Difference between revisions of "Setting up a new node in the cluster"

From Collective Computational Unit
Jump to navigation Jump to search
m (Kubernetes)
m (Post-install configuration)
Line 8: Line 8:
 
Follow instructions [[Initializing_the_Kubernetes_cluster|on this page]] to install Kubernetes and pre-requisites.
 
Follow instructions [[Initializing_the_Kubernetes_cluster|on this page]] to install Kubernetes and pre-requisites.
  
== Post-install configuration ==
+
== Spin up the slave node ==
  
* Add the node to the different namespaces it should belong to.
+
On the master node:
** TODO
+
 
 +
<syntaxhighlight lang="bash">
 +
> cd init
 +
> ./show_join_command.sh
 +
</syntaxhighlight>
 +
 
 +
Copy the output into a new file "join_cluster.sh" in the init directory of the slave.
 +
 
 +
On the slave:
 +
 
 +
<syntaxhighlight lang="bash">
 +
> cd init
 +
> chmod +x join_cluster.sh
 +
> sudo ./join_cluster.sh
 +
</syntaxhighlight>

Revision as of 16:28, 17 June 2019

Docker

  • follow the steps to install docker, nvidia-docker2 and make nvidia-docker the default runtime on the node.


Kubernetes

Follow instructions on this page to install Kubernetes and pre-requisites.

Spin up the slave node

On the master node:

> cd init
> ./show_join_command.sh

Copy the output into a new file "join_cluster.sh" in the init directory of the slave.

On the slave:

> cd init
> chmod +x join_cluster.sh
> sudo ./join_cluster.sh