Lesson 13: Implementing High Availability and Disaster Recovery in the Cloud Flashcards
An organization has a cloud deployment in an availability zone with redundant server hardware, redundant power supplies, and redundant network connectivity. What is the next step that represents the best solution for the organization to mitigate a single point of failure?
A.Replicate the cloud deployment within the zone.
B.Replicate the cloud deployment on-premises.
C.Replicate the cloud deployment to another zone.
D.Replicate the cloud deployment to another region.
C.Replicate the cloud deployment to another zone.
The next step the organization should take is to configure the cloud deployment to replicate to another availability zone, within the same region, to eliminate the availability zone as the single point of failure.
A database administrator works for a critical financial institution which requires the client machines to be informed after multiple servers have a copy of the transaction. What type of replication should the administrator set up?
A.Site-to-site
B.Snapshots
C.Synchronous
D.Asynchronous
C.Synchronous
With synchronous replication, the client machine is not informed that the transaction was successful until both servers have a copy of the data.
A small marketing firm is preparing a business continuity plan and needs to select the right kind of disaster recovery site. The firm’s business relies on created and approved marketing campaign files to run daily campaigns and internet connectivity to run the business on the front end. The firm also uses customer relationship management software, as well as the website, for sales. The firm is in a tornado-prone area, but the firm’s customers reside all across the country. One day of being unable to post a customer’s campaign can cause the firm to lose business. What is the best disaster recovery solution for the marketing firm?
A.Hot site
B.Warm site
C.Cold site
D.Cloud site
D.Cloud site
The marketing firm needs the ability to move operations without any downtime. Since it is a small organization, a cloud solution is the best disaster recovery solution.
Which of the following represents a single point of failure? (Select all that apply.)
A.Cloud direct connection
B.Cluster
C.NIC teaming
D.Load balancer
A.Cloud direct connection
D.Load balancer
A cloud direct connection is a dedicated connection between a cloud service provider and an on-premises network. Having only one cloud direct connection represents a single point of failure.
A load balancer distributes work among servers and represents a single point failure when it becomes unavailable and traffic cannot reach any of the servers that the load balancers service.
An organization has learned that it may be competing with other organizations for access to the cloud service provider’s (CSPs) resources. Which of the following CSP resources might the organization’s be competing for? (Select all that apply.)
A.Storage
B.CPU and Memory
C.Bandwidth
D.VMs
A.Storage
B.CPU and Memory
C.Bandwidth
The organizations may be competing for storage as CSPs may oversell storage capacity. Storage is easy for CSPs to add but competing for the overall read/write speed for storage may be an issue.
The organizations may be competing for CPU and memory, or compute resources, which may be overallocated to VMs.
The organizations may be competing for network bandwidth with CSPs offering more network bandwidth than is actually available. However, virtual private networks do not allow overallocation.
An organization is testing the cloud disaster recovery plan. The goal is to implement the recovery in a non-production environment. What kind of test is the organization running?
A.Parallel recovery
B.Simulated failover
C.Walk-through
D.Live failover
A.Parallel recovery
The organization is running a parallel recovery test by implementing the disaster recovery procedures in a non-production environment.
An organization has concerns about single points of failure for their VMs. What can the cloud administrator use to ensure all the VMs do not fail at the same time?
A.File server replication
B.Anti-affinity
C.Affinity
D.Right-sizing
B.Anti-affinity
The cloud administrator can use anti-affinity for the VMs. Anti-affinity purposely spreads VM instances across multiple hosts to avoid a single point of failure.
What kind of documentation should an organization have on hand to respond to an incident in the cloud? (Select all that apply.)
A.DR playbook
B.Call tree
C.Building diagram
D.Network flow diagram
A.DR playbook
B.Call tree
D.Network flow diagram
The organization should have the DR (Disaster Recovery) playbook on hand. The DR playbook will provide specific, actionable procedures to address the incident.
The organization should have the call tree on hand. A call tree is a notification system that outlines whom employees should call next to notify of an incident.
The organization should have the network flow diagram on hand. This can help administrators visualize how data moves through the network to respond to and contain the incident.
An organization’s cloud service provider (CSP) recently experienced an outage, which caused the organization’s cloud instances to be down for 30 minutes. Where can the organization look to see if this is longer than the CSP’s guaranteed percent availability?
A.BIA
B.RTO
C.SLA
D.MTTR
C.SLA
The organization can check the SLA (Service Level Agreement). The SLA sets the service requirements and expectations and outlines the guaranteed percent availability of hosted services.
An organization is responding to a data breach in the cloud network. The organization has completely restored normal operations to the production environment and has isolated the breach on a specific segment of the network to further investigate and collect evidence. What negative action will eliminate the organization’s chances of taking legal action against the threat actors?
A.Fail to examine and collect the cloud logs.
B.Fail to take snapshots of the breached systems.
C.Fail to use the CSP’s tools to collect evidence.
D.Fail to track the chain of custody
D.Fail to track the chain of custody
If the investigators fail to properly track the chain of custody of all evidence collected during the breach investigation, it can eliminate the organization’s chances of prosecuting the threat actors.
A hybrid cloud environment hosts web services on both on-premises and the cloud. A similar network is set up in an isolated environment for testing disaster recovery scenarios. If IT management wants to test their disaster recovery procedures with the participation of a limited amount of real user accounts, what type of test would be most appropriate to carry out?
A.Paper test
B.Parallel recovery test
C.Table-top
D.Walk-through
B.Parallel recovery test
A parallel recovery test using a non-production test environment involves running a subset of systems in the disaster recovery environment. It would allow for testing with a limited amount of real user accounts, making it the most appropriate type of test for this scenario.
In which instance would an organization NOT employ the disaster recovery playbook?
A.Hurricane
B.Cloud administrator quits
C.Active shooter
D.Data breach
B.Cloud administrator quits
When a cloud administrator quits it can be disastrous for the organization. However, this is not a situation in which an organization would use the disaster recovery playbook.
A solutions architect has performed a risk analysis for disaster recovery and has determined that the likelihood of needing to recover is very low. They also determined that needing some time to bring another center up is acceptable, but not too much time. Which of the following backup solutions would meet the company’s needs? (Select all that apply.)
A.Hot site
B.Warm site
C.Cold site
D.Cloud site
B.Warm site
D.Cloud site
For warm sites to take over from the primary, some equipment needs to be brought in, and some configuration needs to occur. Data has to be migrated to the warm site. This makes the warm site less expensive upfront than a hot site but also requires more time to get the business up and running again if there’s a disaster.
Many organizations now look to the cloud as a replacement for traditional hot, warm, or cold sites.
An organization had to evacuate their primary location due to approaching wildfires. As outlined in the business continuity plan, the organization moved all business operations to another location that they had previously set up with most of the equipment they needed. The organization acquired the rest of the needed equipment, restored data backups to the new location, and resumed operations within a day. What kind of disaster recovery site did the organization use?
A.Hot site
B.Warm site
C.Cold site
D.Cloud site
B.Warm site
The organization used a warm site to recover operations in the face of the wildfires. A warm site is set up with the space and most of the needed equipment to resume operations.
A cloud administrator is configuring a load balancer to manage traffic to an organization’s web servers that house a catalog of products for customers to buy online. What load balancer scheduling type should the cloud administrator configure to ensure that all customers can place orders at all times?
A.Software load balancer
B.Weighted scheduling
C.Most recently used
D.Round-robin
C.Most recently used
The administrator should configure the most recently used scheduling type. This will ensure that if there are not enough connections on the existing machines, it will create new VMs to accommodate all customer orders.
