This guide will show you how can you customize and configure IceCI’s installation for your cluster.
We’ll focus on customizing the deploy by modifying the YAML manifests provided in IceCI’s repository.
Before we start, clone IceCI’s repository so you can make the modifications locally.
git clone https://github.com/IceCI/IceCI.git
The storage configuration manifests can be found in
manifests/operator/pvc-storage.yaml. As you can see, two types of storage are defined - static and dynamic storage. In the provided installation, the volumes are configured as 1Gi host paths. As mentioned in the installation page - it will work fine for single node clusters. You might want to change the type and the requested disk space though, depending on the kind of setup you’ll be deploying the application in.
The names of the storage PVCs are defined in the environment variables in
manifests/operator/operator.yaml - you can create the PVCs with different names, but remember to update them in the operator configuration.
The operator isn’t the only application making use of those PVCs - keep in mind that the sync app uses them as well! You can find it’s manifest in
The default IceCI manifests deploy a PostgreSQL instance for persisting pipeline and build data - it can be found in
manifests/app/postgres.yaml. If you already have a database up and running that you’d like to use with IceCI, you can simply skip the creation of that DB and configure the applications to use your own. All you have to do is configure the following environment variables in the API and sync apps. Their manifests can be found in
- name: ICECI_DB_USER value: <db_user> - name: ICECI_DB_HOST value: <db_host> - name: ICECI_DB_NAME value: <db_name> - name: ICECI_DB_PASS value: <db_pass> - name: ICECI_DB_PORT value: <db_port> - name: ICECI_DB_DIALECT value: <db_dialect>
Resource requests and limits¶
By default, the applications deployed in the cluster don’t have any resources requests nor limits set. You can add them easily by editing the deployment manifests for each of the applications that you decide do deploy in the cluster. You can read more about resource requests and limits in the Kubernetes documentation.
How that you’ve got IceCI installed and configured in your cluster, it’s time to build some pipelines. If you need a quick refresher on how a basic pipeline is built, you can go back to the Quickstart page or visit one of example repositories, like the Quickstart example or the Python Flask API example. For a more comprehensive guide on how the pipeline configuration files are structured - check out the Pipelines page.