Choria Blog

Choria introduced a Centralized AAA model in 2019 that alleviate the need for managing certificates of every user and allow you to integrate Choria into your enterprise identity providers for Authentication, Auditing and Authorization.

For controlled environments this model is a huge boom, but there was one annoying thing - the need to still issue a TLS certificate to communicate with Choria Brokers. In this mode, these certificates do not form part of the security model of Choria but was nonetheless required to exist, you could share them but that was frowned upon.

In our next release we will introduce a new broker type that significantly simplifies the AAA security model by allowing clients holding no certificates to interact, safely, with Choria networks.

In the Scout Announcement blog post I mentioned we are looking to integrate Goss into Scout and I wanted to post an update on that.

Background

Goss is something similar to serverspec - it lets you write unit tests about your nodes actual state rather than code used to build it. Goss definitions are written in YAML or JSON and supports Go templating for customization.

This model is well suited for the purposes of monitoring since you can write really in depth sets of validations and treat them as a single unit.

Goss is written in Go, very fast and thanks to a lot of work I did recently embeddable in other software.

We had a release quite recently but I wanted to release a number of Scout related features to early adopters, these releases are mainly focussed on Scout but includes a few bug fixes and new builds for Ubuntu Focal (20.04 LTS).

The big item here is that we have integrated Goss into the Scout framework and it can now run validations regularly. See the Scout Goss blog post for details.

You’ll also notice a new agent - scout - on your nodes, this gives API access to interact with Scout checks on Choria servers.

Additionally, we are starting to work on our documentation for Scout, an initial cut of this is also published today, this shows our Puppet integration, Prometheus integration and a bit about the events.

Thanks to Romain Tartière for contributions to these releases.

Read on for the full details.

We have a number of releases to announce today, the focus is general quality of life improvements in addition to the features to support out larger Choria Server release that included our announcement of Choria Scout.

With these releases you can create Scout checks on your machines using:

choria::scout_check{"check_typhon":
    plugin            => "/usr/lib64/nagios/plugins/check_procs",
    arguments         => '-C typhon -c {{ o "warn" 1 }}:{{ o "crit" 1 }}',
    remediate_command => "service typhon restart",
}

In addition to this we have fixed mco puppet runall when using Choria Server, I know quite a few people have wanted to see the return of this utility.

Thanks to Romain Tartière for contributions to these releases.

Yesterday I introduced a new Choria component called Scout which helps you build scalable monitoring pipelines. Today, we’ll look a bit at what makes a Scout install and how it is built.

In a follow up post I’ll dive a bit into Autonomous Agents - an infrequently used but very powerful building block found in Choria.

Overview

We’re happy to announce a new project called Choria Scout - a highly scalable system health monitoring framework and monitoring data pipeline released under the Apache 2.0 license.

Initially we support the ability to execute Nagios compatible plugins on Choria managed nodes with results sent centrally in a standard CloudEvents format, and optionally, integrated into Prometheus.

These are framework level building blocks that will in time be used to create a full monitoring stack built on Choria technologies. Checks and value overrides can already be configured using our Puppet modules. You can also use these building blocks to build entirely custom solutions for your own needs.

Scout will be a cloud native project with central components capable of being hosted on Kubernetes and using data formats supported by commercial clouds and projects like KNative. It will have a focus on integration, open data exchange and extensibility.

Despite being cloud native we will of course support monitoring anything where Choria, or the upcoming Scout agent, can run which includes traditional baremetal, VMs, containers and pods and small devices.

We have quite a significant release of the Choria Server today that lays ground work for an upcoming Choria monitoring pipeline.

Read the full post for the details.

It’s been quite some time since we’ve had releases and there’s been a huge list of small improvements.

Thanks to those who contributed to these releases: David Gardner, Mark Frost, Romain Tartière, Yury Bushmelev, @rjd1, Tim Meusel, Alexander Hermes, Vincent Janelle

Yesterday we made some quite minor changes to our app and got it to use JetStream on both the Producer and Consumer side. These changes solved several problems for us, like being able to restart Consumers without losing any messages.

The last remaining issue was around handling messages that fail to be delivered. Imagine the case where our disk is full on the Consumer, wouldn’t it be great if we can somehow communicate our inability to handle messages to the network and have it retry later?

That’s the role of Acknowledgements and JetStream supports several modes. Today we’ll look at those.

In our previous post, we dived a bit into JetStream API, and how to interact with it, many people would not need to know this all to get going. The CLI or Terraform management approaches would be perfectly fine. And today we’ll use the CLI rather than the API.

In this post, we’re back on our codebase, and we’ll see how we might need to change the tools to support JetStream well. To be honest, I could have made some better decisions early on about the shipper design, but that gives us more opportunity to see how some apps might need to adapt.