by Andrie de Vries
I am preparing for a demonstration of functionality of R at a conference next week. For maximum impact, I wanted to use a fast virtual machine in Azure. It is actually very easy to build a fresh machine (cloud or otherwise) that contains R as well as RStudio server. In essence, all you have to do is:
If you were to build this on a virtual machine manager, e.g. VMWare or VirtualBox, this would be all you have to do.
However, in the cloud, you may want to have additional security. This is why Azure virtual machines blocks any incoming traffic, unless you explicitly open the required ports.
In the “classic” Azure virtual machines, this process is well-documented. The process is to create an “endpoint” that opens a specific port. For example, see this video with step-by-step instructions.
However, for the more modern new “Resource manager” virtual machines the process is different.
But first we have to answer the question “what is a resource manager virtual machine”? Well, this is answered concisely in the “Azure Resource Manager overview”. Among the many benefits are that you have more flexibility and power to build groups of machines, and you can also create your machines using scriptable templates.
Back to my problem: I knew I had a resource manager VM (rather than the classic VM). But the term “endpoint” only applies to the classic machines.
The equivalent in the resource manager is to create an “inbound security rule”.
An inbound security rule for RStudio
Since RStudio server uses port 8787, you need to open the port 8787 in the destination range. (This 8787 is the number after the colon in the URL to get to your instance of RStudio, e.g. http://:8787/)
Detailed setup of the security rule
Although it seems complex, you …read more
Source:: r-bloggers.com