On Sunday night we will be moving our authoritative DNS generation and synchronization onto a new dedicated server.  As part of this work we will be reconfiguring the DNS software on all of our authoritative DNS servers, and as such each server may suffer brief service interruptions, however we will keep such downtime to a minimum and ensure that only one server is affected at any given time, as such there should be very little impact to customers.

This work will affect all hosted domains, using primary and secondary DNS, on our own nameservers (ns0/1/2.gradwell.net) as well as any domains customers are hosting on dedicated nameserver addresses.  This work will also affect the gradwell.net domain and so there is potential for minor disruption across all services.

Additionally, during the maintenance, customers may find that DNS updates from the control panel and our DynDNS service may not take effect immediately.

Please accept our apologies for any inconvenience this work may cause, however it is essential maintenance to improve the resilience of our DNS service.

Update: This work was completed at approximately 23:30 on Sunday night with minimal service disruption.

Newred will be unavailable for two separate periods on Wednesday evening whilst we replace a faulty hard drive and rebuild the RAID array.  During this time, changes to DNS will be delayed, dynamic DNS may be unreliable, email collection from remote servers will be unavailable, and customers won’t be able to log into this server via SSH and FTP.

We apologise for the short notice for this maintenance.

Update 18:30 - This issue has now been resolved and an internal issue raised to ensure this does not occur again.  We have identified a single point of failure in our zone configuration and our server team will be working to provision backup.

Update 18:55 - One of our inbound telcos for ported numbers have cached a copy of our resource records which is incomplete.  This is causing inbound calls from this provider towards our network to fail.  We are awaiting an emergency call out from their out of hours engineers to resolve this and will update customers as soon as we have any further news.

Update: 19:30 - A technican at our inbound telco for ported numbers has restarted their DNS cluster to removed the affected DNS zone.  This has resolved the issue for inbound numbers failing.

We apologise for the extended nature of this problem, which is due in part to DNS caches not all adhering to the maximum lifespan we would like to give them.  This is a problem built into the internet by design, however, we should not have released the non-complete zone which triggered this to happen.  As a result, we will be changing the way our DNS servers are configured to try to prevent this issue from happening in the future.