Upgrading Sensor Collector

This article outlines the procedure for upgrading a Sensor Collector.

Instructions are provided for Kubernetes and Docker environments.

How to check if existing Sensor Collectors require software upgrade

The release notes will indicate if a new version of the Sensor Collector is available:

• On Premise Continuous Controlled Availability release
• Cloud Continuous Controlled Availability release

If there is a new version of the Sensor Collector for the software release you are using, you can follow the instructions below to perform the upgrade.

Users can check the software version of each of the Sensor Collectors in their environment by going to Sensors -> Collectors -> Sensor Collectors and looking at the Version column, shown below.

The latest available version is also displayed in the PCA User Interface for easy reference, shown here as an example:

If the Sensor Collectors in the environment require an upgrade, follow the procedure below for each Sensor Collector that needs an upgrade.

Upgrade Procedure: Docker Environment

Important: In order to prevent issues during the upgrade, you must stop the Docker container currently running the existing Sensor Collector before deploying and running the latest version. This ensures a clean transition and avoids conflicts.

To upgrade a Sensor Collector in a Docker environment:

1. Download the Updated Sensor Collector from the user interface (it is the same process for initial download and upgrades):
   1. Navigate to your Sensor Collector by selecting Sensors > Collectors > Sensor Collector.
   2. Select your Sensor Collector from the list and click the ellipsis (three dots) in the upper-right corner and choose Download (Docker). This action triggers the download of the updated package to your local machine.
   3. Extract the Downloaded Tarball (.tar.gz) file at the location of the Sensor Collector you are upgrading.
      Example:

➜ mv sensorCollectorArchive.tar.gz sensorCollectorDir
➜ cd sensorCollectorDir/
➜ gunzip sensorCollectorArchive.tar.gz
➜ tar xvf sensorCollectorArchive.tar

2. Terminate the Existing Docker Container.
   Note: If your Sensor Collector is currently running in an environment (usually the case in an upgrade scenario), you must terminate its Docker container before running the shell script to pick up the new software.

To terminate, view running containers using the docker ps command: docker ps, as shown below:

Example output:

➜ docker ps
CONTAINER ID   IMAGE     COMMAND   CREATED   STATUS    PORTS     NAMES
8a705a524dac  gcr.io/npav-172917/adh-roadrunner:0.616.0  data --config=/tm...  7 seconds ago  Up 8 seconds              aod-connector-for-united.medley.npav.accedian.net-kevin-rr-test-f6ce394a-54f8-427b-be71-a10582d9370b

3. Find your container in the list, and use the container ID to stop it:

➜ docker stop 8a705a524dac

4. If you want to maintain the environmental overrides configured for your sensor collector, copy the .env file from your old sensor collector working directory to your new one.

5. Launch the Upgraded Sensor Collector again using the run.sh script located in the extracted directory:

➜ ./run.sh

Upgrade Procedure: Kubernetes Environment

To upgrade a Sensor Collector in a Kubernetes environment:

Follow the steps outlined in Deploying Sensor Collector in Kubernetes. Ensure your values.yaml uses the same name and namespace for you sensor collector pods. To determine the existing name and namespace, you can run helm list --all-namespaces.

For example:

~# helm list --all-namespaces
NAME                            NAMESPACE               REVISION  CHART                         APP VERSION
sensor-collector-rr-testairgap  cisco-sensor-collectors 1         sensor-collector-0.681.0-fips 0.681.0-fips

When you run the helm helper script, it will prompt you to stop your running Sensor Collector pod so it can be upgraded.

As shown in the last prompt below, answer "Yes" to stop the running sensor collector

~# ./sensor-collector-install.sh --local-chart ./sensor-collector-0.681.0-fips.tgz
Using local helm chart: ./sensor-collector-0.681.0-fips.tgz
Perform 'helm install' of sensor collector instance to Kubernetes cluster

⚠️  ZONE 'RR-TestAirgap' is not Kubernetes-compatible, formatted to 'rr-testairgap'
    Performing preflight checks...
        helm installed ------------------------------ PASS
        kubectl installed --------------------------- PASS

        confirm required environment variables set in local .env file:
            KUBERNETES_INSTANCE_NAME set ------------ PASS
            KUBERNETES_NAMESPACE set ---------------- PASS
            SENSOR_COLLECTOR_TYPE set --------------- PASS

        Verify required support files present:
                values.yaml - PRESENT

    Preflight checks passed


Existing release 'sensor-collector-rr-testairgap' found in namespace 'cisco-sensor-collectors'
This will be an upgrade of the existing installation

    Checking health of existing pods...
    ✓ All pods are healthy

Pods are healthy. Force recreate anyway? [y/N]: