If you want to create a node.js environment on EC2, you have three options during recording.
First, use the Amazon Linux server (it looks like you tried it). In some cases, this can be beneficial for reasons beyond your control of node.js, such as price and the current image for Amazon.
However, as you may have noticed, there is no ready-made node.js package available through the yum package installer for Amazon linux. This means that you need to download the source code, compile it and install it yourself - the node js source is available here: git clone git: //github.com/joyent/node.git. Then, in a typical unix style, you would do something like: ./ configure -> make -> make install. You must do the same with npm. I did this several times without any problems.
The second option, which may not be the easiest for new system administrators, is to use AWS Elastic Beanstalk. Elastic Beanstalk has only recently started to support node, but provides automatic scaling of node.js production environment from tin. It also provides a configuration management approach for deploying multiple environments. Ultimately, if you expect that you will have a large number of users and need to implement a scalable node architecture, this is the approach that you are likely to use to bring your node application to production. Disclaimer: I tried this and found it to be quite complicated.
As already mentioned, the last option is to choose a Linux distribution that already has a ready-made node package for installation.
As for Couchdb and Redis, if you are going to start writing a node.js application, you might want to upgrade to equivalent AWS products. You can look at AWS simpledb, for which you will find a reliable and useful module here https://github.com/rjrodger/simpledb , and instead of using Redis you can use elastic http://aws.amazon.com/elasticache .
Happy hacking node.
source share