We’ve put together a short screencast to showcase some of the features of RethinkDB. The screencast walks through starting RethinkDB, inserting and updating JSON documents through the web UI using the Javascript ReQL driver, building data analysis queries (to get insight into GitHub’s commit information dataset), as well as sharding and replicating data. Check it out!
We identified a filesystem vulnerability this morning that allows read access to arbitrary files on the filesystem through the web administation tool.
A fix has been pushed for all platforms in RethinkDB 1.3.2, so please update immediately: see the install page for instructions.
The built-in web server has a whitelist to allow serving only a very specific set of files required to run the web administration tool (see the whitelist on Github).
However, we inadvertently disabled the whitelist during development. One of our engineers noticed this during a procedural code review (a policy recently introduced after the release of RethinkDB 1.2) and we implemented a fix immediately.
This short video does a quick walkthrough of RethinkDB and its key features. The video showcases the query language, the administration UI, sharding, replication, and other important capabilities of RethinkDB, all in less than two minutes. If you’d like to get a quick sense for what RethinkDB feels like, check it out!
We’ve released RethinkDB 1.3 (Metropolis) which has a large number of enhancements, and adds support for OS X and several other platforms. As of the 1.3 release, the server can now be installed on the following platforms:
Porting to OS X was one of the most requested features, and allows OS X users to try RethinkDB without installing a Linux VM. The story of long and arduous adventure of overcoming the many differences between the Darwin and Linux kernels and their surrounding environments remains to be told by @srh.
When version 1.2 was released, we were surprised (and delighted) by how many people asked about writing RethinkDB clients in their language. RethinkDB’s query language (RQL) is encoded using Google’s Protocol Buffers, which are available in many languages. So writing a client should have been relatively easy, at least in theory. In practice, our first Protobuf spec was a mess. It accreted a number of hacks and inconsistencies as the scope of the query language grew, and the scramble to release didn’t help matters.
RethinkDB Team
Michael Lucy