search

sequenceiq / cloudbreak-shell

CLI shell for the Cloudbreak project
https://cloudbreak.sequenceiq.com/
16 stars 6 forks source link

Help needed - stop cluster with cloudbreak shell #77

Closed malathit closed 9 years ago

malathit commented 9 years ago

Hi,

Can somebody help me figure out how do I stop the cluster in AWS using cloudbreak shell? I find the documentation on creating a cluster but am not sure how to stop the cluster using cloudbreak shell. I want to stop the clusters in an automated way. Hence I need to know how to stop a cluster with cloudbreak-shell.

Thanks, Malathi

lalyos commented 9 years ago

Hi Malathi,

Unfortunately start/stop is currentlynot supported by cloudbreak-shell. It's a valid need. We will check in which upcoming iteration we can fit this in.

malathit commented 9 years ago

Hi lalyos,

Thanks for the reply. In that case is it possible to stop the cluster using rest api? I learnt somewhere that creating a new client to register with UAA server is not allowed at the moment.

Thanks, Malathi

lalyos commented 9 years ago

Yes you can start/stop a cluster through cloudbreak rest api. This is what the UI does.

The actually api call would be an http PUT to the /stacks/{stackId}/cluster resource, see the api docs

So if you are familiar with OAuth2, you should be able to write a couple of lines of bash code, to achieve what you need. But the official way for cli access for cloudbreak remains cloudbreak-shell.

doktoric commented 9 years ago

added to shell with https://github.com/sequenceiq/cloudbreak-shell/pull/78 pull request

malathit commented 9 years ago

Thanks for the quick fix.

But when I tried running the commands in the shell, I get the following error: cloudbreak_1 | com.sequenceiq.cloudbreak.controller.BadRequestException: Cannot update the status of cluster '969' to STARTED, because the stack is not AVAILABLE

I was trying to start a cluster I created via cloudbreak UI.

1) selected my credentials, blueprint(that was used to create the cluster from UI), stack 2) then ran cluster start

If I am missing something, can you help with that?

Thanks.

doktoric commented 9 years ago

Hi @malathit90 , The problem here is that you have to start the stack first and if it is available then you can start the cluster as well. The workflow in cloudbreak is: Stop

  1. stop cluster
  2. stop stack

Start

  1. start stack
  2. start cluster