I85PREP
⌘ K
Alibaba ACA Cloud Operator study materials
Alibaba ACA Cloud Operator study resources

Alibaba ACA Cloud Operator practice questions

One question at a time. Build your Alibaba ACA Cloud Operator understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Cloud Operations Fundamentals

Which Alibaba Cloud service provides a centralized dashboard for monitoring metrics, setting alarms, and viewing resource health across many cloud products?

  1. ActionTrail
  2. Alibaba Cloud CDN
  3. Resource Access Management
  4. CloudMonitor Correct answer.

Explanation

CloudMonitor collects metrics and event data from Alibaba Cloud resources and lets operators build dashboards and alarm rules. It is the primary observability service for operational health tracking.

Question 2Cloud Operations Fundamentals

An operator wants an automated notification when CPU utilization of an ECS instance stays above 80% for five minutes. Which CloudMonitor feature should be configured?

  1. A RAM permission policy
  2. A resource group tag policy
  3. An SLB health check
  4. An alarm rule with a metric threshold and evaluation period Correct answer.

Explanation

Alarm rules in CloudMonitor trigger notifications when a metric crosses a threshold across a defined evaluation period. Operators set the metric, comparison, threshold, and notification targets.

Question 3Cloud Operations Fundamentals

What is the primary purpose of Alibaba Cloud ActionTrail?

  1. To replicate data between regions for disaster recovery
  2. To record API operations on cloud resources for auditing and compliance Correct answer.
  3. To distribute traffic across multiple backend servers
  4. To scan workloads for malware

Explanation

ActionTrail captures API calls made in an account, including the caller identity, time, and source IP, supporting audit and compliance requirements. Its records are used for investigation and governance.

Question 4Cloud Operations Fundamentals

Which service should an operator use to create and manage user identities and grant fine-grained permissions to Alibaba Cloud resources?

  1. Alibaba Cloud DNS
  2. Resource Access Management (RAM) Correct answer.
  3. CloudMonitor
  4. Virtual Private Cloud (VPC)

Explanation

RAM manages users, groups, and roles and controls which actions they may perform on which resources through policies. It is the core identity and access service.

Question 5Cloud Operations Fundamentals

An operator needs to isolate a set of ECS instances in a private network with its own IP address range and route tables. Which service provides this?

  1. Alibaba Cloud CDN
  2. Object Storage Service (OSS)
  3. Server Load Balancer (SLB)
  4. Virtual Private Cloud (VPC) Correct answer.

Explanation

A VPC is a logically isolated network in which operators define CIDR blocks, subnets, and route tables. Resources launched inside the VPC communicate over that private network.

Question 6Cloud Operations Fundamentals

Which component of Server Load Balancer determines whether a backend ECS instance is healthy enough to receive traffic?

  1. The billing method
  2. The instance RAM role
  3. The security group rule priority
  4. The health check configuration Correct answer.

Explanation

SLB health checks periodically probe backend servers, and only servers that pass the check are included in traffic distribution. Operators tune protocol, port, interval, and thresholds in the health check settings.

Question 7Cloud Operations Fundamentals

What does the shared responsibility model imply for an operator running ECS instances?

  1. Alibaba Cloud is responsible for all security including the customer's application code
  2. The customer is responsible for physical datacenter security
  3. Security responsibilities are entirely determined by the chosen billing model
  4. Alibaba Cloud secures the underlying infrastructure while the customer secures the guest OS and application Correct answer.

Explanation

Under the shared responsibility model, the provider secures facilities, hardware, and virtualization, while customers secure their guest operating systems, configurations, and data. Operators must patch and harden their own instances.

Question 8Cloud Operations Fundamentals

Which practice best supports cost visibility for operations teams running many ECS instances?

  1. Assigning a single RAM user to everyone
  2. Using only pay-as-you-go billing for all workloads
  3. Disabling CloudMonitor to reduce overhead
  4. Tagging resources with cost-center and environment labels Correct answer.

Explanation

Consistent tagging lets operators attribute spend to teams, environments, or projects and analyze bills accordingly. It also supports filtering and grouping in management tools.

Question 9Cloud Operations Fundamentals

An operator observes that an ECS instance is unreachable over SSH while the application on port 80 still responds. Which factor is the most likely cause?

  1. A DNS record pointing to the wrong host
  2. A security group rule that permits port 80 but not port 22 Correct answer.
  3. An expired SSL certificate on the web server
  4. A CloudMonitor alarm in the OK state

Explanation

Security groups act as stateful virtual firewalls controlling inbound traffic per port and protocol. If port 80 is allowed but port 22 is not, SSH fails while HTTP succeeds.

Question 10Cloud Operations Fundamentals

What is the effect of associating an elastic IP address with an ECS instance in a VPC?

  1. It automatically creates a load balancer in front of the instance
  2. It provides a public IP that can be remapped to another instance without changing the address Correct answer.
  3. It encrypts all traffic leaving the instance
  4. It converts the instance's private IP into a public subnet address

Explanation

An elastic IP is a public address that can be detached from one resource and attached to another, supporting failover and maintenance. The address itself persists across remapping.

Question 11Cloud Operations Fundamentals

An operator must ensure that a production database is not accidentally deleted by a junior team member. Which RAM-based control is most appropriate?

  1. Grant full administrator permissions but rely on training
  2. Grant read-only permissions to the database resource for that user Correct answer.
  3. Disable ActionTrail to avoid noise
  4. Share the root account credentials with the team

Explanation

Least privilege means granting only the permissions needed to perform the job. Read-only access prevents destructive actions such as deletion while still allowing observation.

Question 12Cloud Operations Fundamentals

A web tier behind SLB shows intermittent 502 errors. Which operational check is most directly relevant?

  1. Whether the SLB billing tier supports the region
  2. Whether the backend application process is listening and responding on the health check port Correct answer.
  3. Whether the ECS instance has a public IP
  4. Whether the domain's WHOIS record is current

Explanation

502 responses from a load balancer typically indicate backends that fail to return valid responses. Verifying the application listener and health check port is the direct diagnostic step.

Question 13Cloud Operations Fundamentals

Which Alibaba Cloud capability helps an operator recover a resource configuration that was changed incorrectly by comparing historical API operations?

  1. Inspecting CDN cache hit ratios
  2. Checking the SLB certificate list
  3. Reviewing ActionTrail logs for the relevant API calls Correct answer.
  4. Reviewing DNS TTL values

Explanation

ActionTrail records the API operations, identities, and timestamps of configuration changes, enabling operators to identify what changed and by whom. This supports root-cause analysis and rollback planning.

Question 14Cloud Operations Fundamentals

An operator wants to be alerted when any ECS instance in a VPC loses network connectivity to a critical database. Which monitoring approach is most suitable?

  1. Enable only host-level CPU alarms
  2. Rely on monthly billing reports
  3. Disable health checks to reduce noise
  4. Configure a site or custom monitoring task that probes the database endpoint and alarm on failure Correct answer.

Explanation

Connectivity to a specific endpoint is best tested with active probes, such as site monitoring or custom monitoring tasks, that alarm when the probe fails. Host CPU alarms do not detect network reachability problems.

Question 15Cloud Operations Fundamentals

What is the operational benefit of placing ECS instances across multiple zones within a region?

  1. It eliminates the need for security groups
  2. It reduces the impact of a single-zone failure on the application Correct answer.
  3. It removes the need for backups
  4. It guarantees lower storage pricing

Explanation

Zones are isolated fault domains within a region, so distributing instances across zones improves availability if one zone experiences a failure. Applications should also be designed to tolerate zone loss.

Question 16Cloud Operations Fundamentals

An operator must grant a third-party auditor temporary access to review resources without creating a permanent user. Which RAM feature fits best?

  1. A RAM role assumed by the auditor with a time-limited policy Correct answer.
  2. Sharing the account's access key secret
  3. Disabling multi-factor authentication for convenience
  4. A permanent RAM user with administrator rights

Explanation

RAM roles allow temporary credentials to be issued to trusted entities, and access can be scoped and time-limited. This avoids long-lived credentials for external parties.

Question 17Cloud Operations Fundamentals

A CloudMonitor alarm fires repeatedly for a metric that briefly spikes but recovers quickly. Which configuration change reduces false alarms?

  1. Remove the notification contact group
  2. Lower the alarm threshold
  3. Increase the number of consecutive evaluation periods required to trigger Correct answer.
  4. Switch the metric to a longer aggregation interval only

Explanation

Requiring the threshold to be breached for several consecutive periods filters out brief spikes. This is the standard approach to reducing alarm noise while keeping detection for sustained issues.

Question 18Cloud Operations Fundamentals

Which statement best describes the role of a resource group in cloud operations?

  1. It replaces CloudMonitor for metric collection
  2. It logically groups resources so that permissions, tags, and billing can be managed together Correct answer.
  3. It replicates data across regions automatically
  4. It provides network isolation equivalent to a VPC

Explanation

Resource groups let operators organize resources by project, team, or environment and apply management and authorization boundaries at that scope. They are an organizational construct, not a network or replication mechanism.

Question 19Cloud Operations Fundamentals

An operator needs to confirm that a scheduled scaling action actually added ECS instances during a traffic spike. Which evidence is most direct?

  1. The SLB certificate expiration date
  2. Scaling activity records showing the triggered rule and instance count changes Correct answer.
  3. The monthly invoice total
  4. The DNS zone file

Explanation

Auto Scaling keeps activity records that show which scaling rule fired, when, and how the instance count changed. These records directly confirm whether the action executed as intended.

Question 20Cloud Operations Fundamentals

Which operational control most directly limits the blast radius of a compromised access key?

  1. Using the same key across all environments
  2. Storing the key in a public repository for easy retrieval
  3. Rotating the key once per year
  4. Scoping the key's RAM policy to the minimum required actions and resources Correct answer.

Explanation

Least-privilege policies restrict what a compromised credential can do, reducing the potential impact. Rotation and secure storage are complementary but scoping is the direct blast-radius control.

Question 21Cloud Operations Fundamentals

An operator is designing an alarm strategy for a multi-tier application. Which approach best balances detection speed with actionable alerting?

  1. Alert only on billing thresholds
  2. Disable alarms during business hours to reduce noise
  3. Alert on service-level symptoms at each tier and route to the owning team with runbook links Correct answer.
  4. Alert on every host metric with a single shared distribution list

Explanation

Symptom-based alarms tied to service health and routed to the responsible team produce actionable alerts. Including runbook links shortens response time and reduces escalation confusion.

Question 22Cloud Operations Fundamentals

A team needs to prove to auditors that no one modified a production security group during a change freeze. Which combination of evidence is most appropriate?

  1. CloudMonitor CPU graphs for the same period
  2. SLB access logs for backend latency
  3. OSS bucket storage class reports
  4. ActionTrail records of security group API calls filtered to the freeze window Correct answer.

Explanation

ActionTrail captures the API operations that alter resources, including identity and time, which directly demonstrates whether changes occurred during the freeze. Other telemetry does not record configuration changes.

Question 23Cloud Operations Fundamentals

An operator must reduce recovery time after a regional outage for a stateful application. Which design choice contributes most directly?

  1. Increasing the instance count in the primary region only
  2. Adding more RAM users for the operations team
  3. Enabling verbose logging on all instances
  4. Replicating data and maintaining standby capacity in a second region with tested failover procedures Correct answer.

Explanation

Cross-region replication plus pre-provisioned standby capacity and rehearsed failover reduces the time to restore service after a regional event. Testing the procedure is essential because untested failover often lengthens recovery.

Question 24Cloud Operations Fundamentals

During an incident, an operator finds that a recent deployment changed a security group to allow all inbound traffic. Which immediate remediation best restores the intended posture while preserving evidence?

  1. Delete the security group entirely and rebuild the instance
  2. Revoke the overly broad rule and retain ActionTrail records of the change Correct answer.
  3. Ignore the rule until the next quarterly review
  4. Disable ActionTrail to stop further logging

Explanation

Removing the permissive rule restores the intended access control, while retaining audit records preserves evidence for investigation. Rebuilding or suppressing logs would lose information or cause unnecessary disruption.

Question 25Cloud Operations Fundamentals

Which metric combination best indicates that an application tier is saturated rather than merely busy?

  1. Sustained high CPU with rising request latency and queue depth Correct answer.
  2. High network throughput with low error rates
  3. Low CPU with stable latency
  4. Increasing storage capacity with flat IOPS

Explanation

Saturation appears when resource demand exceeds capacity, shown by sustained high utilization together with growing latency and queue depth. Utilization alone can be high in a healthy, well-provisioned system.

Question 26Cloud Operations Fundamentals

Which Alibaba Cloud service is the primary centralized monitoring service used to collect metrics and configure alarms for ECS, RDS, and SLB resources?

  1. ActionTrail
  2. Resource Access Management
  3. Alibaba Cloud DNS
  4. CloudMonitor Correct answer.

Explanation

CloudMonitor is the platform service that collects resource and application metrics and evaluates alarm rules against them. The other services handle auditing, identity and permissions, and DNS resolution respectively.

Question 27Cloud Operations Fundamentals

Which Alibaba Cloud service records API calls made in an account so an operator can review who performed an action and when?

  1. Server Load Balancer
  2. ActionTrail Correct answer.
  3. CloudMonitor
  4. Virtual Private Cloud

Explanation

ActionTrail captures API activity in the account and retains event records for auditing and investigation. Metric, traffic distribution, and network isolation services do not provide an API audit trail.

Question 28Cloud Operations Fundamentals

What is the primary purpose of an Alibaba Cloud region?

  1. To group resources by billing account for cost allocation
  2. To cache content closer to end users
  3. To define a geographic area containing one or more isolated zones for deploying resources Correct answer.
  4. To provide a private network boundary for a single department

Explanation

A region is a geographic deployment area that contains multiple physically separated zones. Zones within a region have independent power and network infrastructure, which enables fault isolation and low-latency replication.

Question 29Cloud Operations Fundamentals

Which statement describes the function of a VPC in Alibaba Cloud operations?

  1. It provides an isolated virtual network in which the operator defines subnets, routes, and security controls Correct answer.
  2. It provides a managed relational database engine with automatic backups
  3. It stores object data durably across zones
  4. It distributes inbound traffic across multiple backend servers

Explanation

A VPC is a logically isolated network where the operator controls address ranges, subnets, route tables, and security groups. Databases, load balancing, and object storage are separate services that may be deployed inside a VPC.

Question 30Cloud Operations Fundamentals

Which security control acts as a stateful virtual firewall applied at the instance level in Alibaba Cloud?

  1. NAT gateway
  2. Security group Correct answer.
  3. Route table
  4. Access control list

Explanation

Security groups are stateful virtual firewalls attached to instances that permit or deny traffic by rule. Route tables and NAT gateways handle traffic paths, while access control lists are stateless subnet-level filters.

Question 31Cloud Operations Fundamentals

Which Alibaba Cloud service is designed to distribute incoming traffic across multiple backend ECS instances?

  1. ActionTrail
  2. CloudMonitor
  3. Server Load Balancer Correct answer.
  4. Resource Orchestration Service

Explanation

Server Load Balancer accepts client traffic on a frontend listener and forwards it to a pool of healthy backend servers. It improves availability and throughput by spreading requests across instances.

Question 32Cloud Operations Fundamentals

What is the purpose of an Auto Scaling scaling group?

  1. To record configuration changes for rollback
  2. To cache static content at edge locations
  3. To maintain a defined number of ECS instances and adjust that number according to rules or schedules Correct answer.
  4. To replicate a database to a standby zone automatically

Explanation

A scaling group manages a set of ECS instances and can add or remove them based on scaling rules, scheduled tasks, or alarms. It keeps capacity aligned with demand while respecting minimum and maximum bounds.

Question 33Cloud Operations Fundamentals

Which Alibaba Cloud service lets an operator define infrastructure as code using templates to create and update resources consistently?

  1. Resource Orchestration Service Correct answer.
  2. Alibaba Cloud DNS
  3. ActionTrail
  4. CloudMonitor

Explanation

Resource Orchestration Service uses templates to describe resources so that environments can be created and updated reproducibly. This reduces configuration drift compared with manual console changes.

Question 34Cloud Operations Fundamentals

An operator wants to check whether CPU utilization on a production ECS instance has exceeded 80 percent during the past week. Which approach is most appropriate?

  1. Check the SLB listener configuration for the instance
  2. Inspect the instance's security group rules for changes
  3. Review the historical CPU metric for the instance in CloudMonitor over the past week Correct answer.
  4. Read the ActionTrail events for the instance over the past week

Explanation

CloudMonitor retains time-series metrics, so the operator can query historical CPU data for the chosen interval. Security group rules, API audit events, and load balancer listeners do not contain CPU utilization history.

Question 35Cloud Operations Fundamentals

An operator needs to confirm that a scaling group respected its maximum instance limit during a burst of demand. Which information should be examined?

  1. The DNS record TTL for the application domain
  2. The security group rules attached to the scaling group
  3. The VPC route table entries for the affected subnets
  4. The scaling group's maximum size setting together with its scaling activity history Correct answer.

Explanation

Scaling activity records show each scaling action and its outcome, and the maximum size setting bounds how many instances the group may hold. Comparing the two confirms whether the limit was reached and prevented further scale-out.

Question 36Cloud Operations Fundamentals

An operator must give a new engineer read-only visibility into CloudMonitor alarms without allowing configuration changes. Which approach follows least privilege?

  1. Add the engineer to a group with full write access to all services
  2. Attach a RAM policy granting only the read-level CloudMonitor actions required Correct answer.
  3. Share the account's primary access key with the engineer
  4. Attach the AdministratorAccess policy to the engineer's RAM user

Explanation

Least privilege means granting only the specific read actions needed for the task. Broad administrator policies, shared credentials, and full write access all exceed the requirement and increase risk.

Question 37Cloud Operations Fundamentals

An operator wants to be notified when the number of unhealthy backend servers behind a load balancer rises above a threshold. Which configuration is most direct?

  1. A scheduled scaling task that adds instances every hour
  2. A CloudMonitor alarm on the load balancer's unhealthy backend count metric with a notification contact group Correct answer.
  3. An ActionTrail trail that delivers events to an OSS bucket
  4. A DNS failover record that points to a standby address

Explanation

CloudMonitor exposes load balancer health metrics, and an alarm on the unhealthy backend count with a contact group delivers the required notification. The other options either record events, change capacity, or alter resolution without alerting on health.

Question 38Cloud Operations Fundamentals

An operator needs to identify which RAM user deleted a production security group rule last week. Which service provides this evidence?

  1. CloudMonitor
  2. Object Storage Service
  3. ActionTrail Correct answer.
  4. Server Load Balancer

Explanation

ActionTrail records API calls with the identity of the caller and the time of the request, which supports attribution of configuration changes. Metric, load balancing, and storage services do not provide an API audit trail.

Question 39Cloud Operations Fundamentals

An operator wants a repeatable way to recreate an identical test environment from a known specification. Which practice best supports this?

  1. Exporting CloudMonitor metric data and replaying it in the test account
  2. Copying screenshots of the console settings into a shared document
  3. Defining the environment in a Resource Orchestration Service template and deploying it on demand Correct answer.
  4. Manually rebuilding the environment in the console each time it is needed

Explanation

Infrastructure as code captures the environment definition in a template that can be deployed repeatedly and consistently. Manual rebuilds and documentation introduce drift and human error.

Question 40Cloud Operations Fundamentals

An operator observes that an ECS instance in a VPC cannot reach the internet, although its security group allows outbound traffic. Which configuration should be checked first?

  1. Whether the instance has a descriptive name tag
  2. Whether ActionTrail is enabled in the account
  3. Whether CloudMonitor alarms are enabled for the instance
  4. Whether the subnet's route table has a route to a NAT gateway or an internet-facing gateway Correct answer.

Explanation

Outbound internet access from a private subnet requires a route to a NAT gateway, and from a public subnet it requires a route to an internet gateway. Security group rules alone do not provide a path off the VPC network.

Question 41Cloud Operations Fundamentals

A team wants to confirm that a scaling group added instances during a traffic spike and later removed them. Which evidence best demonstrates this?

  1. The scaling group's scaling activity log showing scale-out and scale-in events with timestamps Correct answer.
  2. The ActionTrail record of a RAM policy change
  3. The VPC flow log showing accepted connections on port 443
  4. The CloudMonitor CPU metric for a single instance

Explanation

The scaling activity log records each scale-out and scale-in action with timing and results, which directly demonstrates the group's behavior. The other records describe traffic, a single instance's load, or an unrelated permission change.

Question 42Cloud Operations Fundamentals

An operator must limit the damage that a leaked access key could cause. Which control most directly reduces the blast radius?

  1. Enabling detailed monitoring on all ECS instances
  2. Attaching a narrowly scoped RAM policy to the identity that owns the key Correct answer.
  3. Increasing the retention period of ActionTrail events
  4. Placing all resources in a single resource group

Explanation

The blast radius of a credential is bounded by the permissions attached to its identity, so a narrowly scoped policy limits what a leaked key can do. Monitoring, audit retention, and grouping improve visibility or organization but do not restrict the key's capabilities.

Question 43Cloud Operations Fundamentals

An operator needs to reduce the time to recover a stateful application if an entire zone becomes unavailable. Which design choice contributes most directly?

  1. Increasing the size of the primary zone's database instance
  2. Enabling detailed monitoring on the primary zone's instances
  3. Adding more instances to the primary zone's scaling group
  4. Replicating the application data to a standby zone and maintaining a tested failover procedure Correct answer.

Explanation

Recovery from a zone failure requires the data and the ability to serve traffic from another zone, which cross-zone replication and a tested failover procedure provide. Monitoring and extra capacity in the failed zone do not help once that zone is unavailable.

Question 44Cloud Operations Fundamentals

An operator wants to receive an alert only when disk usage on a database server stays above 90 percent for ten consecutive minutes. Which alarm setting implements this?

  1. A threshold rule evaluated over multiple consecutive periods before the alarm triggers Correct answer.
  2. An alarm notification with a longer message body
  3. A threshold rule that triggers on a single data point
  4. A scheduled scaling task that runs every ten minutes

Explanation

Requiring several consecutive breaching periods suppresses alerts caused by brief spikes while still catching sustained conditions. A single-point rule would fire on transient peaks, and the other options do not change evaluation logic.

Question 45Cloud Operations Fundamentals

An operator wants to know which resources belong to a specific application so that costs and permissions can be scoped to that application. Which mechanism is intended for this?

  1. Assigning the resources to a dedicated resource group Correct answer.
  2. Creating an additional VPC for each application
  3. Enabling CloudMonitor detailed monitoring
  4. Adding a second security group to each instance

Explanation

Resource groups provide a logical grouping that can be used to scope cost allocation and RAM permissions to a set of resources. VPCs, security groups, and monitoring settings serve different purposes.

Question 46Cloud Operations Fundamentals

An operator must prove that no one changed a production security group during a change freeze. Which evidence combination is most appropriate?

  1. CloudMonitor CPU metrics for the instances in the security group
  2. SLB access logs for the frontend listener
  3. ActionTrail events filtered to the security group API actions for the freeze window Correct answer.
  4. VPC flow logs for the subnet

Explanation

Security group modifications are control-plane API calls, and ActionTrail records those calls with caller identity and time, allowing the operator to show that none occurred during the window. The other sources describe performance or traffic, not configuration changes.

Question 47Cloud Operations Fundamentals

An operator wants to verify that a new alarm will notify the on-call engineer without waiting for a real incident. Which approach is most reliable?

  1. Remove the contact group and observe the alarm state
  2. Lower the threshold permanently so the alarm fires in normal operation
  3. Disable the alarm and check that it no longer fires
  4. Trigger a test notification through the alarm's contact group and confirm receipt Correct answer.

Explanation

Testing the notification path through the configured contact group confirms that recipients and channels work before an incident occurs. Disabling the alarm, forcing false positives, or removing contacts does not validate delivery.

Question 48Cloud Operations Fundamentals

An operator suspects that a recent change opened a security group to the internet and wants to confirm the current effective rules and who made the change. Which combination of services is most appropriate?

  1. Inspect the security group rules directly and query ActionTrail for the related API events Correct answer.
  2. Review OSS bucket policies and inspect scaling activity logs
  3. Query CloudMonitor metrics and inspect SLB listener settings
  4. Inspect VPC route tables and query DNS records

Explanation

The current rule set shows the effective configuration, and ActionTrail attributes the modification to a caller and time. The other combinations examine performance, routing, or unrelated resources and cannot establish both facts.

Question 49Cloud Operations Fundamentals

An application tier shows rising request latency while CPU utilization stays moderate and the queue of pending requests grows. Which interpretation is most consistent with these signals?

  1. The tier is experiencing a network partition to the load balancer
  2. The tier is waiting on a downstream dependency or lock rather than being CPU-bound Correct answer.
  3. The tier's instances are failing health checks
  4. The tier has excess capacity and should be scaled in

Explanation

Growing latency with moderate CPU and a rising backlog points to a bottleneck outside the CPU, such as a slow downstream call or lock contention. CPU saturation would typically show high utilization alongside the latency increase.

Question 50Cloud Operations Fundamentals

An operator needs a repeatable process to roll back a bad configuration change across many resources with minimal manual effort. Which approach best meets this requirement?

  1. Manage the resources with infrastructure-as-code templates and redeploy the last known good version Correct answer.
  2. Take screenshots of the console before each change and restore from them by hand
  3. Rely on engineers to remember the previous settings and reapply them manually
  4. Disable ActionTrail during changes to reduce noise

Explanation

When resources are defined in templates, reverting to the previous template version restores the known good configuration consistently across all managed resources. Manual reapplication and screenshots are slow and error-prone.