Dallas EX9 Outage

Incident Report for PloxHost

Postmortem

To our customers.

With our node back up and running, we are now ready to address the extended service disruption that impacted our Extreme EX59 node in our Dallas location over the past few days. We fully recognize the frustration, inconvenience, and downtime this caused for our clients, partners, and communities, and we take full responsibility for what happened.

The incident began with a sudden loss of network connectivity on the server interface. What initially appeared to be a minor cabling or network card issue evolved into a prolonged outage caused by a physical hardware failure on the primary switch port within our facility. This was further compounded by delays in on-site diagnostic protocols, which prevented us from isolating and resolving the issue at the speed that our operational standards, and you, demand.

To permanently resolve the situation, we made the decision to bypass further port diagnostics and completely migrate the node’s infrastructure to dedicated Ryzen 9 hardware on a verified, clean switch port. This migration successfully restored all services, preserved data across the NVMe drives, and brought network routing back to optimal operating conditions.

This event has served as a critical wake-up call for our team, driving us to implement deep structural improvements. We are directly revising our critical priority support protocols with facility management in Dallas to eliminate any procedural bottlenecks during emergencies.

We offer our most sincere apologies to all impacted users, partners, and clients. We deeply appreciate the patience and continued trust you have shown us throughout these critical hours, and we remain fully committed to delivering the stable, reliable, and transparent infrastructure you deserve.

Posted Oct 29, 2026 - 05:48 UTC

Resolved

This incident has been resolved.
Posted Oct 29, 2026 - 05:44 UTC

Update

We are continuing to monitor for any further issues.
Posted Oct 29, 2026 - 02:48 UTC

Monitoring

The node is back up and running.

We are going to be monitoring the node closely; we will have an RCA and post-mortem soon,.
Posted Oct 29, 2026 - 02:48 UTC

Identified

We want to sincerely apologize to all affected partners, clients, and communities for the extended downtime on node EX59. We fully understand the frustration this delay has caused, and we take complete responsibility for the impact on your services.

Our main drive has been successfully migrated to high-performance hardware. All data, boot systems, and internal configurations are verified healthy and completely intact.

The remaining blocker is an upstream network routing update at the data center level to link our IP block (66.11.123.x) to the new server chassis.

We are actively in direct contact with data center management to complete this routing update. Once applied on their switches, all hosted services and game instances will automatically recover and go live.
Posted Oct 26, 2026 - 21:19 UTC

Update

We have successfully established a direct remote console via KVM to our node EX59. Our engineering team is now actively accessing the system to resolve the boot issue and restore all services. Further updates will be provided shortly.
Posted Oct 25, 2026 - 19:51 UTC

Update

Our datacenter techs configured a KVM overnight; we are waiting for confirmation from our techs to start working on a fix.

SLA will be available upon request once we fix this issue.
Posted Oct 24, 2026 - 16:52 UTC

Update

We have reached out to our data center technicians to try and set up a connection to the node.

We will provide an update promptly.
Posted Oct 24, 2026 - 01:58 UTC

Update

We continue to investigate the issue; we thank you for your patience while we locate the root cause.
Posted Oct 24, 2026 - 01:06 UTC

Investigating

Our team was alerted of our extreme node, us-dal-mc-ex59.ploxhost.com, being unresponsive. We are currently investigating the root cause, and we'll provide an update as soon as we have one.

We appreciate your patience while we look into this.
Posted Oct 24, 2026 - 00:41 UTC
This incident affected: Datacenter Operations: Dallas, Texas (Minecraft Hosting: Dallas, Texas).