We have been informed that Virgin Media will be performing a re-route on one of the backhaul links between London and our Huntingdon Data Centre between the 12/1/2022 and 13/1/2022
We have resilient links to provide service here however this will reduce the resilience for services in Huntingdon and also reduces resilience to Manchester
The planned work has a 10 hour window hpwever we don’t anticipate the link being down that long.
We will update this post if we see any impact and once the line is restored.
We are seeing some VLANs on our core network flap when they pass through a core switch in Telehouse West.
We may need to reset or reload this switch to clear this issue. This MAY cause some instability to the network for a short period and services hosted in Telehouse West may drop tor 15-20 minutes
We will update this as investigations continue
[update 1 – 23:30]
We have reloaded one of our core Juniper switches and are monitoring for stability.
[Update 24/12 9:44]
We have monitored the network after the switch reload at 11pm last night – and this has stayed stable with no further flaps. While the switch had no error logs during the incident it was randomly blocking traffic on some VLANs including a backup vlan between two core routers in Telehouse East and North. This was causing some packet loss / drops as the traffic was switching between links. We apologise for the issues seen however while we are aware this issue started during the day we needed to perform the reboot out of hours since it caused a 10-15 minute outage to services directly connected.
We have an issue with the above router that is causing some issues on our monitoring – and with a couple of internal BGP sessions being down.
We are planning to upgrade and resolve this during a maintenance this evening.
Given the redundancy we now have we do not anticipate this work being noticeable – however it does reduce slightly the resilience within our network while the work is performed. We intend to start this work approx 9pm and it should be completed within approx 1 hour – thought we have a window until midnight just in case.
We will update this post once the work completes
Following on from the packet loss seen earlier today our vendor has suggested an upgrade to be actioned to help resolve the issue we saw.
Given the fault we have decided to start the upgrade process later this evening. This will be approx 9pm – 10pm. As this will be in a controlled manner this should have minimal – if any impact to customer services but there may be brief periods where traffic takes alternative routes to some destinations. While we would prefer to provide more notiece however given the issues seen we would rather ensure this is completed so any follow up work can progress quickly.
We will update this ticket later this evening with an update on the progress
We will be installing additional core routers in out Harbour Exchange Data Centre on Saturday.
During most of the work there will be no impact to any services – However there may be brief periods during the day when traffic takes slightly different routes than normal. In addition we will be upgrading one of our two LINX peering connections over the weekend (the first was upgraded yesterday). As part of this the link being upgraded will be shut down this evening and will be re-enabled over the weekend. Given the other link is now upgraded there will be no traffic congestion but you may see slightly different routes to some locations.
We will update this as the work progresses
We are seeing a number of MAC address learning issues flagged on the logs of one of our core Juniper switches in the Huntingdon Data Centre
This is currently not service affecting – However we would like to clear this before it causes an issue. As such we will upgrade the firmware on the switch and reboot it later this evening.
This will affect access to some services in the Huntingdon Data Centre and some leased lines which are patched via it.
We expect the outage to be no longer than 20 minutes and we do have spare hardware etc on site should anything not go as planned.
We will update this post as the work progresses
We apologise for any issues that you may have as a result
This work was completed we did have an issue with accessing the switch during the upgrade but service disruption to customers was minimal.