Deploying Workflows

Once your workflow is tested and ready, deploy it to make it available for production use. Learn how to deploy, manage versions, monitor executions, and troubleshoot issues.

Deployment Process

Step 1: Prepare for Deployment

Before deploying, ensure:

• All nodes are configured correctly
• Workflow has been tested with various inputs
• Error handling is implemented
• Data source credentials are valid
• Performance is acceptable

Step 2: Deploy Workflow

1. Click Deploy button in the workflow builder toolbar
2. Configure deployment settings
3. Review deployment summary
4. Click Deploy Workflow

Deployment Configuration

Setting	Description
Workflow Name	Unique identifier for the workflow
Description	Purpose and functionality (optional)
Environment	Development or Production
Version Number	Auto-increments by default
Tags	Labels for organization (e.g., v1.0, stable, prod)

Example Configuration:

Name: customer-onboarding
Description: Automated customer onboarding workflow with email verification
Environment: Production
Version: 1.2.0
Tags: prod, stable, customer-facing

Step 3: Post-Deployment

After successful deployment, you'll receive:

For Webhook Triggers:

Webhook URL: https://api.strongly.ai/workflows/webhook/abc123xyz
HMAC Secret: sk_live_abc123...

For REST API Triggers:

API Endpoint: https://api.strongly.ai/api/v1/process
API Key: pk_live_xyz789...
Rate Limit: 100 requests/minute

For Schedule Triggers:

Schedule: 0 9 * * 1-5 (Weekdays at 9 AM)
Next Run: 2024-01-16 09:00:00 EST
Status: Active

For Form Triggers:

Form URL: https://forms.strongly.ai/abc123
Embed Code: <iframe src="..."></iframe>

Environments

Development Environment

Purpose:

• Test new workflows
• Experiment with changes
• Debug issues

Characteristics:

• Separate from production
• Can use test credentials
• Lower performance guarantees
• More verbose logging

When to Use:

• Initial workflow development
• Testing new features
• Debugging issues
• Training team members

Production Environment

Purpose:

• Serve live traffic
• Process real data
• Mission-critical workflows

Characteristics:

• High availability
• Performance optimized
• Production credentials
• Comprehensive monitoring
• SLA guarantees

When to Use:

• Customer-facing workflows
• Business-critical automation
• High-volume processing
• Validated and tested workflows

Production Deployment

Always test workflows in Development environment before deploying to Production. Production deployments should be stable and well-tested.

Version Management

Versioning Strategy

Each deployment creates a new immutable version:

Version	Date	Author	Changes	Status
1.2.0	2024-01-15	john@example.com	Added error handling	Active
1.1.0	2024-01-10	jane@example.com	Updated AI prompt	Previous
1.0.0	2024-01-05	john@example.com	Initial deployment	Archived

Version Numbering

Follow semantic versioning:

MAJOR.MINOR.PATCH

1.2.3
│ │ │
│ │ └─ Patch: Bug fixes, minor changes
│ └─── Minor: New features, backward compatible
└───── Major: Breaking changes

Examples:

• 1.0.0 → 1.0.1: Fixed bug in error handling
• 1.0.1 → 1.1.0: Added new node for notifications
• 1.1.0 → 2.0.0: Changed API response format (breaking)

Version Operations

View Version History

1. Go to Workflows page
2. Click on workflow name
3. Click Versions tab
4. See all deployed versions

Compare Versions

1. Select two versions
2. Click Compare
3. View side-by-side diff
4. See changed nodes and configurations

Rollback to Previous Version

1. Go to Versions tab
2. Click on previous version
3. Click Rollback
4. Confirm rollback action

Version Tags

Use tags like stable, prod, or v1.0 to mark important versions for easy reference.

Monitoring Executions

Viewing Executions

1. Navigate to Workflows
2. Click on deployed workflow
3. Click Executions tab
4. See list of all workflow runs

Execution List

Execution ID	Status	Started	Duration	Trigger
exec_abc123	Success	10:30 AM	2.3s	Webhook
exec_def456	Failed	10:25 AM	1.1s	API
exec_ghi789	Success	10:20 AM	3.5s	Schedule

Filters:

• Status (Success, Failed, Running)
• Date range
• Trigger type
• Duration

Execution Details

Click any execution to view detailed trace:

Overview Tab:

• Execution ID and status
• Start and end time
• Total duration
• Trigger information
• Success/failure reason

Timeline Tab:

• Waterfall chart of node execution
• Node-by-node timing
• Parallel execution visualization
• Bottleneck identification

Logs Tab:

• Complete execution logs
• Node-level output
• Error messages
• Debug information

Data Tab:

• Input data from trigger
• Output data from each node
• Final workflow output
• Data transformations

Performance Metrics

View aggregate metrics across all executions:

Metric	Value	Trend
Total Executions	1,234	↑ 15%
Success Rate	98.5%	↑ 2%
Average Duration	2.1s	↓ 0.3s
P95 Duration	4.5s	↓ 0.8s
Error Rate	1.5%	↓ 1%
Throughput	145/hour	↑ 20%

Charts:

• Execution volume over time
• Success/failure rate trends
• Duration distribution
• Error rate by node

Managing Deployments

Start/Stop Workflows

Pause Workflow:

1. Go to workflow details
2. Click Pause button
3. Confirm pause action

Effects:

• Webhook endpoints return 503
• Scheduled runs are skipped
• API endpoints are disabled
• Form submissions are rejected

Resume Workflow:

1. Go to workflow details
2. Click Resume button
3. Workflow is active again

No Data Loss

Pausing a workflow doesn't delete any data or configuration. Resume anytime without redeployment.

Update Deployed Workflow

To update a production workflow:

1. Create New Version:

   • Make changes in workflow builder
   • Test thoroughly in Development
   • Deploy as new version

2. Zero-Downtime Deployment:

   • New version deploys alongside old
   • Traffic gradually shifts to new version
   • Old version remains available for rollback

3. Automatic Rollback:

   • High error rate triggers automatic rollback
   • Manual rollback always available
   • Previous version restored instantly

Delete Workflow

To delete a workflow:

1. Go to workflow details
2. Click Delete button
3. Confirm deletion (irreversible)

Deletion

Deleting a workflow removes all versions and execution history. This cannot be undone. Consider pausing instead.

Sharing & Permissions

Share Workflow

Grant access to team members:

1. Go to workflow details
2. Click Share button
3. Add users or teams
4. Set permission level
5. Click Share Workflow

Permission Levels:

Level	Can View	Can Edit	Can Deploy	Can Delete
Viewer	✓	✗	✗	✗
Editor	✓	✓	✗	✗
Admin	✓	✓	✓	✓
Owner	✓	✓	✓	✓

Export Workflow

Export workflow as JSON template:

1. Click Export button
2. Download JSON file
3. Share with team or save as backup

Use Cases:

• Backup workflows
• Share with other teams
• Version control in Git
• Duplicate and modify

Import Workflow

Import workflow from JSON:

1. Click Import button
2. Select JSON file
3. Review imported workflow
4. Save or deploy

Alerts and Notifications

Configure Alerts

Set up notifications for workflow events:

Alert Types:

Event	When to Alert
Execution Failed	Any execution fails
High Error Rate	Error rate > threshold
Slow Execution	Duration > threshold
Deployment	New version deployed
Rollback	Version rolled back

Notification Channels:

• Email
• Slack
• Webhook
• PagerDuty
• SMS

Example Configuration:

{
  "alertOn": "execution_failed",
  "threshold": 3,
  "window": "5m",
  "channels": ["email", "slack"],
  "recipients": ["team@example.com"],
  "slackChannel": "#alerts"
}

This sends an alert if 3 executions fail within 5 minutes.

Alert Best Practices

1. Set Appropriate Thresholds: Avoid alert fatigue
2. Use Multiple Channels: Ensure alerts are seen
3. Include Context: Provide execution details
4. Escalation: Different severity levels
5. Test Alerts: Verify notification delivery

Troubleshooting Production Issues

Common Issues

Issue: High Error Rate

Diagnosis:

• Check execution logs for error patterns
• Review recent deployments
• Verify external service status
• Check data source credentials

Solutions:

• Rollback to previous version
• Fix error handling
• Update credentials
• Add retry logic

Issue: Slow Performance

Diagnosis:

• Review execution timeline
• Identify slow nodes
• Check external API latency
• Analyze data volume

Solutions:

• Optimize slow nodes
• Add caching
• Use parallel execution
• Increase timeout limits

Issue: Webhook Not Triggering

Diagnosis:

• Verify webhook URL
• Check HMAC signature
• Review request logs
• Test with curl/Postman

Solutions:

• Update webhook URL in external system
• Verify HMAC configuration
• Check IP allowlist
• Review authentication

Issue: Rate Limiting

Diagnosis:

• Check execution frequency
• Review rate limit settings
• Monitor API usage
• Analyze request patterns

Solutions:

• Implement backoff strategy
• Batch requests
• Increase rate limits
• Use caching

Debug Production Workflows

Enable Debug Logging:

1. Go to workflow settings
2. Enable Debug Mode
3. Redeploy workflow
4. Review enhanced logs

Replay Execution:

1. Find failed execution
2. Click Replay
3. Optionally modify input
4. Execute with same data

Test in Development:

1. Export production workflow
2. Import to Development
3. Test with production data
4. Fix issues
5. Deploy to Production

Best Practices

Deployment

1. Test First: Always test in Development
2. Incremental Changes: Deploy small changes frequently
3. Version Tags: Mark stable versions
4. Documentation: Document changes in version notes
5. Rollback Plan: Know how to rollback quickly

Monitoring

1. Set Up Alerts: Configure notifications early
2. Review Metrics: Regularly check performance
3. Log Retention: Keep execution logs for debugging
4. Performance Baselines: Track normal performance
5. Capacity Planning: Monitor resource usage

Version Control

1. Semantic Versioning: Use consistent version numbers
2. Change Logs: Document what changed
3. Git Integration: Export workflows to Git
4. Backup Regularly: Export critical workflows
5. Tag Releases: Mark production versions

Security

1. Review Permissions: Limit who can deploy
2. Rotate Credentials: Update secrets regularly
3. Audit Logs: Track deployment changes
4. Access Control: Use appropriate permission levels
5. Secure Webhooks: Always use HMAC verification

Production Checklist

Before deploying to production:

Pre-Deployment

• Tested with realistic data
• Error handling implemented
• Performance is acceptable
• Security review completed
• Documentation updated

Deployment

• Deployed to Development first
• Version number updated
• Tags applied
• Change notes documented
• Team notified

Post-Deployment

• Verified workflow is active
• Tested trigger endpoints
• Alerts configured
• Monitoring dashboard created
• Rollback plan documented

Next Steps

• See workflow examples
• Learn about monitoring
• Configure data sources
• Set up user permissions

Gradual Rollout

For critical workflows, consider deploying to a small percentage of traffic first, then gradually increase after validating performance.