Business Continuity and Disaster Recovery at CrowdHandler

CrowdHandler has robust business continuity and disaster recovery planning in place to minimize disruptions in the event of operational issues, infrastructure failures, or regional outages. Our policies include:

Redundant Infrastructure

Our infrastructure utilizes:

  • Multiple AWS regions - We maintain a replica of production environments and data in a secondary AWS region (US-East-2) in case the primary region (US-East-1) experiences a significant outage.
  • Multi-AZ databases - Our RDS database instances are deployed across multiple availability zones within a region to allow for automatic failover if a single AZ is disrupted.
  • Elastic infrastructure - We use auto-scaling groups where possible so compute resources scale based on demand and maintain availability even if individual components require replacement.

Disaster Recovery Testing

We conduct disaster recovery exercises twice annually to validate recovery procedures and minimize downtime in a real disaster scenario:

  1. Simulate a disaster in our primary AWS region (US-East-1) that renders environments and data inaccessible.
  2. Promote replica databases in US-East-2 to become primary production databases.
  3. Update DNS to route all traffic to US-East-2 and update application settings to point to the new databases.
  4. Conduct failover testing to confirm business-critical systems remain available, workloads are balancing and performance is acceptable during to the simulated outage.
  5. Remediate any identified issues or risks to improve the viability of disaster recovery plans.
  6. Test restoration of any altered configurations or settings back to normal operations.

Continuous Monitoring

We employ 24/7 on-call rotations and monitor systems using:

  • CloudWatch dashboards - We track key performance metrics and thresholds for early outage detection across services.
  • CloudWatch alarms - We configure alarms to alert on any deviations from expected values so issues can be promptly investigated.
  • On-call schedules - Staff members carry a 24-hour on-call rotation in case any CloudWatch alerts are triggered out-of-hours. On-call staff have runbooks for common issues ready to mitigate delays.

CrowdHandler's business continuity and disaster recovery planning provide a framework to prepare for and safeguard against system disruptions. Policies aim to minimize downtime by utilizing redundant infrastructure, failover capabilities and restoration scripts supported by a defined plan outlining steps to recover from disaster scenarios. Frequent testing of plans helps guarantee we are able to react swiftly in an actual emergency to avoid data loss or prolonged unavailability. Continuous monitoring acts as an early warning system so we can respond rapidly.