Skip to main content

Morio v0.6.0

· 7 min read
Joost De Cock
Joost De Cock
Morio Maintainer

We have released version 0.6.0 of Morio, a new minor release that bundles a bunch of exciting improvements as well as new features. Let's have a look at the most important changes.

Announcing MorioHub​

One thing we cannot know about Morio is how people will end up using it. What data will be processed, or what business challenges are being addressed. Essentially, when it comes to the data plane, we cannot know for sure what things will look like, so what we want to offer is flexibility to deal with a variety of scenarios.

To some extent, we can foresee such flexibility and incorporate it in our design. But at the end of the day, we need a way for people to be able to plug in their own logic or configuration in an easy way.

To that extent, we currently have 3 different ways you can add your own logic or configuration to Morio:

  • A settings overlay adds certain settings to your Morio setup
  • A client module bundles configuration for the various client agents
  • A stream processor implements business logic for our tap service
tip

Some of these are new since this release, so keep reading to learn more.

To make all of these available -- as well as facilitate contributions from the community -- we have started building out MorioHub.

MorioHub is a centralised repository for these various plugins for Morio. Specifically, it is a dedicated git repository from which you can load client modules, stream processors, or even some of Morio's settings.

To facilitate discoverability, we have also integrated MorioHub in this website, and we plan integration with the Morio UI in a future update.

Stream processing & BYOBL​

Since Morio uses Kafka -- nay, RedPanda -- as its streaming data backbone, you could already do stream processing on Morio's data. But you had to roll your own solution, which is non-trivial.

In this release, we are proud to introduce Morio's new Tap service which allows you to tap into the data stream (hence the name) without having to handle any of the complicated stream processing yourself.

In practical terms, this is a low-code solution that allows you to bring your own business logic ().

In addition to offering the capabilities to deploy your own stream processor with minimal knowledge required, we also publish several stream processors on MorioHub that do a bunch of common tasks, such as caching log data, or metrics, building out an inventory, and so on.

This feature relies on the cache service as well as the new inventory tables, but more on that later.

For more information, please refer to the Tap stream processing guide.

The new cache service facilitates dashboarding (wip)​

In tandem with Morio's new stream processing capabilities, we have added the new cache service.

This allows us to provide dashboarding in the Morio UI, giving real-time insights into the data that is flowing through Morio.

tip

This is a work in progress, so expect more updates in subsequent Morio releases.

New inventory tables and API endpoints​

Another feature set that is tied to the new stream processing capabilities are the new database tables and API endpoints to build out a Morio inventory.

We have created a stream processor for this that you can enable to have Morio build out an inventory over time.

This too is something we'll build upon in future updates.

Improvements to the Morio client and its templates​

We've also made important changes to the Morio client, and sweeping (and breaking) changes to the way we handle client modules, or their templates if you will.

Essentially, we have significantly reduced the amount of boilerplate, making it easy to implement your own client modules for Morio.

We've also changed our client build to generate a statically linked executable, making the client more portable on different (versions of) operating systems.

And much more​

There's a lot more that went into this release. Here are some more highlights:

Support for raw LSCL input in connector config​

The connector service now supports raw LSCL for configuring your data pipelines, thus allowing you to reuse any existing Logstash pipelines you may have.

Support for editing settings by hand through the UI​

The UI now also allows you to hand edit the entire settings file, thus giving you the full power of updating the settings through the API, through Morio's UI.

Changes to preseed settings, and reseeding support​

We've made changes to how we handle preseeding, making it more powerful and versatile. Refer to the preseeding guide for all details.

We're also deprecating preseeding via GitHub or GitLab APIs and will only support preseeding via git going forward.

Add the force_mrt parameter to avoid lockout​

We've added a force_mrt URL parameter to Morio's sign-in page to force the inclusion of the Morio Root Token identity provider on the sign-in form.

Note that this does not bypass the DISABLE_IDP_MRT feature flag. Rather, it prevents you from locking yourself out of the UI if you neglect to add the MRT provider to the sign in page.

Images are now published for our 3 release channels​

We have completed support for our 3 release channels:

  • stable: Our best work
  • canary: Preview releases of what we intend to release as stable
  • testing: Bleeding edge releases for testing and development. Not stable.

All docker images are now tagged as follows (using 0.6 as an example version):

  • v0.6: The 0.6 release in the stable channel.
  • v0.6-canary: The 0.6 release in the canary channel.
  • v0.6-testing: The 0.6 release in the testing channel.
  • stable: The most recent release in the stable channel.
  • canary: The most recent release in the canary channel.
  • testing: The most recent release in the testing channel.
  • latest: The most recent release in the stable channel (this is an alias for stable, and what Docker uses if no specific tag is provided).

For software packages, each release channel has its own repository.

Generic APT repositories​

When we initially created our Debian software repositories, they were tied to Debian Bookworm. We have since recreated these repositories, and now they are not tied to a specific Debian version any more, or even to Debian itself for that matter.

It means that you can use them on a different Debian version, or even on Debian-based distributions such as Ubuntu.

The Morio client is now statically linked​

We have changed the way we build the Morio client to produce a statically linked binary. This makes the client more portable across different (versions of) operating systems.

Extended use of a systemd environment file for moriod​

We have a new moriod.env file to pre-configure the Moriod service. This allows you to change certain Moriod presets, much as you would do by passing custom environment parameters to a container.

Provide guidance for FQDN mismatch during setup​

Morio will not validate a setup when the FQDN used to access Morio in the browser is not included in the list of broken FQDNs. We've now added a helpful message to help identify this issue.

The /up API endpoint is now anonymous​

This endpoint is provided for fast health checks on the API, and as such no longer requires authentication.

Speed up reloading and restarting of Morio​

We've made changes to the various life cycle methods of the services in Morio core, and in doing so have significantly reduced the time needed for Morio to reconfigure itself or restart.