Atlassian Jira Ticketing Support for CI/CD

In 2018, Harness shipped 24 major platform integrations for our customers, from Microsoft Azure to HashiCorp Terraform to Datadog. Today, we have over 40 different integrations to enable end-to-end Continuous Delivery across your apps, clouds, tech stacks, and tools.

One of the most requested integrations in 2018 was Jira ticketing support. Atlassian Jira provides customers (specifically dev teams) with project and issue tracking throughout their software delivery lifecycle. Integration with Harness Continuous Delivery was, therefore, a "no brainer."

Our customers requested the ability to do two simple things:
1. Track the progress of deployment pipelines in Jira.
2. Use Jira to approve/reject deployment pipeline stages or workflows.

I'm therefore very pleased (and relieved) to announce that both use cases are now available in Harness.

Having played with the integration the past week, it's actually pretty slick and easy to set up.

Integrating Jira With Harness

Integrating Jira with Harness takes less than a minute.
To get started go to Setup > Connectors and click Add Collaboration Provider

Next, select Jira from the type dropdown, provide an instance name, and supply the URL, username, and password for the Jira instance you wish to add:

Finally, click the TEST button to verify your Jira connection. When completed, your Jira integration will look something like this:

Note: You can add/manage multiple Jira instances if required.

Tracking Pipeline Progress Using Jira

Harness now gives you the ability to create or update a Jira ticket as many times as you want during the execution of a pipeline, stage, or deployment workflow.

For example, let's edit an existing deployment workflow in Harness so we create a Jira ticket when our deployment workflow execution starts, and update the same Jira ticket when that workflow successfully completes.

To do this, let's open and edit an existing deployment workflow:

To create our Jira ticket, let's click Add Step under the Pre-Deployment Steps dialogue.

Harness now gives us a list of commands we can add. You can see that Jira has been added as a new Collaboration command, so let's select it.

To create a Jira ticket, we simply configure the Pre-Deployment Step as follows:

Notice we can use Harness variables (using $ pre-fix) to populate Jira tickets with workflow, artifact, and pipeline context.

We can also create a new variable called "Jira" from the output so we can reference the same Jira ticket again in our workflow should we want to make updates.

To update the same Jira ticket we can configure a Post-Deployment Step in our deployment workflow as follows:

Notice we use the variable ${Jira.issueId} as a key to reference the same Jira ticket we created in our Pre-Deployment Step. This means we can incrementally add activity to the same Jira ticket without needing to remember ticket ids or identifiers.

Now, when we next run our deployment workflow, we see our Jira steps fully visualized (screenshot below) with context. Notice the deep-link to the Jira ticket we've just created in the bottom right of the screenshot:

Clicking on the deep-link shown in the above deployment workflow takes us directly to the Jira ticket to reveal all the activity we logged during the pre-deployment and post-deployment steps:

So there you have it: full progress tracking of your deployment pipeline in a single Jira ticket!

But wait...there's more!

Approving Pipelines Using Jira Tickets

Up until now, our customers could only approve/reject pipeline stages or workflow execution using big shiny buttons in our Harness UI:

Today, customers can now do all this inside a Jira ticket by simply updating the value of a pre-defined Jira ticket attribute.

For example, changing the Status attribute of a Jira ticket to "Deploy" or "Approved" could automatically approve the execution of a pipeline stage or deployment workflow in Harness.

Again, this literally takes a few minutes to configure and you're all good to go.

Adding Jira Approval For Pipeline Stages

Let's imagine we have a simple 3-stage deployment pipeline consisting of Development, QA, and Production.

Now let's imagine we want a new Jira approval stage to govern how code is promoted from QA to production.

Click on the + symbol in between the QA and Production stage. Now, let's define what our Jira approval criteria should look like:

Select the Jira instance and Project you wish to use, then set the key/issue id for ticket context.
Lastly, define the ticket attribute/value criteria which will approve/reject your pipeline stage.

In the example above changing the ticket Status to Done will approve the pipeline stage and changing the ticket Status to "To Do" will reject the pipeline stage.

Once our Jira Approval is configured our pipeline now has 4 stages:

Adding Jira Approval For Deployment Workflow Execution

Let's go back to our original deployment workflow which we added Jira ticket updates for:

We can actually add a Jira approval to any part of this workflow.

Let's add a new Jira approval step after our Pre-Deployment "Create Jira Ticket" Step:

We can do this by selecting an Approval command from the Others list:

Like we did for the pipeline stage Jira approval, we simply define the approval criteria for our workflow execution:

Now when we run our deployment workflow, execution will wait until someone approves the ticket in Jira.

This can be done by updating the Jira ticket Status attribute to Done. Doing this will cause the deployment workflow to continue:

There we go. In just a few minutes you can configure Jira to approve your pipeline stages or deployment workflow execution. Bob is your Uncle.

What's Coming Next?

In a few weeks, we'll extend our Jira support to allow Jira tickets to be created for the following events:

• Deployment Verification Failure (e.g. new performance or quality regression)
• 24x7 Service Guard Failure (e.g. performance/availability)
• A Rollback

For example, a developer might deploy a new microservice into production and our Continuous Verification might identify a new performance anomaly shortly after that needs to be captured, managed and actioned using a Jira ticket. This ends up providing a closed feedback loop to the developer.

We'll also have identical ServiceNow integration soon as well!

Want to try Harness? You're 2 minutes away from getting started with our trial.

Cheers,
Steve.
@BurtonSays