Skip to main content

Configure Harness Code for pull request ingestion

Last updated on

Harness Code repositories are ingested as a source-control change source automatically. No connector and no webhook setup are required.

Before you begin​

  • Deploy Change Investigator setup: Build and deploy webhook integrations created in AI SRE. Go to Deploy Change Investigator to set up the webhook endpoints.
  • A Harness Code repository: The service you deploy is built from a Harness Code repository.

How Harness Code pull request ingestion works​

The Deploy Change Investigator ingests Harness Code pull requests by polling the internal Harness Code service. Because Harness Code runs inside your Harness account, ingestion uses internal service authentication.

  • Mechanism: AI SRE polls the internal Harness Code API for merged pull requests on the tracked repository.
  • Frequency: The ingestion job runs approximately once per hour.
  • Authentication: None. Harness Code is an internal service, so no connector and no credentials are needed.
No setup required

Unlike GitHub and Bitbucket, Harness Code needs no connector and no webhook. When you create a pull request ingestion, you provide only the repository URL. Ingestion is also created automatically once you send build webhooks that reference a Harness Code repository.

Create a pull request ingestion​

Ingestion is created automatically from your build webhooks. To create one by hand:

  1. In the AI SRE left navigation, go to Integrations.
  2. Open the PR Ingestions tab.
  3. Click + New PR Ingestion.
  4. In the Create New PR Ingestion dialog, under Select Git Provider, choose Harness Code.
  5. Enter the Repository URL of the Harness Code repository, for example https://app.harness.io/ng/account/ACCOUNT_ID/module/code/orgs/ORG_ID/projects/PROJECT_ID/repos/REPO_NAME.
  6. Click Create.

Create New PR Ingestion dialog with Harness Code selected as the Git provider and a repository URL field

No connector selection is required, because Harness Code is an internal service.

Known limitation: missing commit SHA​

For Harness Code pull requests, the ingested commitSha is always empty. The Harness Code API does not return the merge commit SHA to the ingestion job, so pull requests are stored without it.

Commit SHA is not available for Harness Code

Harness Code pull requests are ingested with their branch, PR number, title, merge timestamp, and author, but without a commit SHA. Correlation that relies on matching a build's commit SHA to a pull request does not apply to Harness Code. Pull requests still surface against deployments through service and version matching.

Verify pull request ingestion​

  1. Navigate to AI SRE → PR Ingestions (tab next to Integrations).
  2. Confirm an ingestion job exists for your Harness Code repository with a tracked branch and a recent sync timestamp.
  3. After the next hourly sync, confirm merged pull requests appear against your deployments in Change Management.

Troubleshooting​

Harness Code pull request ingestion job not created after sending build webhooks

Confirm your build webhook payload includes a source.repository_url that points at the Harness Code repository. Harness Code needs no connector, but AI SRE still creates the ingestion job from the first build webhook that identifies the repository.

Harness Code pull requests appear without a commit SHA

This is expected. The Harness Code API does not return the merge commit SHA, so Harness Code pull requests are ingested without one. Pull requests still link to deployments through service and version matching.

Next steps​