diff --git a/menu/navigation.json b/menu/navigation.json index d51d8c9c03..88b2c1a33f 100644 --- a/menu/navigation.json +++ b/menu/navigation.json @@ -1418,10 +1418,6 @@ "label": "Migrate Local Storage volumes to Block Storage", "slug": "migrate-local-storage-to-sbs" }, - { - "label": "Move an Instance to routed flexible IPs", - "slug": "migrate-routed-ips" - }, { "label": "Use Security Groups", "slug": "use-security-groups" diff --git a/pages/instances/faq.mdx b/pages/instances/faq.mdx index 8d17a77af8..095cadde45 100644 --- a/pages/instances/faq.mdx +++ b/pages/instances/faq.mdx @@ -330,4 +330,4 @@ Yes, they can communicate with each other using their public IPs. Currently, additional routed IPv6 addresses do not autoconfigure on CentOS 7, 8, 9, Alma 8, 9, Rocky 8, 9 after migration. Additional routed IPv4 and IPv6 addresses are not autoconfigured post-migration on Ubuntu 20.04 Focal. However, the primary IPv6 continues to be configured via SLAAC. These limitations are currently being addressed. -For detailed migration steps, refer to our [migration guide](/instances/how-to/migrate-routed-ips/). If you encounter connectivity issues with Ubuntu Focal Instances having multiple public IPs, consult our [troubleshooting guide](/instances/troubleshooting/fix-unreachable-focal-with-two-public-ips/). \ No newline at end of file +If you encounter connectivity issues with Ubuntu Focal Instances having multiple public IPs, consult our [troubleshooting guide](/instances/troubleshooting/fix-unreachable-focal-with-two-public-ips/). \ No newline at end of file diff --git a/pages/instances/how-to/migrate-routed-ips.mdx b/pages/instances/how-to/migrate-routed-ips.mdx deleted file mode 100644 index df30b12895..0000000000 --- a/pages/instances/how-to/migrate-routed-ips.mdx +++ /dev/null @@ -1,91 +0,0 @@ ---- -title: How to move an Instance to routed flexible IPs -description: This page explains how to move your Instance to routed flexible IPs -tags: instance flexible ip routed -dates: - validation: 2025-02-06 - posted: 2023-10-11 -categories: - - compute ---- -import Requirements from '@macros/iam/requirements.mdx' - -import image from './assets/scaleway-instance-routedip-transition.webp' - - - - The migration to routed IPs is now complete. The following information is provided for reference only. - - -Prior to April 2024, all Instances got a NAT IP from their hosting node, which was routed to the Instance through a [carrier-grade NAT](/instances/concepts/#carrier-grade-nat-cgnat) solution. This was in addition to the public (flexible) IP. The NAT IP let Instances communicate via an underlying internal network, without public IPs. However, if the underlying hypervisor changed, the NAT IP changed also. This could happen when you restarted your Instance. - -Since October 2024 it is no no longer possible to create Instances with NAT IPs. Instead, Instances' public IPs are **routed** IPs. Routing a public IP directly to an Instance bypasses NAT IPs, sending traffic to its [routed flexible IP](/instances/concepts/#routed-flexible-ip). The benefit is the IP stays consistent even if the Instance is restarted or moved. Routed IPs also let you assign multiple flexible IPv4s and IPv6s to an Instance, as both IPv4 and IPv6 are supported. - -In addition to the benefits explained above, routed IPs allow precise control over network configurations, simplify troubleshooting, ensure compatibility with more applications, and allow for enhanced security monitoring. We recommend that users with Instances still using NAT IPs move them to routed IPs by following the instructions below. - - - * The routed IP feature **is incompatible** with legacy Instance types using the bootscript feature. - * Ensure no static network configuration is in use for your public network interface (usually `ens2` or `eth0`) - * Ensure that you are not using the internal private IP of the Instance for internal communications. Instead, [attach your Instances to a VPC](/vpc/how-to/attach-resources-to-pn/). - * You must update the `scaleway-ecosystem` and `cloud-init` packages before starting transition. - * Before proceeding with the migration, please review the [FAQ on IP autoconfiguration limitations](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature) and the [guide to fix unreachable Ubuntu Focal Instances with multiple public IPs](/instances/troubleshooting/fix-unreachable-focal-with-two-public-ips/). These resources provide important information to ensure a seamless migration process. - - - - -- A Scaleway account logged into the [console](https://console.scaleway.com) -- An [Instance](/instances/how-to/create-an-instance/) that has not been moved to routed IPs - -## Updating the required packages - - - To ensure optimal compatibility and security, especially when working with flexible IPs, we recommend to upgrade your Instance to OS versions that have not reached their end of life. - - -Before proceeding with the migration, consider the following guidelines regarding `scaleway-ecosystem` and `cloud-init`: - -- `scaleway-ecosystem` and `cloud-init`: These packages are not strictly required for routed IPs. However, if they are installed, it is recommended to update them to the latest version. -- Package availability on older Instances: - - `scaleway-ecosystem`: Older Instances may not have this package installed. In such cases, no action is required. - - `cloud-init`: This package is commonly pre-installed on Instances due to its inclusion in upstream cloud images. For older Instances, a legacy version of `cloud-init` may not impact IPv4 functionality, but IPv6 functionality could be inconsistent with outdated versions. - -To check if `scaleway-ecosystem` and `cloud-init` are installed and to update them if needed, [log into your Instance using SSH](/instances/how-to/connect-to-instance/) and run the following commands: - -* For **Ubuntu and Debian** Instances: - ```bash - dpkg-query -W scaleway-ecosystem || echo "scaleway-ecosystem is not installed" - apt upgrade scaleway-ecosystem cloud-init -y - ``` - -* For **CentOS and Fedora** Instances: - ```bash - rpm -q scaleway-ecosystem || echo "scaleway-ecosystem is not installed" - dnf update scaleway-ecosystem cloud-init -y - ``` - -If `scaleway-ecosystem` is not installed, no additional steps are needed for this package. However, it is still recommended to ensure that `cloud-init` is up-to-date to support IPv6 functionality. - -## Moving the Instance to routed IPs - -1. Select **CPU & GPU Instances** from the **Compute** category in the sidebar menu. This will show a list of your Instances. Those eligible to transition to routed IPs are identified with an exclamation mark icon. - -2. Click **Move IP** next to the Instance you wish to move to routed flexible IPs. A pop-up displays. - - Alternatively, you can initiate the transition from your Instance's overview page by clicking the **Move IP** button. - -3. Type **MIGRATE** to confirm your intent and click **Move to routed IP**. This will initiate the transition process, and your Instance will undergo a reboot. - - Post-transition, your Instance will retain its public IP address, but its NAT IP will be removed. - -After the transition finishes, the notification icon will no longer be visible in your Instances list. - - - For detailed guidance on using the routed flexible IP feature via the Scaleway API or CLI tool, refer to our dedicated documentation [Routing public IPs to Instances using the Scaleway API](/instances/api-cli/using-routed-ips/). - - - -## IPv6 support - -Routed IPs provide flexible IPv6 support. This means you can now assign an IPv6 address that will remain the same wherever the Instance is located. Before the advent of routed IPs, IPv6 addresses were assigned per node location, thus whenever an Instance was relocated, its IPv6 address was changed. -Note that for the best experience with IPv6, your Instance must be supported (we do not provide support for EOL Operating Systems). -For IPv6 only, you need `cloud-init` version [23.2](https://github.com/canonical/cloud-init/releases/tag/23.2) or later and `scaleway-ecosystem` version 0.6 or later. diff --git a/pages/instances/troubleshooting/fix-broken-vpn-when-switching-openvpn-vm-to-routed-ip.mdx b/pages/instances/troubleshooting/fix-broken-vpn-when-switching-openvpn-vm-to-routed-ip.mdx index 9d2c574fea..3bcc768982 100644 --- a/pages/instances/troubleshooting/fix-broken-vpn-when-switching-openvpn-vm-to-routed-ip.mdx +++ b/pages/instances/troubleshooting/fix-broken-vpn-when-switching-openvpn-vm-to-routed-ip.mdx @@ -16,7 +16,7 @@ This can now be fixed by installing a new version of the OpenVPN scripts, which You can also avoid this situation altogether by installing the package prior to moving to a routed IP. - This guide addresses specific issues related to OpenVPN Instances. For general information on routed IPs and migration procedures, refer to our [main migration guide](/instances/how-to/migrate-routed-ips/) and the [related FAQ](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature). + This guide addresses specific issues related to OpenVPN Instances. For general information on routed IPs and migration procedures, refer to the [related FAQ](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature). diff --git a/pages/instances/troubleshooting/fix-lost-ip-connectivity-on-debian-buster.mdx b/pages/instances/troubleshooting/fix-lost-ip-connectivity-on-debian-buster.mdx index 0c0abd2c9f..491739633c 100644 --- a/pages/instances/troubleshooting/fix-lost-ip-connectivity-on-debian-buster.mdx +++ b/pages/instances/troubleshooting/fix-lost-ip-connectivity-on-debian-buster.mdx @@ -12,7 +12,7 @@ categories: On older Debian Buster images, the installed custom version of `cloud-init` may interfere with IPv6 connectivity when the Instance transitions to using routed IP. To avoid this, you should install a newer version of `cloud-init` before the migration to routed IP. This procedure also recovers connectivity for an Instance already using routed IP. - This guide addresses specific issues related to older Debian Buster Instances transitioned to routed IP. For general information on routed IPs and migration procedures, refer to our [main migration guide](/instances/how-to/migrate-routed-ips/) and the [related FAQ](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature). + This guide addresses specific issues related to older Debian Buster Instances transitioned to routed IP. For general information on routed IPs and migration procedures, refer to the [related FAQ](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature). 1. Install the new `debian-stable` PPA. diff --git a/pages/instances/troubleshooting/fix-unreachable-focal-with-two-public-ips.mdx b/pages/instances/troubleshooting/fix-unreachable-focal-with-two-public-ips.mdx index 35bf643c7c..c06dbb18a5 100644 --- a/pages/instances/troubleshooting/fix-unreachable-focal-with-two-public-ips.mdx +++ b/pages/instances/troubleshooting/fix-unreachable-focal-with-two-public-ips.mdx @@ -16,7 +16,7 @@ The version of `systemd-networkd` used in Ubuntu Focal does not support the defi A modified `cloud-init` package named `cloud-init_24.2-0ubuntu1~20.04.1+scaleway` has been added to Scaleway's stable PPA to work around this situation. - This guide addresses specific issues related to Ubuntu Focal Instances with multiple public IPs. For general information on routed IPs and migration procedures, refer to our [main migration guide](/instances/how-to/migrate-routed-ips/) and the [related FAQ](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature). + This guide addresses specific issues related to Ubuntu Focal Instances with multiple public IPs. For general information on routed IPs and migration procedures, refer to the [related FAQ](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature). ## Your Instance has not been rebooted diff --git a/pages/instances/troubleshooting/fix-unreachable-ipv6-rhel-based-instance.mdx b/pages/instances/troubleshooting/fix-unreachable-ipv6-rhel-based-instance.mdx index 0b52537123..d93463b727 100644 --- a/pages/instances/troubleshooting/fix-unreachable-ipv6-rhel-based-instance.mdx +++ b/pages/instances/troubleshooting/fix-unreachable-ipv6-rhel-based-instance.mdx @@ -25,7 +25,7 @@ The affected O/S distributions are: This guide addresses specific issues related to RHEL-based Instances after transitioning to routed IP. -For general information on routed IPs and migration procedures, refer to our [main migration guide](/instances/how-to/migrate-routed-ips/) and the [related FAQ](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature). +For general information on routed IPs and migration procedures, refer to the [related FAQ](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature). ## Cause @@ -99,7 +99,7 @@ Activate the web console with: systemctl enable --now cockpit.socket ``` ## Going further -For general information on routed IPs and migration procedures, refer to our [main migration guide](/instances/how-to/migrate-routed-ips/) and the [related FAQ](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature). If you still encounter issues, please provide the following information when creating a support ticket: +For general information on routed IPs and migration procedures, refer to the [related FAQ](/instances/faq/#are-there-any-limitations-on-ip-autoconfiguration-with-the-routed-ip-feature). If you still encounter issues, please provide the following information when creating a support ticket: - Instance ID - Operating System version - Current IP configuration