Leer en español

Using run.sh

The run.sh script automates the deployment process for a k8s environment using minikube. It includes building docker images, customizing postgres versions, enabling SSL, and configuring replicas for the API.

Requirements

• minikube
• docker
• helm

Usage

./run.sh -i  # Interactive mode
./run.sh -p <version> -s <Y|N> -r <number>  # Non-interactive mode
./run.sh -p <version> -s <Y|N> -r <number> -f  # Non-interactive mode with Fluentd enabled

Command Line Options

• -i: Interactive mode
• -p : Specify the postgres version
• -s <Y|N>: Enable or disable SSL (Y for yes, N for no)
• -r : Specify the number of replicas for the API
• -f: Enable fluentd for log aggregation
• -u: Uninstall the chart
• -d: Add basic data to the DB

Examples

Changing postgres version

If you want to modify the database version, you can simply do:

./run.sh -p 13.1

This will update everything necessary for this change.

Specifying number of replicas for the API

./run.sh -r 6

Script Functionality

1. Minikube Status Check:

   The script checks if minikube is already running and starts it if it's not.

2. Postgres Version and Build:

   It prompts the user for the postgres version, builds docker images, and updates the version in the helm values file.

3. TLS Configuration:

   It asks the user if they want to enable TLS and updates the helm values file accordingly.

4. Configuring Replicas for the API:

   It prompts the user for the number of replicas for the API and updates the helm values file.

5. Ingress Configuration:

   It enables the ingress extension if minikube is running.

6. Fluentd Configuration (Optional):

   The user is asked whether to enable fluentd for log aggregation. This can be toggled directly in the values.yaml file. It's worth noting that, to maintain simplicity and avoid "bloat" in values.yaml, an additional file named fluentd.yaml is provided. This file should be passed as a parameter -f to helm to configure fluentd with necessary parameters. It's optional, as Fluentd can function without these configurations, but it's recommended due to performing some container startup conditions to wait for the fluentd service and to collect only logs from exam-* containers (not all logs from k8s).

   The following instruction is used when installing or upgrading helm:

   helm install exam ./helm -n exam -f helm/values.yaml -f helm/fluentd.yaml
   

   To view logs from exam-* containers, you can use:

   kubectl logs -n exam exam-fluentd-0
   

   To configure and send logs to a log backend like elasticsearch, you can edit fluentd.yaml as per your requirements.

7. Secrets Configuration (Optional):

   To avoid keeping secrets in the SVC (if one is used), they can be passed to helm via a secrets.yaml file with the -f option. If ./helm/secrets.yaml exists (in the ./helm folder, i.e., in the same location as values.yaml), these values will be used; otherwise, those defined in values.yaml will be used. These values will override existing ones.

   It's worth noting that upon creating ./helm/secrets.yaml, the run.sh script will automatically detect and use it when running helm. Therefore, no additional actions are required, as the script will handle the detection and passing of this file to helm.