Bitbucket Integration

Jetscale integrates with Bitbucket to automatically create pull requests containing production-ready Terraform code for your cost optimization recommendations.

Overview

The Bitbucket integration allows Jetscale to:

• Push generated Terraform code to your Bitbucket repositories
• Create pull requests with detailed optimization recommendations
• Integrate seamlessly with your existing code review and CI/CD workflows
• Track recommendation implementation status through Bitbucket pipelines

Prerequisites

Before connecting Bitbucket, you'll need:

1. Bitbucket Account (Cloud or Server)
2. Repository Access with write permissions
3. App Password or Personal Access Token with appropriate scopes

Supported Bitbucket Versions

• Bitbucket Cloud (bitbucket.org) - Recommended
• Bitbucket Server / Data Center (self-hosted)

Setup Guide

Step 1: Create a Bitbucket App Password

Image pending: Bitbucket App Password Creation Screenshot placeholder: Bitbucket app password settings

For Bitbucket Cloud

1. Navigate to Bitbucket Settings → App passwords
2. Click Create app password
3. Give it a descriptive label (e.g., "Jetscale Integration")
4. Select the required permissions:
   • ✓ Account: Read
   • ✓ Workspace membership: Read
   • ✓ Projects: Read
   • ✓ Repositories: Write
   • ✓ Pull requests: Write
5. Click Create
6. Important: Copy your app password immediately - you won't see it again

For Bitbucket Server

1. Navigate to Profile Settings → Personal access tokens
2. Click Create a token
3. Give it a name (e.g., "Jetscale")
4. Select permissions:
   • ✓ Repository read
   • ✓ Repository write
5. Set expiration date
6. Click Create
7. Copy your token

Security Note: Store your app password or token securely. Jetscale encrypts credentials at rest.

Step 2: Connect Bitbucket in Jetscale

Image pending: Bitbucket Connection Screen Screenshot placeholder: Jetscale Bitbucket integration page

1. Navigate to Settings → Integrations in Jetscale
2. Click Connect Bitbucket
3. Enter your configuration:
   • Bitbucket Type: Cloud or Server
   • Server URL: (For Bitbucket Server only) e.g., https://bitbucket.yourcompany.com
   • Username: Your Bitbucket username
   • App Password/Token: The credential created in Step 1
4. Click Verify and Connect

Jetscale will verify your credentials and display your accessible repositories.

Step 3: Select Your Repository

Image pending: Repository Selection Screenshot placeholder: Bitbucket repository selection dropdown

1. From the Repository dropdown, select your Terraform infrastructure repository
2. Format: workspace/repository-slug (Cloud) or project/repository (Server)
3. Jetscale will verify you have write access
4. Click Save

The selected repository will be used for all future pull requests.

How It Works

Pull Request Creation

When you approve a cost optimization recommendation in Jetscale:

1. Branch Creation: Jetscale creates a new feature branch from your default branch (usually main or master)
   • Branch naming: jetscale/optimization-{resource-type}-{timestamp}
   • Example: jetscale/optimization-rds-20240115-143022
2. File Push: Generated Terraform files are committed to the branch
   • Files placed in terraform/ directory
   • Commit message includes optimization summary and cost impact
3. Pull Request: A PR is automatically created with:
   • Title: Concise optimization description
   • Description: Detailed cost analysis, performance considerations, implementation steps
   • Reviewers: Automatically added based on repository settings (optional)

Pull Request Structure

# Cost Optimization: RDS Instance Right-Sizing

## Summary
Downgrade RDS instance `production-db` from **db.r5.2xlarge** to **db.r5.xlarge**

## Cost Impact

| Metric | Current | Projected | Savings |
|--------|---------|-----------|---------|
| Monthly Cost | $730.00 | $365.00 | **$365.00 (50%)** |
| Annual Cost | $8,760.00 | $4,380.00 | **$4,380.00** |

## Performance Analysis

✓ Current CPU utilization: 15-25% average
✓ Current memory utilization: 30-40% average
✓ Recommended instance provides 2x current peak usage
✓ **Risk Assessment**: Low - No performance degradation expected

## Files Changed

- `terraform/rds_production_db.tf` - Updated instance class from db.r5.2xlarge to db.r5.xlarge
- `terraform/variables.tf` - Added configurable instance_class variable

## Implementation Checklist

- [ ] Review changes in Terraform plan output
- [ ] Apply to staging environment first
- [ ] Monitor performance for 24-48 hours
- [ ] Verify application health checks pass
- [ ] Apply to production with change management approval

## Testing Recommendations

1. Run `terraform plan` to preview changes
2. Check for any dependent resources
3. Verify backup/snapshot policies
4. Confirm monitoring dashboards are updated

---
🤖 Generated by [Jetscale](https://jetscale.ai) | [View in Jetscale Dashboard](#)

Terraform Code Example

# Optimized RDS instance configuration
# Generated by Jetscale on 2024-01-15

resource "aws_db_instance" "production_db" {
  identifier     = "production-db"
  engine         = "postgres"
  engine_version = "14.7"

  # Previous: db.r5.2xlarge ($730/month)
  # New: db.r5.xlarge ($365/month)
  # Cost Reduction: 50% ($365/month, $4,380/year)
  #
  # Justification:
  # - Average CPU utilization: 15-25%
  # - Average memory utilization: 30-40%
  # - New instance provides 2x headroom above peak usage
  instance_class = var.rds_production_instance_class

  allocated_storage     = 100
  storage_type          = "gp3"
  storage_encrypted     = true

  # ... other configuration unchanged
}

Repository Permissions

Jetscale requires write access to your repository to create branches and pull requests. The integration will:

✓ Create feature branches ✓ Commit Terraform files ✓ Create pull requests ✓ Read repository metadata and branches

✗ Cannot merge pull requests ✗ Cannot delete branches ✗ Cannot modify settings or webhooks ✗ Cannot access repository secrets or pipeline variables

Bitbucket Pipeline Integration

Automatic Validation

Jetscale pull requests trigger your existing Bitbucket Pipelines:

# bitbucket-pipelines.yml example
pipelines:
  pull-requests:
    '**':
      - step:
          name: Validate Terraform
          script:
            - terraform init
            - terraform fmt -check
            - terraform validate
            - terraform plan

Status Checks

Jetscale monitors pipeline status:

• ✅ All checks passed: Ready to merge
• ⚠️ Checks pending: Waiting for pipeline
• ❌ Checks failed: Review pipeline logs

Managing the Integration

View Integration Status

From Settings → Integrations → Bitbucket:

• Connected Bitbucket account
• Server URL (if Bitbucket Server)
• Selected workspace/project and repository
• Last verification time
• Recent pull requests created

Change Repository

1. Navigate to Settings → Integrations → Bitbucket
2. Click Change Repository
3. Select a different repository
4. Click Update

Update Credentials

If your app password or token expires:

1. Generate a new app password/token in Bitbucket
2. Navigate to Settings → Integrations → Bitbucket
3. Click Update Credentials
4. Enter new credentials
5. Click Save and Verify

Disconnect Bitbucket

To remove the integration:

1. Navigate to Settings → Integrations → Bitbucket
2. Click Disconnect
3. Confirm disconnection

Important: Disconnecting will:

• Remove stored credentials from Jetscale
• Stop automatic PR creation
• Preserve existing pull requests in Bitbucket

Troubleshooting

Authentication Failed

Problem: "Invalid credentials" error when connecting

Solutions:

• Verify username is correct (case-sensitive)
• Check app password/token hasn't expired
• Ensure all required scopes are selected
• For Bitbucket Server, verify server URL is correct and accessible
• Try creating a new app password

Repository Not Found

Problem: Can't find repository in selection dropdown

Solutions:

• Verify you have write access to the repository
• Check repository isn't archived
• Ensure workspace/project membership is active
• For Bitbucket Server, confirm project permissions
• Try refreshing the repository list

Pull Request Creation Failed

Problem: Recommendation approved but PR wasn't created

Solutions:

• Verify branch doesn't already exist with same name
• Check you still have write access
• Ensure default branch exists and isn't locked
• Confirm repository isn't in read-only mode
• Review Bitbucket API rate limits

Pipeline Not Triggering

Problem: Bitbucket Pipelines doesn't run on Jetscale PRs

Solutions:

• Verify Pipelines are enabled for the repository
• Check bitbucket-pipelines.yml includes pull-request triggers
• Ensure pipeline quota isn't exhausted
• Review branch pattern matching in pipeline configuration

Best Practices

Credential Management

• Dedicated app password: Create separate credentials for Jetscale
• Set expiration: Use reasonable expiration dates (90-180 days)
• Rotate regularly: Update credentials before expiry
• Revoke unused: Delete old app passwords immediately

Repository Configuration

• Branch protection: Require reviews on default branch
• Required reviewers: Configure CODEOWNERS for automatic review assignment
• Pipeline validation: Enable required pipeline checks before merge
• Merge strategies: Use squash or no-fast-forward for clean history

Pull Request Workflow

1. Automated validation: Let pipelines validate Terraform changes
2. Peer review: Have infrastructure team review cost optimizations
3. Staging deployment: Test changes in non-production first
4. Monitoring: Watch metrics after applying changes
5. Documentation: Update runbooks if infrastructure changes

Bitbucket Cloud vs Server

Feature Comparison

Configuration Differences

Bitbucket Cloud:

• Uses workspace-based permissions
• App passwords have granular scopes
• Automatic HTTPS endpoints

Bitbucket Server:

• Uses project-based permissions
• Requires full server URL
• May need firewall/VPN configuration

Security Considerations

Credential Security

• Jetscale encrypts app passwords and tokens at rest (AES-256)
• Credentials never appear in logs or API responses
• All Bitbucket API calls use HTTPS with TLS 1.2+
• Credentials stored separately with restricted access

Audit Trail

All Jetscale actions are logged:

• Bitbucket API calls
• Pull request creation per recommendation
• Failed authentication attempts
• Repository access events

Audit logs available in Jetscale dashboard under Settings → Audit Log.

Revoking Access

To immediately revoke access:

Bitbucket Cloud:

1. Go to App passwords
2. Click Revoke next to Jetscale password

Bitbucket Server:

1. Navigate to Personal access tokens
2. Click Revoke next to Jetscale token

Jetscale:

1. Go to Settings → Integrations → Bitbucket
2. Click Disconnect

API Reference

For programmatic access, see our API Documentation.

Key endpoints:

• POST /api/v1/bitbucket/action/connect - Connect Bitbucket account
• POST /api/v1/bitbucket/action/select-repo - Select repository
• GET /api/v1/bitbucket/repos - List accessible repositories
• POST /api/v1/bitbucket/action/pr/create - Create pull request
• POST /api/v1/bitbucket/action/disconnect - Disconnect Bitbucket

Support

Need help with Bitbucket integration?

• Email: support@jetscale.ai
• Documentation: FAQ
• GitHub Issues: Report a problem