I85PREP
⌘ K
Exoscale Certified Cloud Professional study materials
Exoscale Certified Cloud Professional study resources

Exoscale Certified Cloud Professional practice questions

One question at a time. Build your Exoscale Certified Cloud Professional 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 1Compute

Which Exoscale compute offering provides pre-configured virtual machine templates that can be deployed directly from the Exoscale portal, CLI, or API without installing an operating system manually?

  1. Private networks
  2. Instance templates Correct answer.
  3. Anti-affinity groups
  4. Block storage volumes

Explanation

Instance templates are the pre-built images (operating system plus optional preinstalled software) that Exoscale uses to boot compute instances, so a customer can deploy a working VM immediately. Private networks, block storage, and anti-affinity groups are supporting resources that do not contain a bootable operating system image.

Question 2Compute

An operator wants to stop an Exoscale compute instance temporarily while keeping its disk contents intact and preserving the ability to start it again later. Which instance state best matches this goal?

  1. Starting
  2. Migrating
  3. Destroyed
  4. Stopped Correct answer.

Explanation

A stopped instance is powered off but its disk and configuration remain allocated, so it can be started again with its data intact. Destroying an instance releases the resources and removes the instance from the account, which is a different and irreversible operation.

Question 3Compute

Which Exoscale feature lets you place several instances of the same service on different physical hypervisors so that a single host failure does not take down the whole service?

  1. Elastic IP addresses
  2. Instance pools
  3. Anti-affinity groups Correct answer.
  4. Security groups

Explanation

Anti-affinity groups instruct the platform to schedule member instances on distinct hypervisors, reducing correlated failure. Elastic IPs, security groups, and instance pools address addressing, filtering, and scaling respectively, not hypervisor placement.

Question 4Compute

A developer needs to run a containerized workload on Exoscale without managing the underlying virtual machines. Which compute service is designed for this?

  1. Block storage
  2. SKS (Scalable Kubernetes Service) Correct answer.
  3. DNS
  4. Instance templates

Explanation

SKS is Exoscale's managed Kubernetes service, which provisions and operates the control plane so customers deploy containerized workloads without administering the cluster's VMs directly. Instance templates, block storage, and DNS are lower-level or unrelated services.

Question 5Compute

Which Exoscale compute instance family is intended for workloads that need a high ratio of CPU to memory, such as batch processing or web front ends?

  1. Memory-optimized
  2. Storage-optimized
  3. GPU
  4. Standard Correct answer.

Explanation

The Standard family offers a balanced but CPU-leaning ratio suitable for general web and batch workloads. GPU, memory-optimized, and storage-optimized families target acceleration, large in-memory datasets, and high disk throughput respectively.

Question 6Compute

Which Exoscale compute capability allows an instance to be created from a snapshot of another instance's disk, effectively cloning a configured system?

  1. Instance pool
  2. Custom templates Correct answer.
  3. Elastic IP
  4. Private network

Explanation

Custom templates are created from an instance's disk or snapshot and can then be used to launch new instances with the same preconfigured state. Elastic IPs, private networks, and instance pools do not capture disk contents.

Question 7Compute

Which Exoscale compute resource is required before an instance can be attached to a private network?

  1. A second block storage volume
  2. A private network in the same zone as the instance Correct answer.
  3. A DNS A record for the instance
  4. A public IP address on the instance

Explanation

Private networks are zone-scoped, so the instance and the target private network must reside in the same zone before attachment is possible. Public IPs, DNS records, and extra volumes are independent of private network attachment.

Question 8Compute

Which statement describes the purpose of an Exoscale instance pool?

  1. It assigns static public IP addresses to instances.
  2. It encrypts the disks of member instances.
  3. It provides a shared filesystem across instances.
  4. It maintains a desired number of identical instances and can scale them automatically. Correct answer.

Explanation

An instance pool manages a group of identical instances from one template, keeps the desired count, and can scale based on rules or metrics. Shared filesystems, IP assignment, and disk encryption are separate features.

Question 9Compute

An instance running in the CH-GVA-2 zone needs to be resized to a larger instance type. Which action accomplishes this on Exoscale?

  1. Stop the instance, change its service offering, then start it. Correct answer.
  2. Create an elastic IP and reassign it to a larger instance.
  3. Move the instance to a different anti-affinity group.
  4. Attach an additional block volume to increase CPU.

Explanation

Resizing requires stopping the instance, updating its service offering (instance type), and starting it again so the new CPU and memory allocation takes effect. Elastic IPs, volumes, and affinity groups do not change compute capacity.

Question 10Compute

A team deploys three web servers in a single instance pool and wants the failure of one hypervisor to affect at most one member. Which configuration should they apply?

  1. Place all instances in one security group.
  2. Attach a shared block volume to each instance.
  3. Attach the pool to an anti-affinity group. Correct answer.
  4. Assign each instance an elastic IP.

Explanation

Anti-affinity groups ensure pool members are scheduled on separate hypervisors, so a single host failure impacts at most one member. Elastic IPs, security groups, and block volumes do not influence hypervisor placement.

Question 11Compute

A developer wants to automate instance creation using the Exoscale API. Which identifier is needed to specify the operating system image when creating the instance?

  1. The private network ID
  2. The template ID Correct answer.
  3. The elastic IP ID
  4. The security group ID

Explanation

Instance creation requires a template identifier that selects the bootable image. Security groups, elastic IPs, and private networks are networking resources optionally attached to the instance after the image is chosen.

Question 12Compute

An operator must ensure an instance receives a stable public IP address that can be quickly remapped to a standby instance during failover. Which Exoscale resource fits this requirement?

  1. Elastic IP Correct answer.
  2. Private network
  3. Anti-affinity group
  4. Instance template

Explanation

Elastic IPs are movable public addresses that can be detached from one instance and attached to another, enabling fast failover. Private networks, templates, and affinity groups do not provide a remappable public address.

Question 13Compute

A customer needs an instance with GPU acceleration for machine learning inference. Which Exoscale instance family should they select?

  1. Storage-optimized
  2. Standard
  3. GPU Correct answer.
  4. Memory-optimized

Explanation

GPU instance types include NVIDIA GPUs for accelerated workloads such as ML inference and rendering. Standard, memory-optimized, and storage-optimized families provide CPU, RAM, or disk emphasis without GPU acceleration.

Question 14Compute

An administrator wants to run a startup script automatically the first time an instance boots. Which Exoscale mechanism is designed for this?

  1. Block volume snapshots
  2. Elastic IP reassignment
  3. User data (cloud-init) Correct answer.
  4. Security group egress rules

Explanation

User data is passed to the instance at creation and consumed by cloud-init to run configuration or scripts on first boot. Egress rules, elastic IP reassignment, and volume snapshots do not execute boot-time configuration.

Question 15Compute

A team wants to deploy a second identical instance pool in the DE-FRA-1 zone. What must they do to make the existing custom template usable there?

  1. Copy the template to the DE-FRA-1 zone. Correct answer.
  2. Create a new security group in DE-FRA-1.
  3. Attach an elastic IP in DE-FRA-1.
  4. Move the anti-affinity group to DE-FRA-1.

Explanation

Templates are zone-scoped, so a custom template must be copied to the target zone before instances can be launched from it there. Elastic IPs, security groups, and affinity groups do not make an image available in another zone.

Question 16Compute

A workload requires very high local disk throughput for a caching tier that stores large datasets on NVMe. Which Exoscale instance family is most appropriate?

  1. Memory-optimized
  2. Standard
  3. GPU
  4. Storage-optimized Correct answer.

Explanation

Storage-optimized instance types provide large, high-throughput local NVMe storage suited to caching and data-intensive tiers. Standard, GPU, and memory-optimized families do not emphasize local disk throughput.

Question 17Compute

An operator wants to scale an instance pool automatically based on CPU utilization. Which Exoscale feature should be configured on the pool?

  1. Elastic IP failover
  2. Anti-affinity rules
  3. Autoscaling rules with metric thresholds Correct answer.
  4. Private network DHCP ranges

Explanation

Instance pools support autoscaling rules that adjust the number of members when a metric such as average CPU crosses a threshold. Anti-affinity, elastic IP failover, and DHCP ranges do not change pool size.

Question 18Compute

A developer wants to launch an instance and have it join an existing private network in the same zone at creation time. Which parameter must be supplied in the creation request?

  1. The DNS domain name
  2. The private network ID Correct answer.
  3. The elastic IP ID
  4. The snapshot ID

Explanation

Supplying the private network identifier in the creation request attaches the instance to that network immediately, provided it is in the same zone. Elastic IPs, DNS names, and snapshots are unrelated to network membership.

Question 19Compute

An administrator needs to build a golden image from a running instance's disk so that future instances start with the same software. Which two-step approach is correct on Exoscale?

  1. Add a private network, then enable DHCP.
  2. Create a security group, then attach it to the instance.
  3. Create a snapshot of the instance disk, then register a custom template from it. Correct answer.
  4. Attach an elastic IP, then create an instance pool.

Explanation

A snapshot captures the disk state, and registering a custom template from that snapshot makes the image reusable for new instances. The other options configure networking or scaling and do not produce a bootable image.

Question 20Compute

A service must survive the loss of an entire availability zone. Which Exoscale compute design best achieves this?

  1. Run all instances in one instance pool with autoscaling enabled.
  2. Place all instances in one anti-affinity group in a single zone.
  3. Attach all instances to one elastic IP in a single zone.
  4. Deploy identical instance pools in at least two zones behind a global load-balancing layer. Correct answer.

Explanation

Zone-level resilience requires resources in multiple zones so that the loss of one zone leaves capacity elsewhere, typically combined with global load balancing. Anti-affinity, a single elastic IP, and a single pool all remain confined to one zone and cannot survive its loss.

Question 21Compute

An application running on an Exoscale instance must reach a database on a private network without exposing the database to the internet. Which configuration achieves this?

  1. Attach both instances to the same private network and allow the port in the security group. Correct answer.
  2. Place both instances in the same anti-affinity group.
  3. Attach a shared block volume to both instances.
  4. Assign both instances elastic IPs and allow the port publicly.

Explanation

Private networks provide internal connectivity, and security group rules permit the required port between members without public exposure. Elastic IPs would expose the database, while affinity groups and shared volumes do not provide network paths.

Question 22Compute

A team observes that an instance pool repeatedly launches instances that fail health checks. Which change most directly addresses a misconfigured boot image rather than the pool's scaling policy?

  1. Increase the pool's maximum instance count.
  2. Lower the autoscaling CPU threshold.
  3. Rebuild the custom template with corrected cloud-init and update the pool. Correct answer.
  4. Add more elastic IPs to the pool.

Explanation

If instances boot into a broken state, the underlying template or its cloud-init configuration must be fixed and the pool pointed at the corrected template. Scaling counts, elastic IPs, and CPU thresholds change capacity or addressing but not the boot behavior.

Question 23Compute

An operator plans to resize a production instance to a larger type. Which consideration is most important for minimizing downtime?

  1. Resizing automatically moves the instance to another zone.
  2. Resizing requires deleting and recreating the private network.
  3. The instance must be stopped during the resize, so plan a maintenance window or use a rolling replacement. Correct answer.
  4. The resize happens live with no interruption on all instance types.

Explanation

Changing an instance's service offering requires stopping it, so production impact should be handled with a maintenance window or by replacing instances in a rolling fashion. Resizes are not universally live, and they do not require network deletion or zone migration.

Question 24Compute

A team wants to reduce cost for a batch job that runs a few hours each night. Which approach best matches Exoscale compute economics?

  1. Attach additional elastic IPs to idle instances.
  2. Resize instances to the largest type permanently.
  3. Keep instances running 24/7 in an anti-affinity group.
  4. Start instances on a schedule and destroy them when the job finishes. Correct answer.

Explanation

Compute billing is tied to allocated running instances, so scheduling start and destroy around the batch window avoids paying for idle capacity. Keeping instances running, adding IPs, or permanently upsizing increases cost without benefit.

Question 25Compute

An instance in an anti-affinity group cannot be started because the group's constraint cannot be satisfied. What is the most likely cause?

  1. The instance lacks an elastic IP address.
  2. The instance's private network has exhausted its DHCP range.
  3. Too many members already occupy distinct hypervisors, leaving no compliant host for the new instance. Correct answer.
  4. The instance's security group has no egress rules.

Explanation

Anti-affinity requires each member to land on a different hypervisor, so when compliant hosts are exhausted the start fails. Missing elastic IPs, egress rules, or DHCP addresses cause connectivity problems, not placement failures.

Question 26Compute

Which Exoscale compute component defines the CPU, memory, and local storage capacity that an instance receives at launch?

  1. The zone
  2. The template
  3. The security group
  4. The instance type Correct answer.

Explanation

An instance type is the hardware profile that fixes the number of vCPUs, the amount of RAM, and the size of local disk presented to a compute instance. The template supplies the operating system image, while networking and placement are handled separately. Choosing the type therefore determines the resource envelope of the guest.

Question 27Compute

Which statement describes what an Exoscale compute instance template provides when launching a new instance?

  1. A billing plan tied to monthly commitment
  2. A reserved public IP address
  3. A preconfigured bootable operating system image Correct answer.
  4. A firewall rule set for inbound traffic

Explanation

A template is a bootable disk image, typically a distribution such as a Linux variant or a custom image registered by the customer. It defines what the instance boots into, and the instance type defines the hardware. Networking and billing are configured through other mechanisms.

Question 28Compute

Which Exoscale CLI command family is used to manage compute instances from a terminal?

  1. exo dns domain
  2. exo iam key
  3. exo storage bucket
  4. exo compute instance Correct answer.

Explanation

The Exoscale CLI groups resources under service subcommands, and compute instances are managed through the compute instance group of commands. Storage, DNS, and IAM each have their own command groups. Using the correct group is required before the instance-specific verbs are accepted.

Question 29Compute

What does the Exoscale compute concept of an anti-affinity group constrain?

  1. Which public IPs the member instances may receive
  2. Which zones the member instances may use
  3. Which hypervisors may host the member instances Correct answer.
  4. Which templates the member instances may boot

Explanation

An anti-affinity group ensures that member instances are placed on different physical hypervisors so that a single host failure affects a limited number of them. It is a placement constraint, not a network or image constraint. Zone selection and addressing remain independent decisions.

Question 30Compute

A developer needs to start an instance and pass a small amount of configuration data that a cloud-init enabled template will read on first boot. Which Exoscale mechanism should be used?

  1. A security group rule
  2. A private network lease
  3. User data attached to the instance Correct answer.
  4. An elastic IP

Explanation

User data is a payload supplied at instance creation that cloud-init consumes during initial boot to configure the guest. It is the standard way to inject scripts or configuration without building a custom template. Security groups, elastic IPs, and private networks serve networking purposes instead.

Question 31Compute

An operator wants to protect an instance from accidental deletion while it is still in use. Which Exoscale instance setting addresses this directly?

  1. Attaching an elastic IP to the instance
  2. Enabling delete protection on the instance Correct answer.
  3. Assigning a larger instance type
  4. Adding the instance to an anti-affinity group

Explanation

Delete protection is a flag on the instance that causes the API to reject deletion requests until the flag is removed. It guards against accidental or scripted removal. Anti-affinity affects placement, elastic IPs affect addressing, and instance type affects sizing.

Question 32Compute

A team wants a new instance to appear in the Exoscale console with a meaningful label that identifies its role, separate from its hostname. Which instance attribute should they set?

  1. The instance name Correct answer.
  2. The template
  3. The zone
  4. The instance type

Explanation

The instance name is a human-readable label used for identification in the console and API, and it is independent of the guest hostname configured inside the operating system. Instance type, zone, and template are fixed infrastructure choices. Setting the name improves operational clarity without affecting the running workload.

Question 33Compute

A developer creates an instance from the Exoscale API and needs the request to succeed only if the specified instance type is available in the target zone. Which request element determines whether the type is accepted?

  1. The user data payload alone
  2. The instance type identifier combined with the zone Correct answer.
  3. The security group identifier alone
  4. The template identifier alone

Explanation

Instance types are offered per zone, so the API validates the requested type against the zone in the same request. Supplying a type that is not available in that zone causes the call to fail. Template, security group, and user data do not influence type availability.

Question 34Compute

An operator must stop an instance without destroying its disk so that it can be started again later with the same state. Which Exoscale instance state transition accomplishes this?

  1. Scale the instance pool to zero
  2. Delete the instance
  3. Detach the elastic IP
  4. Stop the instance, leaving it in a stopped state Correct answer.

Explanation

Stopping an instance powers it off while retaining the instance and its disk, so a later start resumes from the preserved state. Deleting removes the instance, and pool scaling affects group membership rather than an individual instance's disk. Address detachment is a networking action.

Question 35Compute

A customer wants to run a containerized workload on Exoscale without managing the underlying virtual machines directly. Which Exoscale compute offering fits this requirement?

  1. Exoscale Elastic IP
  2. Exoscale Object Storage
  3. Exoscale Managed Kubernetes (SKS) Correct answer.
  4. Exoscale DNS

Explanation

SKS is a managed Kubernetes service in which Exoscale operates the control plane and node lifecycle, letting teams deploy containers without administering the nodes directly. Object Storage, DNS, and Elastic IP are supporting services rather than container orchestration platforms. SKS is the compute product aligned with this requirement.

Question 36Compute

A team wants to run a short-lived job on a schedule each night and pay only for the compute consumed. Which Exoscale compute approach best matches this pattern?

  1. Keep a large instance running continuously
  2. Reserve a dedicated host permanently
  3. Attach the instance to an anti-affinity group
  4. Create the instance before the job and delete it after completion Correct answer.

Explanation

Exoscale compute instances are billed for the time they exist, so creating an instance for the job window and deleting it afterwards limits cost to actual usage. Leaving an instance running continuously incurs charges outside the job window. Placement and dedication options do not reduce idle cost.

Question 37Compute

An operator needs to change the instance type of a running instance to a larger size. What does Exoscale require for this operation?

  1. The instance must be stopped before the type can be changed Correct answer.
  2. The instance must be detached from all security groups
  3. The instance must be moved to another zone first
  4. The instance must be deleted and recreated

Explanation

Changing the instance type requires the instance to be stopped, after which the new type is applied and the instance started again. The disk is preserved through this cycle, so data remains intact. Deleting and recreating, zone migration, and security group changes are not prerequisites.

Question 38Compute

A team needs to attach additional block storage to an instance for data that must persist independently of the instance lifecycle. Which Exoscale resource should be created?

  1. A second elastic IP
  2. A new private network
  3. An additional security group
  4. A block storage volume attached to the instance Correct answer.

Explanation

Block storage volumes are separate resources that can be attached to and detached from instances, so the data survives instance replacement. Elastic IPs, security groups, and private networks are networking constructs without persistent data capacity. A volume is the correct resource for lifecycle-independent storage.

Question 39Compute

A workload must run on hardware dedicated to a single customer for compliance reasons. Which Exoscale compute option addresses this requirement?

  1. A dedicated host / private compute offering Correct answer.
  2. A shared instance type with more vCPUs
  3. An elastic IP with a static address
  4. An anti-affinity group

Explanation

Dedicated compute options provide hardware reserved for one customer, which addresses isolation and compliance requirements that shared tenancy cannot satisfy. Adding vCPUs to a shared type does not change tenancy. Anti-affinity and elastic IPs affect placement and addressing but not hardware isolation.

Question 40Compute

An operator notices that instances in a pool are being replaced frequently even though the application responds correctly to manual checks. Which pool configuration should be examined first to reduce unnecessary replacements?

  1. The pool's disk size
  2. The health check method and its interval and grace settings Correct answer.
  3. The pool's instance name prefix
  4. The pool's security group description

Explanation

Pool health checks determine whether instances are considered healthy; an overly aggressive method, interval, or insufficient grace period can mark working instances unhealthy and trigger replacement. The name prefix, security group description, and disk size do not influence health verdicts. Tuning the check is therefore the first step.

Question 41Storage

A team wants block storage they can attach to a single Compute instance and format with any filesystem, persisting independently of the instance lifecycle. Which Exoscale storage product fits this requirement?

  1. Managed Database service storage
  2. Object Storage buckets
  3. Simple Object Storage with versioning
  4. Block Storage volumes Correct answer.

Explanation

Block Storage provides raw volumes that attach to a Compute instance and can be partitioned and formatted with the filesystem of your choice. The volume persists independently, so it survives instance stop/start and can be reattached to another instance. Object Storage is accessed over HTTP and is not a block device.

Question 42Storage

Which Exoscale Object Storage characteristic distinguishes it from Block Storage volumes?

  1. It exposes an S3-compatible HTTP API for storing objects in buckets Correct answer.
  2. It is only reachable from inside a private network
  3. It can be formatted with ext4 and mounted as a root filesystem
  4. It must be attached to exactly one Compute instance at a time

Explanation

Object Storage is an S3-compatible service where data is stored as objects inside buckets and accessed over HTTP(S). It is not a block device and is not attached to instances. Block Storage volumes are the ones that get attached and formatted.

Question 43Storage

A Block Storage volume is created in zone CH-GVA-2 but the target Compute instance runs in CH-DK-2. What is the outcome?

  1. The volume attaches but with reduced IOPS
  2. The volume must first be converted to Object Storage
  3. The volume can be attached only to instances in the same zone Correct answer.
  4. The volume is automatically replicated to the instance's zone

Explanation

Block Storage volumes are zonal resources and can only be attached to Compute instances in the same zone. An instance in a different zone cannot see or attach the volume. To use storage in another zone you create a new volume there.

Question 44Storage

What is the primary purpose of a snapshot in Exoscale Block Storage?

  1. To encrypt the volume with a customer-managed key
  2. To increase the maximum IOPS of the attached volume
  3. To migrate a volume between zones automatically
  4. To capture a point-in-time copy of a volume for backup or cloning Correct answer.

Explanation

A snapshot records the state of a Block Storage volume at a point in time, enabling backup and restoration or creation of new volumes. It does not change performance characteristics or zone placement. Encryption is a separate concern from snapshotting.

Question 45Storage

Which statement correctly describes Exoscale Object Storage bucket naming within a zone?

  1. Bucket names may contain uppercase letters and spaces
  2. Bucket names must be unique within the zone and follow DNS-compatible rules Correct answer.
  3. Bucket names are assigned automatically and cannot be chosen
  4. Bucket names must be globally unique across all Exoscale zones

Explanation

Exoscale Object Storage buckets follow S3-style naming: lowercase, DNS-compatible characters, and unique within the zone where they are created. Because the namespace is per-zone, the same name can exist in different zones. Users choose the bucket name at creation time.

Question 46Storage

Which access method is standard for programmatic interaction with Exoscale Object Storage?

  1. NFS mounts exported by the zone
  2. An iSCSI initiator over the private network
  3. The S3-compatible API using access key and secret key Correct answer.
  4. SSHFS over the instance's public IP

Explanation

Exoscale Object Storage exposes an S3-compatible API, and clients authenticate with an access key and secret key. Tools such as the AWS CLI or s3cmd can be pointed at the Exoscale endpoint. iSCSI, NFS and SSHFS are not the supported access path for Object Storage.

Question 47Storage

What happens to data on a Block Storage volume when the attached Compute instance is stopped and later started?

  1. The data persists only if the instance was rebooted, not stopped
  2. The data is moved to Object Storage automatically
  3. The data is erased when the instance stops
  4. The data persists because the volume is independent of the instance Correct answer.

Explanation

Block Storage volumes exist independently of Compute instances, so stopping or starting an instance does not erase the volume's contents. The volume can be detached and reattached to another instance in the same zone. Instance lifecycle actions do not relocate data to Object Storage.

Question 48Storage

Which Exoscale feature lets you define lifecycle rules that transition or expire objects in a bucket?

  1. Anti-affinity groups
  2. Block Storage snapshot schedules
  3. Object Storage lifecycle policies Correct answer.
  4. Instance pools

Explanation

Object Storage supports lifecycle configuration that can expire objects or transition them according to rules. This is an object-level management feature. Snapshot schedules apply to Block Storage volumes, and anti-affinity and instance pools are compute concepts.

Question 49Storage

An application writes many small objects to Object Storage and reads them back frequently. Which design choice most directly reduces request latency for repeated reads?

  1. Convert the bucket to a Block Storage volume
  2. Disable bucket versioning to reduce object count
  3. Enable a CDN or caching layer in front of the bucket endpoint Correct answer.
  4. Increase the Block Storage volume size attached to the client

Explanation

Object Storage is optimized for durable object access, and repeated reads benefit from a caching layer or CDN in front of the endpoint. Volume size and block storage are unrelated to object read latency. Versioning affects retained copies, not read speed.

Question 50Storage

A team needs a volume larger than the default maximum offered for a single Block Storage volume. What is the appropriate approach?

  1. Enable Object Storage versioning on the volume
  2. Stripe data across multiple volumes or revisit the size limit with the provider's current offering Correct answer.
  3. Convert the volume to a snapshot and attach the snapshot
  4. Attach the same volume twice to double its capacity

Explanation

When a single volume cannot meet capacity needs, the usual pattern is to use multiple volumes and stripe or partition data across them, or confirm the current maximum size offered. A volume cannot be double-attached to increase capacity, and snapshots are not attachable storage.