The source code for what follows can be found on Github.
In my last article on running a lua script against redis with lettuce, we just sent the entire script [that redis will execute atomically] along with the arguments every time. For very small scripts this is unlikely to be a problem, but there is definitely a more efficient way to do this, using EVALSHA.
The source code for what follows can be found on Github.
Running a lua script against redis is done using EVAL. The primary benefit of using a lua script is that the entire script is guaranteed to be run at once, and nothing else will interfere with it [it’s atomic]. This allows for operating on multiple keys, or check-then-set type operations on the same key.
The source code for what follows can be found on Github.
Subscribing to topics in redis allows for a fanout behavior, where any number of subscribers can be notified of a message from a publisher.
The source code for what follows can be found on Github.
Redis does not allow you to set the expiration on individual members in a set, it only allows you to set an expiration on the entire set itself. If you want to have a sort of expiry on individual elements in a set, this article shares a workaround to that problem that works well in practice. Because I have already written a lot of boilerplate code for testing any redis operation using lettuce, I’m going to be showing you this technique using a reactive lettuce client, however the basic concept should transfer easily to any other client.
The source code for what follows can be found on Github.
MSETNX when you’re working with a single redis primary node is simple enough to understand: it sets all of the key/value pairs, or none at all. If one of the keys already exists in the cluster, then all of them are rejected.
Sorted Sets in redis are one of my personal favorite tools when operating at scale. As of this writing, there are over 30 unique operations you can perform against sorted sets in redis. This article will focus on some of the more common ones you’re going to need to know, and it will use a reactive lettuce client to demonstrate them.
In the last article, we showed how to do some of the most common single set operations against redis, this article will focus on operating on multiple sets using a lettuce client against redis. Specifically, we’ll focus on subtracting, intersecting, and adding sets. The source code for what follows can be found on Github.
There are, as of this writing, about 15 distinct operations available to someone wanting to work with sets in redis. This article seeks to cover some of the more basic ones using a reactive lettuce client, and a follow up article will seek to deal with explaining some of the more common operations against multiple sets, rather than a single set in this case.
There are about 15 or so commands you can execute against redis for hash types as of this writing. This article will demonstrate some of the more common operations you’re likely to need when using lettuce as your client.
In clustered redis, any non hash tagged key can be sent unpredictably [well, actually predictably, if you know the formula] to any given primary node in the cluster. The very basic way it works is:
The source code for this article can be found on Github.
Using embedded dynamodb for testing is, in my experience, kind of flakey and unpredictable. Because of the weird way it pulls in SQLite on a per operating system basis, it can sometimes work locally and not work on the build server. Sometimes it’s just not working for some unexplained reason and wiping the directory that the code is in and re-cloning fixes it. Not a fun time.
The source code for this post can be found on Github.
Lettuce supports reading from redis replicas, but with the caveat that it doesn’t [out of the box] provide you with the fine-grained control over when to read from the replicas that you’re likely to want.
