在Libcloud中部署现有节点

Question

I'm working with libcloud but meet the difficulty:

I can use the method deploy_node() to deploy file/shell command to the cloud node. It work perfectly except that I have to pass the create node data inside the function. That means I cannot deploy to the existing node.

For example, now I can :

node = driver.deploy_node(name='myname', size='s1', image='i1', deployment='deployment1'...)

But what I want:

node = driver.create_node(name='myname', size='s1', image='i1')
node.deploy(deployment = 'deployment1)
OR:
driver.deploy(node, deployment = 'deployment1')

This is quite important because we can get the existing node with libcloud via list_nodes() but we cannot deploy via libcloud independently.

If anyone of you have idea with this matter, please help me.

Thank you very much!!!

Answer 1

If I understand your questions, you would like to perform a configuration action on an existing node that is running (eg an AWS instance that is already up and not just instantiated with libcloud). This is not the goal of the deploy.node function in libcloud (see http://libcloud.readthedocs.org/en/latest/compute/deployment.html#deployment ). The recommendation would be to use a CM tool that can do interactive operations such as chef, puppet, saltstack, etc.