Update - 25% of VMs have already been rebooted. You can check your system's uptime: if it was rebooted (by dataforest) recently, the maintenance for this VM has been completed. If not, the reboot is still pending.
Oct 17, 2026 - 22:34 CEST
Oct 17, 2026 - 22:34 CEST
In progress - Scheduled maintenance is currently in progress. We will provide updates as necessary.
Oct 17, 2026 - 16:30 CEST
Oct 17, 2026 - 16:30 CEST
Scheduled - A critical security vulnerability known as "PwnEscape" has been discovered in KVM, the virtualization technology powering our infrastructure. This flaw potentially affects all virtual servers across our infrastructure. Over the past 24 hours, our team has analyzed the threat and developed an immediate mitigation plan.
This is one of the most severe vulnerabilities reported in KVM to date. While there are currently no public exploits available to compromise host systems, we expect this to change imminently. To ensure the security of our infrastructure and our customers, we need to carry out this maintenance without unnecessary delays.
Affected products
This maintenance applies to every type of virtual server in our infrastructure, including:
- Seeds in the dataforest cloud
- Virtualized Avoro products (VPS, Rootserver, Power Rootserver, NVMe vServer, Compute vServer, Cloud Server)
- Every PHP-Friends vServer; both managed and unmanaged
- Our Plesk web hosting systems and TeamSpeak 3 servers (as these are hosted on the same virtualization platform)
What we are doing
Our engineers will work in multiple shifts to apply a security patch to all host systems, which requires a reboot of all virtual servers. The maintenance will begin in approximately 2 hours and is expected to take up to 48 hours. Given the scale of the operation, we are unfortunately unable to arrange individual maintenance windows. Should we need to adjust the schedule, we will update this post.
Expected impact
- We expect a brief downtime of 5 to 15 minutes per virtual server. If your server is unreachable for more than 15 minutes, please check this post for updates regarding potential host-specific issues. If there are no known issues listed, please contact our support.
- For unmanaged server customers: Ensure your VM is reboot-safe. Configure critical software to autostart on boot and terminate cleanly upon a shutdown signal from the host. Please understand that we cannot take responsibility for the internal configuration of unmanaged servers.
- For managed, webhosting, and TeamSpeak 3 customers: You do not need to take any action. Our team manages the configuration and safe reboot of your services automatically.
Oct 17, 2026 16:30 - Oct 19, 2026 16:30 CEST
This is one of the most severe vulnerabilities reported in KVM to date. While there are currently no public exploits available to compromise host systems, we expect this to change imminently. To ensure the security of our infrastructure and our customers, we need to carry out this maintenance without unnecessary delays.
Affected products
This maintenance applies to every type of virtual server in our infrastructure, including:
- Seeds in the dataforest cloud
- Virtualized Avoro products (VPS, Rootserver, Power Rootserver, NVMe vServer, Compute vServer, Cloud Server)
- Every PHP-Friends vServer; both managed and unmanaged
- Our Plesk web hosting systems and TeamSpeak 3 servers (as these are hosted on the same virtualization platform)
What we are doing
Our engineers will work in multiple shifts to apply a security patch to all host systems, which requires a reboot of all virtual servers. The maintenance will begin in approximately 2 hours and is expected to take up to 48 hours. Given the scale of the operation, we are unfortunately unable to arrange individual maintenance windows. Should we need to adjust the schedule, we will update this post.
Expected impact
- We expect a brief downtime of 5 to 15 minutes per virtual server. If your server is unreachable for more than 15 minutes, please check this post for updates regarding potential host-specific issues. If there are no known issues listed, please contact our support.
- For unmanaged server customers: Ensure your VM is reboot-safe. Configure critical software to autostart on boot and terminate cleanly upon a shutdown signal from the host. Please understand that we cannot take responsibility for the internal configuration of unmanaged servers.
- For managed, webhosting, and TeamSpeak 3 customers: You do not need to take any action. Our team manages the configuration and safe reboot of your services automatically.
Oct 17, 2026 16:30 - Oct 19, 2026 16:30 CEST
In progress - Scheduled maintenance is currently in progress. We will provide updates as necessary.
Oct 12, 2026 - 08:00 CEST
Oct 12, 2026 - 08:00 CEST
Scheduled - Start: 12.10.2026, 08:00 CEST
End: 23.10.2026, 17:00 CEST
Power infrastructure upgrades will be carried out at FC-FRA4 across two phases. The work is intended to improve redundancy and operational stability while providing additional capacity for future growth. No downtime is planned for connected systems.
From 12.10.2026 to 16.10.2026, the power feeds of distribution systems A and B will be modified to improve selectivity and operational safety. The work will be performed sequentially, with only one power path being worked on at a time. During each respective phase, the affected power path will be supplied by the datacenter's diesel generator system, while UPS protection remains active throughout generator operation.
As this phase involves switching and work on the active power infrastructure, a small residual risk remains. In the (unlikely and unexpected) event of a disruption affecting the respective path, some rack feeds will become unavailable.
From 19.10.2026 to 23.10.2026, additional electrical supply lines will be installed, tested and commissioned to expand the site's power infrastructure and provide additional capacity.
We will provide an update should any unexpected impact occur.
Oct 12, 2026 08:00 - Oct 23, 2026 17:00 CEST
End: 23.10.2026, 17:00 CEST
Power infrastructure upgrades will be carried out at FC-FRA4 across two phases. The work is intended to improve redundancy and operational stability while providing additional capacity for future growth. No downtime is planned for connected systems.
From 12.10.2026 to 16.10.2026, the power feeds of distribution systems A and B will be modified to improve selectivity and operational safety. The work will be performed sequentially, with only one power path being worked on at a time. During each respective phase, the affected power path will be supplied by the datacenter's diesel generator system, while UPS protection remains active throughout generator operation.
As this phase involves switching and work on the active power infrastructure, a small residual risk remains. In the (unlikely and unexpected) event of a disruption affecting the respective path, some rack feeds will become unavailable.
From 19.10.2026 to 23.10.2026, additional electrical supply lines will be installed, tested and commissioned to expand the site's power infrastructure and provide additional capacity.
We will provide an update should any unexpected impact occur.
Oct 12, 2026 08:00 - Oct 23, 2026 17:00 CEST
dataforest Network
Operational
Edge Network (affects all locations)
Operational
DDoS Protection (affects all locations)
Operational
Dedicated Servers (EQX-FR4 & DRT-FRA8)
Operational
Datacenter MC-FRA01
Under Maintenance
Datacenter Core Routing & Switching
Operational
Ceph Cluster
Operational
Virtual Servers
Under Maintenance
Dedicated Servers
Operational
Plesk Web Hosting
Under Maintenance
TeamSpeak3 Servers
Under Maintenance
Network Storage
Operational
Facilites
Operational
Datacenter FC-FRA4
Under Maintenance
Datacenter Core Routing & Switching
Operational
Dedicated Servers
Operational
Facilites
Under Maintenance
General Services
Operational
Avoro CP & Support
Operational
PHP-Friends CRM & Support
Operational
Hotline
Operational
Dedicated Control Panel
Operational
IPMI VPN
Operational
DDoS Manager
Operational
Operational
Degraded Performance
Partial Outage
Major Outage
Maintenance
Scheduled Maintenance
DDoS Protection Software Update Oct 19, 2026 15:15-16:15 CEST
Start: 19.10.2026, 15:15 CEST
End: 19.10.2026, 16:15 CEST
We will carry out maintenance on our DDoS protection infrastructure.
What we are doing
We will deploy software updates to our redundant scrubbing cluster. Filter appliances will be taken out of production and updated one by one to ensure mitigation capacity is always sufficient. After each update, functionality will be verified before we move on to the next filter.
Expected impact
No customer impact is expected, but it cannot be ruled out entirely. Idle (inactive) connections/sessions may time out; active ones should not, as they are synchronized during the startup of a new software release. If you notice anything unusual, such as timeouts of active sessions or the inability to connect to your server, please contact our support.
Affected customers
All IP prefixes routed through our DDoS protection (which is the default).
Fallback plan
Should the update fail and cause issues, we will disconnect the respective filters within a few seconds.
Posted on Oct 17, 2026 - 16:24 CEST
End: 19.10.2026, 16:15 CEST
We will carry out maintenance on our DDoS protection infrastructure.
What we are doing
We will deploy software updates to our redundant scrubbing cluster. Filter appliances will be taken out of production and updated one by one to ensure mitigation capacity is always sufficient. After each update, functionality will be verified before we move on to the next filter.
Expected impact
No customer impact is expected, but it cannot be ruled out entirely. Idle (inactive) connections/sessions may time out; active ones should not, as they are synchronized during the startup of a new software release. If you notice anything unusual, such as timeouts of active sessions or the inability to connect to your server, please contact our support.
Affected customers
All IP prefixes routed through our DDoS protection (which is the default).
Fallback plan
Should the update fail and cause issues, we will disconnect the respective filters within a few seconds.
Posted on Oct 17, 2026 - 16:24 CEST
[MC-FRA01] Software Update on Dedicated Server Rack Switches (BD09) Oct 22, 2026 00:00-01:00 CEST
We will update the software of the switches that provide network connectivity to the dedicated server rack(s) mentioned below at MC-FRA01 to resolve known bugs and ensure long-term stability. After the update, a reboot is required.
Expected impact
During the maintenance window, there will be a network interruption during each switch reboot. The reboot is expected to take up to 20 minutes per switch.
- Servers with only one uplink will lose network connectivity during that reboot. Connectivity should recover automatically once the switch is back online.
- Servers with two uplinks are designed to remain reachable during this maintenance, as one physical uplink will remain available while the other switch is rebooting.
- For servers with two uplinks, correct LACP configuration on the customer side is always required. Two physical uplinks only provide redundancy if they are configured as one LACP channel. If LACP is missing or incorrectly configured on the customer side, the server will become unreachable during the maintenance, even though redundant uplinks are physically available.
dataforest does not verify customer-side LACP configuration for unmanaged dedicated servers. Customers are responsible for ensuring that both provided uplinks are correctly configured and operational.
Affected rack(s)
BD09
How to check if you are affected
In our control panel at https://dedicated.dataforest.cloud, open the server and check "Statistics". If one of your ports shows a description containing one of the rack identifiers listed above, your server is part of the maintenance scope.
You can see whether your server has one or two uplinks in your contract or product details. If two uplinks are included, LACP must be configured and actively used on your side.
We will complete the work as quickly as possible and aim to keep the interruption to the minimum stated above. Thank you for your understanding.
Posted on Sep 29, 2026 - 17:06 CEST
Expected impact
During the maintenance window, there will be a network interruption during each switch reboot. The reboot is expected to take up to 20 minutes per switch.
- Servers with only one uplink will lose network connectivity during that reboot. Connectivity should recover automatically once the switch is back online.
- Servers with two uplinks are designed to remain reachable during this maintenance, as one physical uplink will remain available while the other switch is rebooting.
- For servers with two uplinks, correct LACP configuration on the customer side is always required. Two physical uplinks only provide redundancy if they are configured as one LACP channel. If LACP is missing or incorrectly configured on the customer side, the server will become unreachable during the maintenance, even though redundant uplinks are physically available.
dataforest does not verify customer-side LACP configuration for unmanaged dedicated servers. Customers are responsible for ensuring that both provided uplinks are correctly configured and operational.
Affected rack(s)
BD09
How to check if you are affected
In our control panel at https://dedicated.dataforest.cloud, open the server and check "Statistics". If one of your ports shows a description containing one of the rack identifiers listed above, your server is part of the maintenance scope.
You can see whether your server has one or two uplinks in your contract or product details. If two uplinks are included, LACP must be configured and actively used on your side.
We will complete the work as quickly as possible and aim to keep the interruption to the minimum stated above. Thank you for your understanding.
Posted on Sep 29, 2026 - 17:06 CEST