github/kubernetes
4
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update multi-node docker instructions:

Browse Source 
* Update etcd and add a warning about Docker 1.7.0
   * Update kubernetes to 0.21.2, Flannel to 0.5.0
   * Remove --sync-nodes which is removed.
This commit is contained in:
Brendan Burns
2015-07-13 20:58:33 -07:00
parent bdbcbe2e2f
commit ab4448a0d6
5 changed files with 25 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
docs/getting-started-guides/docker-multinode/worker.md
View File

@@ -14,6 +14,8 @@ certainly want the docs that go with that version.</h1>
<!-- END MUNGE: UNVERSIONED_WARNING -->
## Adding a Kubernetes worker node via Docker.
These instructions are very similar to the master set-up above, but they are duplicated for clarity.
You need to repeat these instructions for each node you want to join the cluster.
We will assume that the IP address of this node is ```${NODE_IP}``` and you have the IP address of the master in ```${MASTER_IP}``` that you created in the [master instructions](master.md).
@@ -26,6 +28,11 @@ For each worker node, there are three steps:
### Set up Flanneld on the worker node
As before, the Flannel daemon is going to provide network connectivity.
_Note_:
There is a [bug](https://github.com/docker/docker/issues/14106) in Docker 1.7.0 that prevents this from working correctly.
Please install Docker 1.6.2 or wait for Docker 1.7.1.
#### Set up a bootstrap docker
As previously, we need a second instance of the Docker daemon running to bootstrap the flannel networking.
@@ -59,7 +66,7 @@ or it may be something else.
Now run flanneld itself, this call is slightly different from the above, since we point it at the etcd instance on the master.
```sh
sudo docker -H unix:///var/run/docker-bootstrap.sock run -d --net=host --privileged -v /dev/net:/dev/net quay.io/coreos/flannel:0.3.0 /opt/bin/flanneld --etcd-endpoints=http://${MASTER_IP}:4001
sudo docker -H unix:///var/run/docker-bootstrap.sock run -d --net=host --privileged -v /dev/net:/dev/net quay.io/coreos/flannel:0.5.0 /opt/bin/flanneld --etcd-endpoints=http://${MASTER_IP}:4001
```
The previous command should have printed a really long hash, copy this hash.
@@ -107,14 +114,14 @@ systemctl start docker
Again this is similar to the above, but the ```--api_servers``` now points to the master we set up in the beginning.
```sh
sudo docker run --net=host -d -v /var/run/docker.sock:/var/run/docker.sock gcr.io/google_containers/hyperkube:v0.18.2 /hyperkube kubelet --api_servers=http://${MASTER_IP}:8080 --v=2 --address=0.0.0.0 --enable_server --hostname_override=$(hostname -i)
sudo docker run --net=host -d -v /var/run/docker.sock:/var/run/docker.sock gcr.io/google_containers/hyperkube:v0.21.2 /hyperkube kubelet --api_servers=http://${MASTER_IP}:8080 --v=2 --address=0.0.0.0 --enable_server --hostname_override=$(hostname -i)
```
#### Run the service proxy
The service proxy provides load-balancing between groups of containers defined by Kubernetes ```Services```
```sh
sudo docker run -d --net=host --privileged gcr.io/google_containers/hyperkube:v0.18.2 /hyperkube proxy --master=http://${MASTER_IP}:8080 --v=2
sudo docker run -d --net=host --privileged gcr.io/google_containers/hyperkube:v0.21.2 /hyperkube proxy --master=http://${MASTER_IP}:8080 --v=2
```
### Next steps