diff --git a/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/guide.en-gb.md b/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/guide.en-gb.md new file mode 100644 index 00000000000..5bfcb46f5c9 --- /dev/null +++ b/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/guide.en-gb.md @@ -0,0 +1,56 @@ +--- +title: Improving Network Resilience on Bare metal servers +excerpt: Discover how to reinforce the resilience of public and private interfaces on your OVHcloud Bare metal servers with the new network aggregation architecture +updated: 2025-10-07 +--- + +> [!primary] +> +> This document is related to a deployment planned for **October 31st, 2025**, which will change the way link aggregation (LACP) functions within OVHcloud infrastructures. +> OVHcloud customers affected by these changes will also receive a communication by e-mail. + +## Objective + +This guide presents the new network aggregation architecture implemented on OVHcloud Bare metal servers. This evolution aims to strengthen fault tolerance and improve service continuity for your critical environments. + +## Requirements + +- A [dedicated server](/links/bare-metal/bare-metal). +- Have configured link aggregation (LACP) on your public or private interfaces (excluding OLA). + +## Instructions + +### What's Changing + +Until now, the aggregation (LACP) of network interfaces was performed on ports belonging to the **same Network Interface Card (NIC)**. This configuration already ensured redundancy in case of a ToR switch failure but did not cover the risk of a NIC failure. + +From now on: + +- Logical aggregations will be spread across **two distinct network cards** (with no physical cabling modification). +- Existing aggregations are not modified. + +![Resilience comparison before/after](images/resilience_comparison.png){.thumbnail} + +If you do not use LACP link aggregation on your server, **no action is required** and **no change will be visible for your services**. + +### How to Apply the Change on Your Servers + +When reconfiguring aggregations, the new rule will be applied automatically. To activate the new rule, you can also switch to **OLA** mode, then return to the default mode. +Once the new rule is activated, it will no longer be possible to revert to the old configuration. Servers delivered after the deployment date will directly benefit from this new rule. + +> [!warning] +> For the changes to take effect, you must **reconfigure the declared MAC addresses** in your operating system. +> +> In case of incorrect configuration in the OS, resilience might not be effective + +### Benefits + +Subject to correct configuration on the OS side, this development provides: + +- **Enhanced Availability**: better tolerance to hardware failures (network cards, switches, connectivity). +- **Uninterrupted Connectivity**: your services remain accessible even in the event of a network card failure. +- **Transparent Evolution**: no modification required for existing aggregations, apart from specific cases mentioned above. + +## Go Further + +Join our [user community](/links/community). diff --git a/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/guide.fr-fr.md b/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/guide.fr-fr.md new file mode 100644 index 00000000000..54045ccb013 --- /dev/null +++ b/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/guide.fr-fr.md @@ -0,0 +1,56 @@ +--- +title: Amélioration de la résilience réseau sur les serveurs Bare metal +excerpt: Découvrez comment renforcer la résilience des interfaces publiques et privées sur vos serveurs Bare Metal OVHcloud avec la nouvelle architecture d'agrégation réseau +updated: 2025-09-19 +--- + +> [!primary] +> Ce document concerne un déploiement prévu le **31 octobre 2025**, qui changera la manière dont fonctionne l'aggrégation de liens (LACP) au sein des infrastructures OVHcloud. +> Les clients OVHcloud concernés par ces changements recevront également une communication par mail. + +## Objectif + +Ce guide présente la nouvelle architecture d’agrégation réseau mise en place sur les serveurs Bare metal OVHcloud. Cette évolution a pour but de renforcer la tolérance aux pannes et d’améliorer la continuité de service pour vos environnements critiques. + +## Prérequis + +- Posséder un [serveur dédié](/links/bare-metal/bare-metal); +- Avoir configuré l’agrégation de liens (LACP) sur vos interfaces publiques ou privées (hors OLA). + +## En pratique + +### Ce qui change + +Jusqu’à présent, l’agrégation (LACP) des interfaces réseau était réalisée sur des ports appartenant à la **même carte réseau (NIC)**. Cette configuration assurait déjà une redondance en cas de défaillance d’un switch ToR, mais ne couvrait pas le risque de panne d’une carte réseau. + +Désormais : + +- Les agrégations logiques seront réparties sur **deux cartes réseau distinctes** (sans modification physique du câblage). +- Les agrégations existantes ne sont pas modifiées. + +![Resilience comparison before/after](images/resilience_comparison.png){.thumbnail} + +Si vous n’utilisez pas l’agrégation de liens LACP sur votre serveur, **aucune action n’est requise** et **aucun changement ne sera visible pour vos services**. + +### Comment effectuer le changement sur vos serveurs + +Lors d’une reconfiguration des agrégations, la nouvelle règle sera appliquée automatiquement. Pour activer la nouvelle règle, vous pouvez également effectuer un passage en mode **OLA**, puis un retour en mode par défaut. + +Une fois la nouvelle règle activée, il ne sera plus possible de revenir à l’ancienne configuration. Les serveurs livrés après la date de déploiement bénéficieront directement de cette nouvelle règle. + +> [!warning] +> Afin que les modifications soient effectives, vous devez **reconfigurer les adresses MAC déclarées** dans votre système d’exploitation. +> +> En cas de configuration incorrecte dans l’OS, la résilience pourrait ne pas être effective. + +### Bénéfices + +Sous réserve d’une configuration correcte côté OS, cette évolution permet : + +- **Disponibilité renforcée** : meilleure tolérance aux pannes matérielles (cartes réseau, switches, connectique). +- **Connectivité ininterrompue** : vos services restent accessibles même en cas de défaillance d’une carte réseau. +- **Évolution transparente** : aucune modification requise pour les agrégations existantes, hors cas spécifiques mentionnés ci-dessus. + +## Aller plus loin + +Rejoignez notre [communauté d'utilisateurs](/links/community). diff --git a/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/images/resilience_comparison.png b/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/images/resilience_comparison.png new file mode 100644 index 00000000000..ebba1f31bb8 Binary files /dev/null and b/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/images/resilience_comparison.png differ diff --git a/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/meta.yaml b/pages/bare_metal_cloud/dedicated_servers/lacp-resilience-scale-hg/meta.yaml new file mode 100644 index 00000000000..e69de29bb2d diff --git a/pages/bare_metal_cloud/dedicated_servers/storage-wipe-nist-800-88/guide.en-gb.md b/pages/bare_metal_cloud/dedicated_servers/storage-wipe-nist-800-88/guide.en-gb.md new file mode 100644 index 00000000000..8016f7a8ebc --- /dev/null +++ b/pages/bare_metal_cloud/dedicated_servers/storage-wipe-nist-800-88/guide.en-gb.md @@ -0,0 +1,91 @@ +--- +title: How to erase your data on a Dedicated Server +excerpt: Learn how to securely erase your data on an OVHcloud Dedicated Server in compliance with NIST SP 800-88 Revision 1. +updated: 2025-10-07 +--- + +## Objective + +This guide explains how to securely wipe your data from OVHcloud Dedicated Servers in accordance with [NIST Special Publication 800-88 Revision 1](https://csrc.nist.gov/pubs/sp/800/88/r1/final). + +Wiping ensures that your data cannot be recovered when you reinstall, transfer, or decommission your server. + +> [!warning] +> Once the data has been wiped, **it cannot be recovered**. Always back up any information you wish to keep before proceeding. +> + +## Requirements + +- A [dedicated server](/links/bare-metal/bare-metal) in your OVHcloud account with administrative access +- Access to the [OVHcloud Control Panel](/links/manager) + +## Concepts from NIST 800-88r1 + +- **Clear**: overwrites storage using the standard interface (for example, through a reinstallation). +- **Purge**: more thorough sanitization that protects against advanced forensic recovery (for example, secure erase on SSD/NVMe). +- **Destroy**: physical destruction of the media. + +On Bare Metal Cloud, you will most often use **Clear** (via reinstallation) or **Purge** (via secure erase in rescue mode). + +## Instructions + +### Method 1 — Reinstallation (Clear) + +1. Log in to the [OVHcloud Control Panel](/links/manager). +2. Go to `Bare Metal Cloud` > `Servers`. +3. Select your server. +4. Click **Reinstall**. +5. Choose your operating system or template. +6. Enable the **Erase existing partitions** option. +7. Confirm the reinstallation. + +This overwrites existing partitions and user-accessible data (*Clear*). + +### Method 2 — Secure Erase (Purge) + +if you require a more thorough deletion of data: + +1\. Boot your server into **Rescue Mode** from the Control Panel. You can find out how to access Rescue Mode in [this guide](/pages/bare_metal_cloud/dedicated_servers/rescue_mode). + +2\. Connect to the server via SSH, or using the Remote KVM feature. + +3\. Identify your disks with the following command: + +```bash +lsblk +``` + +4\. Run a secure erase command depending on the disk type: + +**For HDD/SSD:** +```bash +shred -vzn 3 /dev/sdX +``` + +This command uses several arguments: + + - *-v* : verbose, increases the detail in the output. + - *-z* : adds a final overwrite with all zeros. + - *-n 3* : overwrite three times (default). Can be set to 1 (faster, less secure) or up to 25 (maximum number of overwrite patterns). + +**For NVMe:** + +```bash +nvme format /dev/nvme0n1 --verbose --ses=1 +``` + +This command uses several arguments: + + - *--verbose* : increases the detail in the output. + - *--ses=1* : performs a User Data Erase, overwriting the data with a fixed pattern determined by the storage device's firmware (e.g., all zeros). Can be set to 2 to perform a Cryptographic Erase instead, which deletes the encryption key, effectively rendering the data inaccessible. + +5\. Exit rescue mode and reboot the server. + +## Verification + +- After reinstallation, connect to the server to ensure the system boots with a fresh OS and no old data remains. +- After secure erase, check the disk with `lsblk` — the drive should appear unallocated or with a new filesystem. + +## Go further + +Join our [community of users](/links/community). diff --git a/pages/index.md b/pages/index.md index 1cc4a665581..567b9f83af4 100644 --- a/pages/index.md +++ b/pages/index.md @@ -235,6 +235,7 @@ + [Configuring the network on Windows Server with Hyper-V](bare_metal_cloud/dedicated_servers/hyperv-network-HG-Scale) + [Configuring the network on Proxmox VE](bare_metal_cloud/dedicated_servers/proxmox-network-HG-Scale) + [How to configure Link Aggregation with LACP in Debian 12 or Ubuntu 24.04](bare_metal_cloud/dedicated_servers/lacp-enable-netplan) + + [Improving Network Resilience on Baremetal Servers](bare_metal_cloud\dedicated_servers\lacp-resilience-scale-hg) + [Configuring OVHcloud Link Aggregation in the OVHcloud Control Panel](bare_metal_cloud/dedicated_servers/ola-enable-manager) + [How to Configure Your NIC for OVHcloud Link Aggregation in Debian 9](bare_metal_cloud/dedicated_servers/ola-enable-debian9) + [How to Configure Your NIC for OVHcloud Link Aggregation in Windows Server 2019](bare_metal_cloud/dedicated_servers/ola-enable-w2k19)