We're not exposing the json_dump from the UI, so to fit our current retention policy and still fit into the AWS ES infra we need to start redacting the json_dump.

We will need a strategy on how to actually get some of that data into ES in the future and use it. There are multiple bugs open and questions coming in repeatedly on use cases for searching at least parts of that json_dump data.

We have been discussing this yesterday during the Socorro meeting, and I'll come up with a proposal sometime soon. At first we are going to remove the json_dump entirely though, because of disk space issues.

PR: https://github.com/mozilla/socorro-infra/pull/158