System instability xeon5,6,7,8 on 7 Jul 2026 14:04:04 (UTC+00:00)
Resolved on 8 Jul 2026 03:51:36 (UTC+00:00)
Resolved - 19:02 CEST
The incident has been marked as resolved because all affected systems are operational again. We will continue monitoring the situation to confirm everything remains stable. If your VM is still experiencing issues, please contact support.
Monitoring - 15:41 CEST
The fix has been applied, and the affected KVM servers are coming back online. We are monitoring the nodes to confirm everything remains stable.
Identified - 15:28 CEST
We identified the instability on xeon5, xeon6, xeon7, and xeon8 after the CPU type change. We are reverting that change and applying the new Proxmox kernel to resolve the issue.
Reported - 15:23 CEST
We detected instability on xeon5, xeon6, xeon7, and xeon8 after the CPU type change and started investigating immediately.
Resolved - 19:02 CEST
The incident has been marked as resolved because all affected systems are operational again. We will continue monitoring the situation to confirm everything remains stable. If your VM is still experiencing issues, please contact support.
Monitoring - 15:41 CEST
The fix has been applied, and the affected KVM servers are coming back online. We are monitoring the nodes to confirm everything remains stable.
Identified - 15:28 CEST
We identified the instability on xeon5, xeon6, xeon7, and xeon8 after the CPU type change. We are reverting that change and applying the new Proxmox kernel to resolve the issue.
Reported - 15:23 CEST
We detected instability on xeon5, xeon6, xeon7, and xeon8 after the CPU type change and started investigating immediately.
Monitoring on 7 Jul 2026 14:04:04 (UTC+00:00)
Monitoring - 15:41 CEST
The fix has been applied, and the affected KVM servers are coming back online. We are monitoring the nodes to confirm everything remains stable.
Identified - 15:28 CEST
We identified the instability on xeon5, xeon6, xeon7, and xeon8 after the CPU type change. We are reverting that change and applying the new Proxmox kernel to resolve the issue.
Reported - 15:23 CEST
We detected instability on xeon5, xeon6, xeon7, and xeon8 after the CPU type change and started investigating immediately.
Monitoring - 15:41 CEST
The fix has been applied, and the affected KVM servers are coming back online. We are monitoring the nodes to confirm everything remains stable.
Identified - 15:28 CEST
We identified the instability on xeon5, xeon6, xeon7, and xeon8 after the CPU type change. We are reverting that change and applying the new Proxmox kernel to resolve the issue.
Reported - 15:23 CEST
We detected instability on xeon5, xeon6, xeon7, and xeon8 after the CPU type change and started investigating immediately.
Emergency security mitigation on 7 Jul 2026 05:21:46 (UTC+00:00)
Update on 7 Jul 2026 05:22:17 (UTC+00:00)
The emergency maintenance across our KVM infrastructure has now been completed.
This maintenance was required because of a critical Linux KVM hypervisor vulnerability that was publicly disclosed. The issue could allow a guest virtual machine to crash the host system or, in more severe cases, escape from the guest environment to the host. You can read the technical disclosure here: https://thehackernews.com/2026/07/16-year-old-linux-kvm-flaw-lets-guest.html
Due to the severity of this vulnerability, we had to act quickly to protect customer data and the stability of our infrastructure. The work was carried out during the night to keep the impact on active services as low as possible. During the rollout, virtual machines were shut down gracefully where possible so operating systems could stop cleanly.
As part of the mitigation, the CPU type for affected virtual machines has been changed from "Host" to a standard compatible CPU profile. Nested Virtualization is currently disabled as a precaution. We are monitoring upstream kernel patches closely, and once a reliable patch is available and tested, we aim to make Nested Virtualization available again upon request.
We apologize for the short notice and for any disruption this emergency maintenance may have caused. If your server has not recovered as expected, or if you are experiencing any related issues, please open a support ticket and we will assist you as soon as possible.
The emergency maintenance across our KVM infrastructure has now been completed.
This maintenance was required because of a critical Linux KVM hypervisor vulnerability that was publicly disclosed. The issue could allow a guest virtual machine to crash the host system or, in more severe cases, escape from the guest environment to the host. You can read the technical disclosure here: https://thehackernews.com/2026/07/16-year-old-linux-kvm-flaw-lets-guest.html
Due to the severity of this vulnerability, we had to act quickly to protect customer data and the stability of our infrastructure. The work was carried out during the night to keep the impact on active services as low as possible. During the rollout, virtual machines were shut down gracefully where possible so operating systems could stop cleanly.
As part of the mitigation, the CPU type for affected virtual machines has been changed from "Host" to a standard compatible CPU profile. Nested Virtualization is currently disabled as a precaution. We are monitoring upstream kernel patches closely, and once a reliable patch is available and tested, we aim to make Nested Virtualization available again upon request.
We apologize for the short notice and for any disruption this emergency maintenance may have caused. If your server has not recovered as expected, or if you are experiencing any related issues, please open a support ticket and we will assist you as soon as possible.
Resolved on 7 Jul 2026 05:21:46 (UTC+00:00)
We performed a short reboot of the KVM servers during the night due to a required security mitigation on the hypervisor level.
The downtime was only around 30 seconds to 1 minute. This should be a single short interruption, and we do not expect any further downtime.
As part of the mitigation, we also need to change the CPU type of the virtual machines from "Host" to a standard compatible CPU profile. Nested Virtualization, running a VM inside a VM, will be disabled for now as a precaution.
If an official kernel patch becomes available later, we will apply it and can re-enable Nested Virtualization for your server on request.
We apologize for the short interruption and thank you for your understanding.
We performed a short reboot of the KVM servers during the night due to a required security mitigation on the hypervisor level.
The downtime was only around 30 seconds to 1 minute. This should be a single short interruption, and we do not expect any further downtime.
As part of the mitigation, we also need to change the CPU type of the virtual machines from "Host" to a standard compatible CPU profile. Nested Virtualization, running a VM inside a VM, will be disabled for now as a precaution.
If an official kernel patch becomes available later, we will apply it and can re-enable Nested Virtualization for your server on request.
We apologize for the short interruption and thank you for your understanding.
Issue Xeon1-20 on 14 May 2026 14:09:45 (UTC+00:00)
Resolved
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Update 14:36:
All nodes but xeon19 are working again.
Update 14:52:
Everything is now back up and running.
We are currently performing a thorough root cause analysis to determine the exact cause of the interruption, and we will provide further updates as soon as we have more information.
As compensation for the downtime, all affected services have been credited with two additional days of runtime.
If you are still experiencing any issues, please contact us through the ticket system, and our team will be happy to assist you.
We sincerely apologize for the inconvenience caused and appreciate your patience and understanding.
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Update 14:36:
All nodes but xeon19 are working again.
Update 14:52:
Everything is now back up and running.
We are currently performing a thorough root cause analysis to determine the exact cause of the interruption, and we will provide further updates as soon as we have more information.
As compensation for the downtime, all affected services have been credited with two additional days of runtime.
If you are still experiencing any issues, please contact us through the ticket system, and our team will be happy to assist you.
We sincerely apologize for the inconvenience caused and appreciate your patience and understanding.
Issue Xeon1-20 on 14 May 2026 14:09:42 (UTC+00:00)
Resolved
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Update 14:36:
All nodes but xeon19 are working again.
Update 14:52:
Everything is now back up and running.
We are currently performing a thorough root cause analysis to determine the exact cause of the interruption, and we will provide further updates as soon as we have more information.
As compensation for the downtime, all affected services have been credited with two additional days of runtime.
If you are still experiencing any issues, please contact us through the ticket system, and our team will be happy to assist you.
We sincerely apologize for the inconvenience caused and appreciate your patience and understanding.
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Update 14:36:
All nodes but xeon19 are working again.
Update 14:52:
Everything is now back up and running.
We are currently performing a thorough root cause analysis to determine the exact cause of the interruption, and we will provide further updates as soon as we have more information.
As compensation for the downtime, all affected services have been credited with two additional days of runtime.
If you are still experiencing any issues, please contact us through the ticket system, and our team will be happy to assist you.
We sincerely apologize for the inconvenience caused and appreciate your patience and understanding.
Issue Xeon1-20 on 14 May 2026 12:55:05 (UTC+00:00)
Investigating
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Update 14:36:
All nodes but xeon19 are working again.
Update 14:52:
Everything is now back up and running.
We are currently performing a thorough root cause analysis to determine the exact cause of the interruption, and we will provide further updates as soon as we have more information.
As compensation for the downtime, all affected services have been credited with two additional days of runtime.
If you are still experiencing any issues, please contact us through the ticket system, and our team will be happy to assist you.
We sincerely apologize for the inconvenience caused and appreciate your patience and understanding.
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Update 14:36:
All nodes but xeon19 are working again.
Update 14:52:
Everything is now back up and running.
We are currently performing a thorough root cause analysis to determine the exact cause of the interruption, and we will provide further updates as soon as we have more information.
As compensation for the downtime, all affected services have been credited with two additional days of runtime.
If you are still experiencing any issues, please contact us through the ticket system, and our team will be happy to assist you.
We sincerely apologize for the inconvenience caused and appreciate your patience and understanding.
Issue Xeon1-20 on 14 May 2026 12:55:02 (UTC+00:00)
Investigating
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Update 14:36:
All nodes but xeon19 are working again.
Update 14:52:
Everything is now back up and running.
We are currently performing a thorough root cause analysis to determine the exact cause of the interruption, and we will provide further updates as soon as we have more information.
As compensation for the downtime, all affected services have been credited with two additional days of runtime.
If you are still experiencing any issues, please contact us through the ticket system, and our team will be happy to assist you.
We sincerely apologize for the inconvenience caused and appreciate your patience and understanding.
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Update 14:36:
All nodes but xeon19 are working again.
Update 14:52:
Everything is now back up and running.
We are currently performing a thorough root cause analysis to determine the exact cause of the interruption, and we will provide further updates as soon as we have more information.
As compensation for the downtime, all affected services have been credited with two additional days of runtime.
If you are still experiencing any issues, please contact us through the ticket system, and our team will be happy to assist you.
We sincerely apologize for the inconvenience caused and appreciate your patience and understanding.
Issue Xeon1-20 on 14 May 2026 12:47:36 (UTC+00:00)
Investigating
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Update 14:36:
All nodes but xeon19 are working again.
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Update 14:36:
All nodes but xeon19 are working again.
Issue Xeon1-20 on 14 May 2026 11:06:19 (UTC+00:00)
Investigating
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
We are currently investigating an issue with xeon1-20.
Update 13:01:
Gold7 is up again. It was not reported as down before, but it was also affected. Xeon13, Xeon14, Xeon15, and Xeon16 are also up.
Issue Xeon1-20 on 14 May 2026 10:07:51 (UTC+00:00)
Investigating
We are currently investigating an issue with xeon1-20.
We are currently investigating an issue with xeon1-20.
Quick maintenance on 4 May 2026 06:14:50 (UTC+00:00)
Completed on 4 May 2026 06:17:22 (UTC+00:00)
Quick maintenance on luxvps.net & billing and other main sites. (The control panel is not affected.)
Should not last longer than 5-10 minutes
Quick maintenance on luxvps.net & billing and other main sites. (The control panel is not affected.)
Should not last longer than 5-10 minutes
In progress on 4 May 2026 06:14:50 (UTC+00:00)
Quick maintenance on luxvps.net & billing and other main sites. (The control panel is not affected.)
Should not last longer than 5-10 minutes
Quick maintenance on luxvps.net & billing and other main sites. (The control panel is not affected.)
Should not last longer than 5-10 minutes
Network outage EQ FR5 Xeon Cluster on 18 Apr 2026 08:58:48 (UTC+00:00)
Completed on 18 Apr 2026 10:03:35 (UTC+00:00)
There appears to be a network outage at the EQ FR5 Xeon Cluster site; only the Xeon KVM servers at this location are affected. The upstream provider has already been notified, and we are awaiting a response.
11:14:
Xeon1 to xeon20 are affected. Update from AS203446: It seems like the MX480 router in FR5 is down, Equinix will check the router.
11:39:
The downtime appears to have been resolved. Four systems were located in a rack that was experiencing power issues (Xeon 13, 14, 16). We are now restarting these systems, including all VMs. All other systems are accessible and only experienced connectivity issues.
Customer satisfaction is important to us. Therefore, we are offering compensation in the form of three days of runtime on the respective nodes that experienced issues. This runtime will be added to the services today.
There appears to be a network outage at the EQ FR5 Xeon Cluster site; only the Xeon KVM servers at this location are affected. The upstream provider has already been notified, and we are awaiting a response.
11:14:
Xeon1 to xeon20 are affected. Update from AS203446: It seems like the MX480 router in FR5 is down, Equinix will check the router.
11:39:
The downtime appears to have been resolved. Four systems were located in a rack that was experiencing power issues (Xeon 13, 14, 16). We are now restarting these systems, including all VMs. All other systems are accessible and only experienced connectivity issues.
Customer satisfaction is important to us. Therefore, we are offering compensation in the form of three days of runtime on the respective nodes that experienced issues. This runtime will be added to the services today.
Completed on 18 Apr 2026 10:02:15 (UTC+00:00)
There appears to be a network outage at the EQ FR5 Xeon Cluster site; only the Xeon KVM servers at this location are affected. The upstream provider has already been notified, and we are awaiting a response.
11:14:
Xeon1 to xeon20 are affected. Update from AS203446: It seems like the MX480 router in FR5 is down, Equinix will check the router.
Customer satisfaction is important to us. Therefore, we are offering compensation in the form of three days of runtime on the respective nodes that experienced issues. This runtime will be added to the services today.
There appears to be a network outage at the EQ FR5 Xeon Cluster site; only the Xeon KVM servers at this location are affected. The upstream provider has already been notified, and we are awaiting a response.
11:14:
Xeon1 to xeon20 are affected. Update from AS203446: It seems like the MX480 router in FR5 is down, Equinix will check the router.
Customer satisfaction is important to us. Therefore, we are offering compensation in the form of three days of runtime on the respective nodes that experienced issues. This runtime will be added to the services today.
Update on 18 Apr 2026 09:15:59 (UTC+00:00)
There appears to be a network outage at the EQ FR5 Xeon Cluster site; only the Xeon KVM servers at this location are affected. The upstream provider has already been notified, and we are awaiting a response.
11:14:
Xeon1 to xeon20 are affected. Update from AS203446: It seems like the MX480 router in FR5 is down, Equinix will check the router.
There appears to be a network outage at the EQ FR5 Xeon Cluster site; only the Xeon KVM servers at this location are affected. The upstream provider has already been notified, and we are awaiting a response.
11:14:
Xeon1 to xeon20 are affected. Update from AS203446: It seems like the MX480 router in FR5 is down, Equinix will check the router.
Investigating on 18 Apr 2026 08:58:48 (UTC+00:00)
There appears to be a network outage at the EQ FR5 Xeon Cluster site; only the Xeon KVM servers at this location are affected. The upstream provider has already been notified, and we are awaiting a response.
There appears to be a network outage at the EQ FR5 Xeon Cluster site; only the Xeon KVM servers at this location are affected. The upstream provider has already been notified, and we are awaiting a response.
Maintenance on epyc2 on 2 Dec 2025 06:01:34 (UTC+00:00)
Completed on 4 Dec 2025 11:06:56 (UTC+00:00)
Quick maintenance on epyc2 (including our main, billing site and webhosting customers)
from 03.12.2025 5:30PM to 7PM ((UTC+01:00) Europe/Berlin)
Quick maintenance on epyc2 (including our main, billing site and webhosting customers)
from 03.12.2025 5:30PM to 7PM ((UTC+01:00) Europe/Berlin)
Scheduled on 2 Dec 2025 06:01:34 (UTC+00:00)
Quick maintenance on epyc2 (including our main, billing site and webhosting customers)
from 03.12.2025 5:30PM to 7PM ((UTC+01:00) Europe/Berlin)
Quick maintenance on epyc2 (including our main, billing site and webhosting customers)
from 03.12.2025 5:30PM to 7PM ((UTC+01:00) Europe/Berlin)