Together" blog, please be sure to keep this source http://wujingfeng.blog.51cto.com/5725921/1792490
VMware's Docker barrier