Skip to main content

Upgrade via Portainer

Upgrading Chatty AI through Portainer Server (vendor-managed deployment).

Overview​

For Portainer deployments, upgrades are vendor-managed and performed by the Chatty AI team through Portainer Server. Customers do not need to manually update images or configurations.

Upgrade Process​

Vendor-Managed Upgrades​

  1. Vendor initiates upgrade from Portainer Server
  2. New images pushed to customer Edge Agent
  3. Stack updated with new versions
  4. Services restarted automatically
  5. Verification performed by vendor

Customer involvement: Minimal - just monitoring and verification

Pre-Upgrade Checklist​

Before vendor performs upgrade:

  • Backup completed (vendor responsibility)
  • Maintenance window scheduled and communicated
  • Monitoring prepared to track upgrade progress
  • Support contact available during upgrade
  • Rollback plan confirmed with vendor

During Upgrade​

What Happens​

  1. Notification received from vendor about upcoming upgrade
  2. Maintenance window begins (typically 30-60 minutes)
  3. Services briefly unavailable during container restart
  4. New versions deployed automatically
  5. Services come back online with new version

Customer Monitoring​

Monitor the upgrade through Portainer UI:

# Access Portainer Edge Agent UI (if available)
https://your-portainer-edge-agent:9443

# Or check Docker directly
docker ps
docker compose ps

Watch for:

  • Container restarts
  • New image versions
  • Service health status
  • Log messages

Post-Upgrade Verification​

1. Check Service Status​

# Check all containers are running
docker ps

# Check service health
docker compose ps

# View recent logs
docker compose logs --tail=100

2. Verify Web Access​

Test each service:

# Chatty AI
https://YOUR_CHATTYAI_DOMAIN

# n8n
https://YOUR_N8N_DOMAIN

# Databases
https://YOUR_DATABASES_DOMAIN

3. Test Core Functionality​

  • Login with admin credentials
  • Create new chat conversation
  • Upload document and test RAG
  • Test n8n workflow execution
  • Test Databases query
  • Check integrations (MS365, Jira)

4. Check Version Numbers​

Verify new versions are running:

# Check Chatty AI version
docker inspect chatty-app-chattyai-1 | grep -i version

# Check all image versions
docker images | grep chattyai-org

5. Review Logs​

Check for errors:

# All services
docker compose logs --tail=200

# Specific service
docker compose logs chattyai --tail=100

# Follow logs in real-time
docker compose logs -f

Troubleshooting​

Services Not Starting​

# Check container status
docker ps -a

# View logs for failed service
docker compose logs <service-name>

# Restart specific service
docker compose restart <service-name>

# Contact vendor if issues persist

Version Mismatch​

# Check current versions
docker images | grep chattyai-org

# Verify environment variables
docker compose config

# Contact vendor to verify correct images deployed

Data Issues After Upgrade​

# Check database connectivity
docker compose exec db psql -U chattyAdmin -d chattydb -c "SELECT version();"

# Check Qdrant
curl http://localhost:6333/collections

# Contact vendor for data verification

Performance Degradation​

# Check resource usage
docker stats

# Check disk space
df -h

# Check logs for errors
docker compose logs | grep -i error

# Contact vendor for performance analysis

Rollback Procedure​

If issues occur after upgrade:

1. Contact Vendor Immediately​

Provide:

  • Description of issue
  • Error messages from logs
  • Service status output
  • Timeline of when issue started

2. Vendor Initiates Rollback​

Vendor will:

  1. Stop current deployment
  2. Deploy previous version images
  3. Restore configuration if needed
  4. Verify services are working
  5. Confirm with customer

3. Customer Verification​

After rollback:

  • Services accessible
  • Login working
  • Data intact
  • Functionality restored

Communication During Upgrade​

Before Upgrade​

Vendor provides:

  • Upgrade schedule (date/time)
  • Expected duration (typically 30-60 minutes)
  • New features being deployed
  • Known issues or breaking changes
  • Rollback plan if needed

During Upgrade​

Vendor provides:

  • Progress updates at key milestones
  • Issue notifications if problems occur
  • Estimated completion time updates

After Upgrade​

Vendor provides:

  • Completion notification
  • Version confirmation
  • Release notes for new version
  • Known issues (if any)
  • Support contact for questions

Upgrade Schedule​

Typical Upgrade Frequency​

  • Minor updates: Monthly (bug fixes, small features)
  • Major updates: Quarterly (new features, major changes)
  • Security patches: As needed (critical fixes)
  • Emergency fixes: Immediate (critical bugs)

Maintenance Windows​

Typical schedule:

  • Day: Sunday or off-peak hours
  • Time: 2:00 AM - 4:00 AM (customer timezone)
  • Duration: 30-60 minutes
  • Notification: 7 days advance notice

Customer Responsibilities​

Before Upgrade​

  • Review upgrade notification from vendor
  • Confirm maintenance window acceptable
  • Notify end users of planned downtime
  • Ensure monitoring is active
  • Have support contact ready

During Upgrade​

  • Monitor service status
  • Do not make configuration changes
  • Do not restart services manually
  • Report any issues to vendor immediately

After Upgrade​

  • Verify services are accessible
  • Test core functionality
  • Review release notes
  • Report any issues to vendor
  • Notify end users services are restored

Vendor Responsibilities​

Before Upgrade​

  • Create full backup
  • Test upgrade in staging environment
  • Prepare rollback plan
  • Notify customer 7 days in advance
  • Confirm maintenance window with customer

During Upgrade​

  • Execute upgrade procedure
  • Monitor service health
  • Provide progress updates
  • Handle any issues that arise
  • Verify successful deployment

After Upgrade​

  • Verify all services running
  • Confirm with customer
  • Provide release notes
  • Monitor for 24 hours post-upgrade
  • Be available for support

Emergency Upgrades​

For critical security patches:

  1. Vendor notifies customer of critical issue
  2. Shorter notice period (24-48 hours)
  3. Expedited upgrade process
  4. Immediate verification required
  5. Extended monitoring post-upgrade

Upgrade Checklist​

Pre-Upgrade​

  • Upgrade notification received from vendor
  • Maintenance window confirmed
  • End users notified
  • Monitoring prepared
  • Support contact available

During Upgrade​

  • Maintenance window started
  • Services monitored
  • No manual interventions performed
  • Issues reported to vendor (if any)

Post-Upgrade​

  • Services accessible
  • Login tested
  • Core functionality verified
  • Version numbers confirmed
  • Logs reviewed
  • Issues reported to vendor (if any)
  • End users notified