We will make a small network change to update the MTU between a number of our core devices to improve network performance.
This will likely cause a brief blip as the change is applied.
We will update this post as the work completes
[Update 18/2/23]
This work completed just after 10pm yesterday evening and was a sucess
We are planning a window to extend our network to two new Data Centres (Cloud House and Slough
There should be no impact on the services, however this is classed as an “At Risk” maintenance window, due to the nature of the works on core network infrastructure.
We will update this post as the work progresses and completes
[Update 21/2/23]
This wok is completed with no impact
We have had an alert that the UPS in one of our racks is failing – this is discrete UPS as a backup in one of our core cabinets and not the main UPS. We will investigate this and swap out as needed.
Most equipment in the rack is dual fed so there will be no impact. However a few devices are single fed so there may be a small number of devices that briefly power down if the UPS needs swapping out
We plan to do this today as while these is short notice we feel it better to work at the weekend rather than a change or possible issue on a week day.
We will update this post as we are on site and investigate
[Update 13/2/23]
This work completed yesterday the faulty UPS was swapped out and replaced. There was a short downtime to a number of services in Huntingdon as one switch that was used to interconnect several racks was fed from the UPS that needed to be swapped
We are making a small change to the config of our new LNS servers for broadband services
This may cause a very brief drop on some PPPoE sessions to us – and your connection may drop. This drop should last more than a few seconds while the router re-connects.
We will update this page as the work progresses
[update]
The main config change here has been completed – the only noticeable difference for customers is they may see a different upstream IP on a traceroute in some cases). Some sessions did reset as predicted but we saw all sessions that dropped re-connect. If you do see any issues please reboot your router and if that fails contact support the normal way
Following the advice of Juniper we are deploying an upgrade to the Juniper Software to our core switch in telehouse North and East and rebooting to clear an at risk issue on the switch.
Expected downtime per site is approx 10 minutes – we will update the sites one at a time.
While the switches are working find now we the upgrade clears a known issue that can cause instability which we have seen in the Lab.
Through a customer report we have become aware of an issue with a MPLS link across our network. To resolve this we need to change the MTU across some of our core links.
We need to make this change this evening to fully restore service on the link. We will make the change link at a time – Due to the resilience built into our network we don’t anticipate any issue on our network – However with any config change there is a risk of a small number of very brief dropouts or you may see slightly different routing between sites.
We will update this as the work completes
As we have previously announced due to the closure of Equinix Reynolds House in Manchester we have moved our Manchester PoP to a new location
This Friday our carrier is moving our 10G Link from Huntingdon to Manchester to the new PoP in Manchester
Through this we will be using our primary link to London so we do not expect any issues – However for a short period we will be without redundancy / backup to Huntingdon
Note in this case the work is through the daytime – as his means additional engineering resources are available to ensure the change is smooth and without issues
We will update this post as the work progresses and completes
[Completed]
Virgin Media completed the circuit move Friday evening – we now have moved all links back to our net resilient network core.
Following on from the switch replacement in London Harbour Exchange square recently the vendor had advised us of a potential fan issue on the switch supplier and have supplied is with a second replacement switch,
We will swap this switch (again) on Saturday evening to ensure that all fans etc are fully working
While we have allowed a four how window here we expect the outage to be no longer than 10-20 minutes all being well.The new switch is all pre-configured so this is a simple chassis swap
We apologise for the extra work here but we would prefer to ensure there’s no potential unplanned issue here
This work was completed at 22:16 on Saturday with an outage of less than 10 minutes
We have been infirmed that Virgin media are working on one of the 10G Links to our Huntingdon Data Centre
Planned Work Start (Local UK): 01/09/2022 00:01
Planned Work End (Local UK): 01/09/2022 06:01Impact: There will be a loss of service at some point during the timeframe stated above whilst we complete the maintenance work.
The service outage should only last for up to 4 hours; however a 6 hour timeframe has been given in the event of any unforeseen circumstances.
This Emergency planned work is required to replace a hard spliced fibre patch at our site in Northampton that is causing degraded services
Corresponding Fault ticket reference = F010066288.
We do have resilient routes to the data centre however the data centre should be deemed to be at risk during these works
Virgin Media have some planned work over the next few days which may affect one of the two core links to our Huntingdon Data Centre. We have spent the last few weeks improving and strengthening our network with a new EVPN based core.
We had planned a full failover test to be announced for late next week – however with this planned work – we have decided to run at least one part of the test later today.
In theory this will only leave a drop of a few seconds as traffic will re-route – However there’s a risk if there’s any issues some connectivity to Huntingdon MAY be lost while we troubleshoot any issues. We will have full control of this and can easily restore service if we need more time to troubleshoot
We are sorry for the short notice here but we’d rather be prepared and happy that failover works exactly as it is designed – as opposed to finding this out in a real issue on a link.
We will update as this work progresses / completes