Posted: 14th Feb, 2012 By: MarkJ
Customers of internet providers Be Broadband and O2 UK appear to have been hit by a serious
peering problem that causes intermittent
difficulty with gaining access to websites. The issue results in some pages being either incredibly slow or completely failing to come up on the first load (subsequent refresh attempts will often bring the site up).
Unfortunately the difficulties, which appear to have been growing over the past couple of weeks, are nothing new to some of BE and O2's customers. Most recently it seems to have gotten much worse and is now causing quite a bit of frustration among those affected.
Mark Nichols, BE's Head of Marketing, told ISPreview.co.uk:"We're aware that some of our customers have recently been experiencing slower-than-normal speeds on their broadband service and had difficulty in accessing certain websites.
Over the weekend we resolved an issue with a section of our connection to the LINX peering platform which has made an improvement but we will continue to work on this until all [is] resolved. Customers can monitor the latest network status at our website (
BE Network Status) and we're sorry for any inconvenience this is causing until the fault is resolved."
BE's official service status page was updated yesterday afternoon to say: "
The issue is caused by [a] piece of equipment which will be replaced as soon as possible". In the meantime BE's unbundled ( LLU ) broadband network, which is also used by O2 UK, will continue to suffer from "
latency, slow browsing and packet loss". No firm ETA for a fix has been set.
Last week BE confirmed that it would begin migrating customers over to its
new core network by the end of February 2012 (
here). The major upgrade, which will not complete until early 2013, is designed to improve the ISPs capacity, resiliency and support new features. Hopefully it won't suffer from the same peering problems as their current platform.
UPDATE 15th Feb 2012This morning's latest status update from BE suggests that the problem has been identified and should now be fixed.
BE Broadband Status Update
We have received numerous complaints of packet loss and high latency. The issue is caused by a malfunction of one of our core network routers. Engineers are dispatched and dealing with the said equipment. UPDATE: As per the latest information, the faulty equipment will be replaced during the night and the service should be restored back to normal by 2 - 3 AM.
At the time of writing, about 6am, no change to the status has been made (i.e. no word on if this fixed it).
UPDATE 15th Feb 2012 @ 1:40pmThe game of musical chairs that is
BE's Service Status page, which frequently switches the problem from being marked as "
resolved" to "
not resolved" and back again, appears to be continuing. Despite BE's efforts their customers are continuing to report problems and the situation remains on-going.
We've noted that BT's overall satisfaction rating, such as in the
recent survey by Which? magazine, has seen some worrying declines of late. Problems that continue on for days, weeks or even months probably aren't going to help reduce O2 and BE's overall level of customer churn.
UPDATE 16th Feb 2012BE now claims that, as some customers were suggesting awhile ago, the, "
issue has been traced to the network of one of our peering providers. A case with them has been raised." Still no solution. Meanwhile, over on the
BE Usergroup, some customers have found that the problem can be fixed by swapping between static and dynamic IP addresses.
UPDATE 17th Feb 2012Customers appear to be reporting that the "
the situation with the packet loss is a lot better" (BE's Status Quote), which follows the ISPs attempts yesterday to chase the problem with a peering provider. However the difficulty has still not been completely resolved and remains an open issue.
UPDATE 18th Feb 2012BE claims to have finally resolved the problem and has even issued a wider public statement to explain a bit more about what has been happening.
BE Statement
Over the course of the last week, there have been a number of problems across the BE network, resulting in significant packet loss for a large number of customers and preventing certain web sites from being accessed.
We apologise for the inconvenience and difficulty this caused. Although we undertook extensive investigations to identify the root cause of the problem, a number of conflicting test results made it difficult to pin point the problem as quickly as we would normally expect.
We can now report that the issues, principally hardware faults, have now been resolved and service across the network has returned to normal levels.
Over the coming days, we’ll be working to better understand how these problems occurred and any lessons we can learn for the future.
Some reports indicate that a few customers might still be suffering from problems, such as when attempting to access the BBC's website. Needless to say that we'll be keeping an eye on this.
UPDATE 1st March 2012BE has finally admitted what most people already knew, that the BBC problems are related to a lack of capacity and heavy load from customers using iPlayer video streaming.
BE Statement:
"Some BE customers are continuing to experience problems when running BBC’s iPlayer. This is being caused by our link to the [Akamai] network, through which BE customers access BBC’s video services, operating close to capacity during periods of heavy demand. Although we are increasing the capacity of the link to resolve this problem, the earliest date this can be done is end-April.
Obviously, we’ll try and get this done earlier if at all possible, and in the meantime we’re working with other peering platforms to route overflow traffic away from Akamai during peak usage until the upgrade is completed; from the feedback we’ve already received this approach has improved the performance of iPlayer for those customers who’ve been affected."