Cloud Native Development in Practice

Zero Downtime Upgrades

You will demonstrate how to achieve zero downtime upgrades of your pal-tracker application running on Tanzu Application Service.

After completing the lab, you will be able to:

Check out the Blue/Green slides.

You must have completed (or fast-forwarded to) the Scaling lab. You must have your pal-tracker application associated with the scaling-availability-solution codebase deployed and running on Tanzu Application Service.
In a terminal window, make sure you start in the ~/workspace/pal-tracker directory.

Cherry-pick the start point of this lab:

git cherry-pick rolling-upgrade-solution

In this lab you will exercise your pal-tracker application under load.

You can monitor the pal-tracker application through the following:

Command Line via the following cf commands:
- cf app pal-tracker
- cf events pal-tracker
Apps Manager user interface.

If you will monitor via command line you will need a minimum of four terminal windows open.

If you choose to monitor with Apps Manager you will need only one.

What if you need to perform a zero downtime upgrade? How would you do that?

If you have not already, stage and commit your changes from this lab locally.
Pull in a script which will demonstrate a rolling upgrade:

git cherry-pick rolling-upgrade-solution

You can view the script here:
```
git show rolling-upgrade-solution:scripts/rolling-update.sh
```
Run a local build to compile, test and regenerate the pal-tracker.jar file:

./gradlew clean build
Run the script from the pal-tracker project directory, and make sure to supply the ${DOMAIN} and ${UNIQUE_IDENTIFIER} parameters you used in the Deployment Pipelines lab when specifying the route in your manifest.yml file:

./scripts/rolling-update.sh ${UNIQUE_IDENTIFIER} ${DOMAIN}
Follow the directions the script gives you to step through the process of a zero downtime upgrade.

Read about how you can use a newer cf command line client to perform a rolling update:

Report an issue