Knife ssh not work when a node is not bootstrapped
I have a node which is registered to chef server when its first time run the chef-client, so I can see the node in the web interface.
But I can’t run knife ssh to it, return error
So, in this case, what is the best way to copy the ssh key?
Chef server failover
I am evaluating chef for our company and so far it looks great. One question that comes to my mind from looking at the chef architecture is how to make it HA. I’ve started reading documentation about scaling couchdb and failure-proofing it, but I can’t find anything chef-server specific on the documentation site. Can anyone point me to a good source of docs on how to make sure that the chef-server is not a single point of failure in your cluster?
I appreciate your help,
How to pass attributes from knife command via ssh?
I am wondering if it is possible to pass attributes from knife command (via ssh), e.g. https://github.com/opscode/cookbooks/tree/master/java, I want to pass the
knife ssh ‘role:app-server’ ‘sudo chef-client’
ValidationFailed on AMI Ubuntu 12.04 and Chef server 11
I installed a fresh chef-server_11.0.6-1.ubuntu.12.04_amd64.deb on Amazon EC2 using Ubuntu 12.04 64 bit images. The server seems running OK, I can login through webUI.
I try to run to knife configure -i , I get
When I run knife client list, I did get two client return.
Chef sever on Debian
I see there is no Chef 11 server package for Debian. Is there any plan to provide on in the future? I can always try the Ubuntu debs, but I assume that isn’t supported.
Edit: I went ahead and used the Ubuntu 11.04 deb. Turns out 12.04 uses a newer version of glibc than Debian 7, so that one won’t work.