As some of you may know, traffic within AWS is free, and traffic going outside of AWS has a small cost. We had never restricted outgoing traffic because it makes sense for our customers to access their database from their local environments. Everyone needs to run administrative commands every now and then, and that's totally fair.
Recently, we noticed an increase in traffic going outside of AWS, to the point that we were operating at a loss. We tried to contact all our customers to let them know that we would restrict that kind of usage, and effectively did so today. After all, we thought, restricting traffic to AWS was not that bad, as everyone would be able to connect to their databases via their EC2 instances.
Turns out we made a big mistake. For one, we thought an email was enough warning, and it clearly wasn't. Then we didn't offer an explanation of the possible workarounds, and today people just couldn't connect to their instances with redis-cli, and had no idea about what was happening.
That's why we did two things to mitigate this problem:
1. We decided to restore the traffic outside of AWS until we had a solution.
2. We found a solution that will work for everyone.
Starting today, every customer will have two ways of accessing their
instances: an OPENREDIS_URL that can be used within AWS, and that
has unlimited traffic, and a public REDIS_URL that is rate limited
to 10 GB per month.
You can get your proxy REDIS_URL by logging into
your dashboard and clicking on "Options".