Hello, my name is Victor I am the product manager of the environment group at gitlab, and we are the group working on the gitlab to flux Integrations in this demo. Video I would like to show you how easy it is to get immediate flux synchronizations by using the gitlab integration with flux by default.

Without this integration flux either updates from a git Repository at the configured interval, or you need to set up a receiver, a vac hook like an Ingress together with that, and you should manage token to authenticate your web hook, codes with flux and then flux will trigger the synchronization. Now you don't need any configuration at all. If you use Agent K, together with flux from the project where Agent K is installed, let me show you how it works here.

I have a very simple project that only installs flux and has an airbag configured I installed, in this case Agent K separately, just to simplify the example demo, but you could install Agent K with flux as well, and I have agent care already set up in my cluster. You can see the in the cluster deflect system. Namespace. That's where flux is running and the gitlab agent name space where the agent is running. My agent configuration has nothing specific to the feature that I'm demoing right now.

This is actually configuration I'll use a month ago when we released another feature- and this is everything I have in my cluster- the core flux, installation and an airbag again from month ago. Now, let's see how actually the cluster will be. Synchronized I will show you one more thing here in the flux installation. This intel was back under git repository. This is set to one day, which means that flux given its configuration.

We check my grid repository once a day and I would need to wait 24 hours to get any updates from this repository to the cluster or I need to set up a receiver.

Let's see how it works actually with gitlab, as we have this receiver behind the hood in the integration I'm going to modify this airbag file here, I will just rename it to something else.

And then I will switch to the other tab where this is the previous reconciliation. Sorry, for that, where we see how the reconciliation will actually play out so once I commit the changes, it's immediately picked up by flux and the new cluster role binding was created, the old one was removed and the reconciliation is done without any configuration at all. Actually, I just need agent a and flux. Now, in the current release, this is the first iteration we have one limitation, namely actually two I would say, namely it only works with Git repositories.

We have plans to add support for oci Registries for the gitlab oci registry. Specifically, so, if you use the gitlab container registry with OCR images to store your manifest for flux, you will be able to use the same trigger mechanism there. The other limitation today is that it only works. If the agent is registered. With that specific git repository, they have plans to extend it to multiple repositories, so you will be able to reference other repositories in the agent configuration that needs to trigger this update mechanism as well. This is to come.

We would love to hear your feedback in the original issue. That's uh part of the release post, so you can find it easily if you have any feedback, please share it with us.