Running RStudio on Digital Ocean, AWS etc Using Tutum and Docker Containers

Via RBloggers I noticed a tutorial today on Setting Rstudio server using Amazon Web Services (AWS).

In the post Getting Started With Personal App Containers in the Cloud I described how I linked my tutum account to a Digital Ocean hosting account and then launched a Digital Ocean server. (How to link tutum to Amazon AWS is described here: tutum support: Link your Amazon Web Services account.)

Having launched a server (also described in Getting Started With Personal App Containers in the Cloud), we can now create a new service that will fire up an RStudio container.

First up, we need to locate a likely container – the official one is the rocker/rstudio image:

Having selected the image, we need to do a little bit of essential configuration (we could do more, like giving the service a new name):

Specifically, we need to publish the port so that it’s publicly viewable – then we can Create and Deploy the service:

After a minute or two, the service should be up and running:

We can now find the endpoint, and click through to it (note: we need to change the URL from a tcp:// address to an http:// one. (Am I doing something wrong in the set up to stop the http URL being minted as the service endpoint?)

URL tweaked, you should now be able to see an RStudio login screen. The default user is rstudio and the default password rstudio too:

And there we have it:-)

So we don’t continue paying for the server, I generally stop the container and then terminate it to destroy it…

And then terminate the node…

So, assuming the Amazon sign-up process is painless, I’m assuming it shouldn’t be much harder than that?

By the by, it’s possible to link containers to other containers; here’s an example (on the desktop, using boot2docker, that links an RStudio container to a MySQL database: Connecting RStudio and MySQL Docker Containers – an example using the ergast db. When I get a chance, I’ll have a go at doing that via tutum too…

PS see also: How to Run An R Shiny App in the Cloud Using Tutum, Digital Ocean and Docker Containers.