Exam AWS Certified Solutions Architect - Professional topic 1 question 462 discussion

A B\D: Too costly and not pilot light. C: ECS should not be used.

I would go with "A". B: it is not pilot, as it has working nodes in DR region. C: comparing Lambda to ECS snapshot job, it is better to use Lambda. D: EBS does not have auto regional replication!!

A is the right answer D-> I think that active-actice will not reduce costs and it's not pilot light.

A: write unnecessary lambda seems odd and incurs costs B: again a lamda C: long jobs on ec2 even more expensive D: cross region snapshots are possible and avoid costs for lambda/ec2. I do not like the active active as it bears the unnecessary risk of two systems beeing up at the same time and datacoruption... . but it does have the benefit that with autscaling at 0 the system is down and will immediately be used as soon as it comes up. Pull the primary down and pull the other up and dns will deliver the ip of the sendary. Difference to active passiv in dns is only in case both systems are up at the same time in which case both systems will be used for name resolution. So the whole setup is active-pasive, only dns is active-active which is ok as it will use only the up of the active system.. So cheapest is indeed D. (May bear a risk of datacorruption in case of misconfigured autoscaling and both systems beeing up at the same time. But correctly configured it works)

A is best answer due to pilot and cost.

It's actually D. R53 Active-Active with 0 instances on the DR region will result in a pilot-light configuration. A B and C have "daily" snapshot copy that won't meet the RPO/RTO requirements. D does not specify how often the snapshots are created/copied on the DR region, and the minimum interval is 1 hour, which will meet RPO and RTO requirements.

trick of the question is "pilot light", which makes A better than D

A. A is correct , we need active-passive failover configuration in Route 53.

Based on all comments

i think it is easy to implement EBS cross-region rather than lambda, actually today i use AWS backup service but "'D" answer has active-active which is not the pilot light strategy as active-active is the strategy of multi-site https://aws.amazon.com/blogs/architecture/disaster-recovery-dr-architecture-on-aws-part-i-strategies-for-recovery-in-the-cloud/ So, Answer A is best choice in this case

As it's pilot light approach on backup region, anything which has active-active doesn't make sense

Only A is correct. How D can be correct? its active-active are you going to shift traffic to an instance which does not exists? also in the question states "pilot light" this is clearly a active-passive deployment. another point is that you cannot just cross-region copy snapshots by itself you need to have a lambda or a script that will do it.

Why i think A is wrong...The company's RTO is six hours and its RPO is twenty-four hours. A talks about a daily backup which means once after 24 hrs. What if the system fails at 23rd hour. Will you meet RTO of 6 hours? I prefer D although its Active-Active the autoscaling is set to 0.

D is perfect, if it's not "active-active failover". So it's A

EBS & EFS don't have cross-region replication in-built; you can use AWS Backup for that.

A is correct. Pilot light.

A. Use AWS Lambda to create daily EBS and RDS snapshots, and copy them to the disaster recovery region. Use Amazon Route 53 with active-passive failover configuration. Use Amazon EC2 in an Auto Scaling group with the capacity set to 0 in the disaster recovery region.