I have a Java Servlet/JSP application which requires the user to upload an archive file (either .rar or .zip). This archive file is then extracted, and the extracted files are parsed. After parsing the files, the data in them is added to the database and the files are deleted again.
On my local machine, this works perfectly, since you just use the filesystem provided by the OS. But now I'd like to run this application on Heroku and I'm unsure on how to do the file uploads.
Since these files are user specific, and not permanent, my initial thoughts were that I could just use the ephemeral file system provided by Heroku and I do not require the use of S3.
At the moment, my application runs on only 1 web dyno and no worker dyno's but in the future this may get scaled to multiple web dyno's, depending on the amount of users that are going to use it.
Can I use the ephemeral file system for my specific use case, and will it scale properly?
I am currently writing using ServletFileUpload, and am writing to java.io.File; Can I just change the path of my java.io.File to a path in the ephemeral file system? What would be an example of such path?
1 Answer 1
I guess you can use the ephemeral fileystem in your specific case, as it's just a temporary usage for parsing the file.
You can use the /tmp directory but keep in mind that the file will be destroyed after the request is complete.
This is discussed in this post https://stackoverflow.com/a/12416923/476782