When you are developing a service that needs to communicate with other services or has to support integration with customer services, you should consider implementing a proper retry mechanism. Network connections can fail, other services can be temporarily unavailable to do maintenance or rescheduling of Pods to another node. Even running your services on Kubernetes these days doesn’t guarantee 100% availability. Redelivery is the king!
You can ask an obvious question why do I need this when I can use a built-in backpressure mechanism of my library/framework/etc.
The problem with backpressure is that you are stuck, you are stopping processing…
In my previous post I showed how you can implement an event bus using Akka and how to handle incoming and outgoing messages using Websockets. Now it’s time to feed the event bus with some internal events and also support scaling it up on multiple instances.
Instead of depending directly on the event bus I will introduce a publisher object, an intermediate component, which then can be used by various different components to publish events into the event bus.
Adding additional abstractions allows to easily mock it in unit tests and also replace undergoing logic without changing the users of…
You’ve probably read such posts many times and you are wondering what more I can tell you about this topic. Maybe not much but my idea for this post was inspired by a real case scenario I had implemented in my project. I will start with events going in/out with Websockets which are then propagated to Akka actors. In the following post I will add RabbitMQ to support multiple pods/instances of the app.
In my previous post I started a preparation process of allowing to run a Hubot instance in the Kubernetes cluster. I presented how to set up a Docker image, how to use MySQL as Hubot’s brain and so on. Right now we are ready to jump into Helm and Charts and do a real Kubernetes stuff!
Just one remark: I assume you have a Kubernetes cluster running plus you have an account on Docker Hub to push your images there. Also, DevOps already installed Jenkins on the cluster which you can use for deployment.
The very first option to setup and run your own Hubot is Heroku. There is a whole section in Hubot’s documentation how to do it. This is a very easy and straightforward solution, yet it’s true as far as you are running just Hubot and you don’t have any other services to maintain.
In a modern company, consolidating your services and resources on one platform reduces maintenance and operations costs, not to mention better knowledge sharing inside the IT department. You do not need to hire multiple specialists to support each platform, you just need one YAML developer ;-)
In a project I’m involved in, we started using RabbitMQ to support internal communication between different microservices. At the beginning we have been using rabbitmqadmin to create all the queues at startup — this isn’t a production ready solution. Let’s read how you should configure RabbitMQ to fully control who changed what.
As I mentioned we started with a simple case, right now we are extending the usage of RabbitMQ and adding support for collecting all information and events happening in different parts of our system. …
You all know these little images showing the latest version of an artefact or if the tests are passing or showing coverage of the project. We all love them, they are small but so informative. With just one image you convey a bunch of data.
The Maven Badges project was started by Jakub Jirutka few years ago and as such was hosted on Heroku, on a free dyno.
The main purpose was to provide missing features of shields.io. Right now shields.io …
Yes, you can be surprised but after almost 18 years on the market the Apache Struts project is still maintained and under active development. Ok, this is not exactly truth as Struts 1 was announced End Of Life a few years ago and Struts 2 (Struts 1 successor) is a totally different framework. Anyway, when you started using Struts 2 few years ago you can still be sure that you will get the latest patches or security fixes in a timely manner.
Oh, and Struts is a Java web framework so if you are not a Java developer you can…
A few days ago I gave my first presentation on Bielsko Biala JUG. It wasn’t my first presentation but it was my first time in front of the new audience and with a new topic.
Yes, I was talking about Halinka and Watson, showing some code examples and explaining how we can use bots and how we can connect them to an Artificial Intelligence to escalate the bots’ possibilities. There was also a plenty of discussions about different use cases. One of the most interesting is about bot’s behaviour:
should the bot behave like a human? should it be treated…
I would like to share three hacks that you should consider when implementing a bot using Hubot. They can make your daily use of Hubot a better experience.
OSS enthusiast, ASF committer, Apache Struts lead, developer, husband and father and biker :-)