Overview

Exoscale Elastic IPs (EIPs) provide static IP addresses that can be persistently attached to multiple Compute instances within the same zone, allowing seamless transition and consistent IP configurations across different instances.

Terminology

• Native IP Address
  The primary IP address automatically assigned to a Compute instance. It is distinct from any Elastic or additional IPv6 addresses that might be assigned.
• Elastic IP
  A static public IP address that an organization can attach to one or multiple Compute instances in addition to their native IP addresses. Elastic IP addresses are created for the organization and remain available until they are explicitly deleted.
• Manual Elastic IP
  A type of Elastic IP that requires manual configuration on Compute instances. It provides more flexibility, allowing traffic to originate from the manual Elastic IP address if configured correctly.
• Managed Elastic IP
  A type of Elastic IP that does not require manual configuration. When attached, it automatically routes traffic to Compute instances.
• Elastic IPv6 Prefix
  A Compute resource providing an entire IPv6 prefix (with a prefix length of /96) that can be assigned to one or several instances. It offers multiple IPv6 addresses for use within an organization.
• Traffic Distribution
  The process by which incoming traffic is distributed among multiple instances attached to a single Elastic IP. There is no guarantee of even load distribution across instances.
• Zone
  A specific geographical location where Exoscale infrastructure is available. Compute instances and Elastic IPs are bound to a certain zone, meaning resources must be within the same zone to be attached to each other.

Features

All Exoscale instances are provisioned with a native IPv4 address, leased from a global pool and intrinsically tied to the Compute instance. However, upon instance termination, the IP address is released back to the pool with no guarantee of reacquisition.

• Elastic IPs
  Ensure continuity with Elastic IPs, designed for persistent address allocation across instances. Elastic IPs supplement native addresses and can attach to multiple instances simultaneously.
• Persistent Usage
  Achieve IP persistence through Elastic IPs, mitigating address changes upon instance destruction. Seamlessly reroute traffic with an Elastic IP when transitioning between underlying instances.
• Multi-IP Scenarios
  Elastic IPs address complex requirements like multi-TLS certificate usage or running numerous service instances on a single machine.

NOTE
Leverage Elastic IPs for operational continuity and flexibility across your network architecture.

Elastic IP Types

• Manual Elastic IPs
• Managed Elastic IPs

Basic Behavior of Elastic IPs (EIPs)

• Creation and Availability
  Elastic IPs are created for your organization. They remain available until you decide to delete them.
• Attachment and Zone Binding
  Elastic IPs can be attached to one or multiple instances at any time, but can only be attached to Compute instances within the same zone.
• Traffic Characteristics
  All incoming traffic towards an Elastic IP is redirected to the attached instances. All outgoing traffic is sent by the internal IP address attached to the instance. It will not appear as coming from the Elastic IP. When multiple instances are attached to an Elastic IP, traffic is distributed among them with no guarantee of even load distribution.

Customizable Behavior
Both manual and managed Elastic IPs have distinct characteristics, their behavior can be customized further based on requirements. Distinct Characteristics of Manual and Managed Elastic IPs are:

Manual Elastic IPs

• Configuration
  Manual Elastic IPs need to be manually enabled on the instance, this type offers more flexibility but requires more manual configuration.
• Traffic Perception
  Internally, instances attached to a manual Elastic IP see traffic as coming directly to the Elastic IP address, this is different from managed Elastic IPs, where incoming traffic is seen as coming to the native IP address of the attached instance.
• Outgoing Traffic
  Outgoing traffic can originate from the manual Elastic IP, allowing the instance to use it as a traffic source, albeit with some limitations.

Managed Elastic IPs

• Configuration
  Managed Elastic IPs do not require manual action on your part for the instance itself. Once attached, a managed Elastic IP transparently routes traffic to the instance.
• Traffic Routing
  Includes basic health check capabilities, ensuring traffic is routed only to healthy instances.
• Traffic Perception
  Instances see incoming traffic as if it’s from the native IP address, making it impossible to differentiate traffic based on the target IP.
• Outgoing Traffic
  Outgoing traffic will always appear from the native IP address, unlike manual Elastic IPs.

Elastic IPv6 Prefix

Exoscale Compute instances can optionally be assigned a public IPv6 address either during creation or afterward. This IPv6 address is partially derived from the instance’s unique MAC address, making it intrinsically tied to the instance. Therefore, when an instance is destroyed, its public IPv6 address is permanently released.

In scenarios where a persistent IPv6 address is desirable, such as employing multiple IPv6 addresses for a single use case, an Elastic IPv6 prefix can be used. The Elastic IPv6 prefix is a Compute resource that provides your organization with an entire IPv6 prefix (with prefix length /96), which can be attached to one or several instances in addition to their primary IPv6 address.

To utilize these prefixes, instances must have IPv6 enabled. Your organization can attach multiple Elastic IPv6 prefixes as needed. Unlike its IPv4 counterpart, an Elastic IPv6 prefix contains over 4 billion addresses, providing ample resources to cater to your instances’ networking requirements.

Availability

Limitations

• Elastic IPs are initially limited to 5 per organization. You can ask for a quota iIP addresses are not managed by Exoscale APIs, but are parameters in operations with platform products like instances (LIST, CREATE) increase in the quotas section of the Portal.
• Instances sharing a common Elastic IP cannot communicate with each other by using the Elastic IP address.
• Elastic IP will natively work only for incoming traffic. Outgoing traffic will still use the native instance’s IP address as a source. This behavior can be partially circumvented manually with manual Elastic IPs.
• Instances associated with a managed Elastic IP will see incoming Elastic IP traffic destined to their primary interface IP address.
• With managed Elastic IPs, it is currently not possible to retrieve the health status of an instance: while the system will not direct traffic to a unhealthy instance, there is no information about which instance is in an unhealthy state.
• HTTP health checks do not follow redirects.
• Elastic IPv6 prefixes have the same limitations as Elastic IPs.
• For managed Elastic IPv6 prefixes, only the first-plus-one IPv6 address of that prefix is transparently forwarded to the associated instances.