The short version is as follows: /dev/random is basically the entropy pool. There isn't much data in there and whatever lands in it must be carefully chosen to prevent entropy going down (Debian accidentally broke that part a few years ago). Since there isn't much data you probably shouldn't take too much of it when you can help it because if it runs out, it blocks until there is enough entropy back in the pool.

/dev/urandom on the other hand is a CSPRNG that is regularly re-seeded from /dev/random. The purpose is to stretch the avilable entropy to more (pseudo-)random numbers. Since it's a CSPRNG the sequence is deterministic but impractical to predict, re-seeding helps in preventing others from observing too much output to reverse-engineer the seed.

So advice is generally to only use /dev/random when you're implementing something like /sev/urandom and for SSL keys and the like you just use /dev/urandom.