Home
 » ISP News » 
Sponsored Links

Peak Time Latency Strikes Part of Virgin Media’s UK Network UPDATE2

Tuesday, Apr 12th, 2022 (4:13 pm) - Score 11,928
virgin media o2 network engineer at work 2021

Some broadband customers on UK ISP Virgin Media’s (VMO2) fixed line network are reporting a noticeable increase in peak time latency – usually between around 5:30pm and 11pm, which appears to have begun at around the end of last week. This can see latency times become very slow (double or more vs the usual baseline).

Latency is a measure of the time that it takes for a packet of data to travel from your computer to a remote server and then back again (ping). The delay is usually measured in milliseconds (e.g. 1000ms = 1 second) and modern broadband connections will often have an average latency of anything from around c.5ms to 40ms (what is normal for your connection will depend on lots of different factors – see below)

A lower score (shortest time) is always best for latency, although the times can be affected by various things, such as the performance of remote internet servers, the connection technology being used, network congestion at your ISP, peering / routing problems and the setup of your own home network etc.

Complaints about latency spikes and surges are of course nothing new in this market, which makes it difficult to spot whether something is reflective of a growing trend or merely related to an isolated incident in a particular part of the country. But since the weekend we have received a number of emails about this, and some seemingly related topics can be found on Virgin Media’s Community Forum (examples here, here and here).

Consumers don’t always notice problems with latency unless they’re using applications that are more likely to be impacted by it, such as while playing fast-paced online video games or making time sensitive live video calls or streaming broadcasts. Suffice to say, we have seen a small but noticeable up-tick in complaints from such groups, which has now been confirmed by data from Thinkbroadband’s network monitoring tools (here).

We contacted Virgin Media about this during the early afternoon and are still awaiting their response. Hopefully it will be resolved soon.

UPDATE 14th April 2022 @ 11:17pm

We’ve had an update from Virgin Media.

A Virgin Media spokesperson said:

“We’ve investigated the reports and found no issues with latency on our network. It appears the erroneous reports were the result of how traffic was being routed to ThinkBroadband’s monitoring tool, which has now been resolved.”

UPDATE 6th May 2022 @ 10:04am

After chasing, the operator has provided an updated statement.

A Virgin Media spokesperson said:

“We’ve now determined the root cause of the issue and can confirm that it’s an issue exclusively impacting traffic passing through an interconnect from our network to the ThinkBroadband server site.

General traffic on our network is not impacted and as a result, the TB BQM is not reporting accurate latency data for Virgin Media customers right now. Whilst we’re looking to address this, we recommend that customers use the SamKnows RealSpeed application or Ookla’s Speedtest.net’s ping test if they wish to check their latency or network performance.”

In response, Andrew Ferguson OBE of Thinkbroadband, said: “It is good that Virgin Media has identified the root cause but we are not happy with the spokesperson appearing to suggest that the BQM is not accurate, it most definitely is accurate and was reflecting the state of all the links between the BQM systems and the customers router. Similar with the speed test we have always run that in a state that will highlight speed issues for individuals rather than trying to counter congestion by opting for a closer server or adding more download/upload links to compensate.”

Share with Twitter
Share with Linkedin
Share with Facebook
Share with Reddit
Share with Pinterest
Tags:
Mark-Jackson
By Mark Jackson
Mark is a professional technology writer, IT consultant and computer engineer from Dorset (England), he also founded ISPreview in 1999 and enjoys analysing the latest telecoms and broadband developments. Find me on X (Twitter), Mastodon, Facebook and .
Search ISP News
Search ISP Listings
Search ISP Reviews
Comments
33 Responses
  1. Avatar photo M says:

    To be fair this is a bottleneck somewhere specifically along the BQM route. Obviously, it could affect other traffic too.

    However, on a related note, I have previously raised serious issues with routes to Cloudflare (including weeks/months of packet loss) and they’re *completely* ignored by Virgin despite providing evidence that it’s a nationwide issue. It’s very difficult to get directed to someone who can actually help with this kind of problem and that was probably quite a lot more impactful than what this appears to be.

    I am no longer a customer.

    1. Avatar photo AQX says:

      @M – There’s been issues with Cloudflare since before 2020, there was even a major forum post about it but never anything done.

    2. Avatar photo M says:

      I am probably the author of said post 😉

    3. Avatar photo CarlT says:

      It’s not even VMO2 ignoring it, it disappears into the abyss of Liberty Global, Schipol, where it goes to die.

      I diagnosed the issue and showed where on the network it was but it then goes off to LG who seem to consider nationwide degraded service a minor inconvenience to be dealt with when someone can be bothered with the 5 minutes to fix it.

      With that back to my hidey-hole.

    4. Avatar photo M says:

      @CarlT I’m not sure that’s accurate?

      The conversations I had with an engineer (don’t know his exact role) say the loss is happening at a peering router. The Cloudflare edge is London, suggesting it’s something within the UK that’s (or was, as I said, not a customer now, so I don’t know) at fault.

    5. Avatar photo CarlT says:

      I’m referring to the fault handling, not the traffic itself, M.

      5089 is supposed to be behind 6830, 6830 is managed from Schipol.

      This issue is an overloaded port on the LINX Juniper LAN. Liberty aren’t fans of settlement free peering, preferring transit or shaking other ASNs down to connect.

      Traffic to NCUK for example is going outbound via Aorta and NTT, inbound is direct to 5089 via LINX. The inbound isn’t what Liberty want, free interconnection to a smaller company, eww, and is resourced appropriately.

    6. Avatar photo Chris says:

      I was having issues the other week for ~ 4 days connecting to cloudflare based systems and google services like mail, chat and calendar.

      No such problems when using apples private relay.

      I eventually went online to report it, the vm check my line tool said there was something wrong with my connection, the following morning it was all sorted.

      Other colleagues on vm where not impacted, but they live in other parts of the country.

      At one point I could not get a ping response from 8.8.8.8 but 8.8.4.4 worked fine, I mainly use cloudflare DoH but I also noted huge problems with adguard dns DoH profile which has also resolved, I assume that transited cloudflare too.

    7. Avatar photo Chris says:

      @carlt,

      Liberty global runs an actual tier 1 network which defines them as not paying for peering.

      https://en.wikipedia.org/wiki/Tier_1_network#List_of_Tier_1_networks

      There has been lots of discussion over the decades about the open and free internet / net neutrality and that if large content providers where to pay for priority access that would hinder the minnows and hamper the net entrepreneurs.

      We need tier 1’s to stop the likes of google and Facebook from skewing the connectivity landscape. It’s tenuous as google already provide intercontinental links but they at least have the excuse of needing connectivity for their gcp products and yes Amazon and Microsoft do the same with their private intercontinental connections they also use for their customers cloud offerings.

      But Wikipedia is often stuffed with stale information, have you references that liberty charge some smaller providers for peering?

    8. Avatar photo CarlT says:

      Every day is a school day and my knowledge was clearly out of date and wrong. My thanks!

  2. Avatar photo Richard says:

    I’m not sure that this is just a TB BQM issue as I’ve been having issues with AWS and Amazon for a while too. M is right it’s pretty pointless for a mere customer to try to raise this issue because the people you talk to are just not set up to understand or recognise this as an issue.

    They were looking for a DevOps Engineer a while back and I was kind of interested in at least interviewing with them because as a long term discontented customer I could bring a that perspective to VM and ideas about what could be put in place to remedy some of the glaring CS and technical failings.

    1. Avatar photo Riza says:

      I used to work for Virgin granted it was in their centres and there were no cares for issues reported, last I was there we used Citrix and we had nothing but issues with 3 days in a row being off but reporting network issues was impossible for us work monkeys to report anything

    2. Avatar photo Chris says:

      I doubt a dev ops engineer would be let to influence how they do their CS or resolve their network technical failings.

      perhaps some kind of director of cs or something in that area of seniority.

      To be fair they have millions of customers telling them how much they suck and don’t listen to those that pay their revenue, why would they listen to a minion who they pay to dev ops stuff on how to do their cs?

  3. Avatar photo TrustMe says:

    Mark – Have you reported on the major Zen issue being discussed over on TBB? Zen core network engineers out at customers trying to work out what the problem is when they migrate FTTP circuits from BT Wholesale backhaul to Zen’s own network. 2 weeks and still not fixed!!!!

    1. Mark-Jackson Mark Jackson says:

      I’ve not had any such complaints sent to myself or our forum recently (at least not enough to peak my interest), so no. But Zen has had a few issues with FTTP provision and speeds over the past 18 months. In any case, this is not relevant to the topic above.

  4. Avatar photo MikeyMole says:

    OK this is interesting and somewhat comforting – I was monitoring my BQM graph (like all the other sadcases on here :P) and seen same spikes over last 2 days. Was worried it was a sudden local overutilisation issue – but if it’s a story on the mighty ISPReview.co.uk then maybe it’s a national thing and thus will surely go away soon.

    I hope it’s not a permanent thing, I’ve always had very decent latency from VM so far.

    P.S. obviously using in modem mode. Their CPE is awful.

  5. Avatar photo FibreAddict says:

    After 20yrs I left VM after Cityfibre came to my area.

    My last phone call to disconnect I thought would be the last as it was horrible. No they sent me a bill for £40 after they did not send return equipment bag. After another long wait it was eventually collected, I don’t expect this to go smooth either.

    If you can leave VM the customer service is poor at best

    1. Avatar photo John says:

      Similarly after 16 or so years I left when fttp was available in my area and been with andrew&arnold 160 and never had an issue, they monitor the network for me all the time, and I got a very powerful opnsense doing bandwidth sharing, latency is 0 unloaded and around 3ms loaded… the last I checked with virgin it was 10ms at best, and can go uk to 30ms during peak times.

      By the way I never used up all the 5tb, most was around 3tb, because I was getting worried I might exceed the limit.

      I can remain with these guys for ever, does everything I need. And there was no price hikes, I’m still in contract, but I’ve read even after the contract ends, it’s still sane price.

  6. Avatar photo Tazz says:

    @FibreAddict Same here best thing I ever did was moving to Giganet over Cityfibre

  7. Avatar photo Richard says:

    Looks like they have remedied the situation this evening according to my BQM graph. I’ll see tomorrow what the situation with AWS is for me

  8. Avatar photo BigJim says:

    My Virgin 220Mb fibre tracert tests were reporting 48ms to a gaming server… After the price increase I left to a temporary Three Network 4G and now getting 60ms…crazy how bad Virgin’s latency is on some parts of their mish mash network.

  9. Avatar photo Steven says:

    Looks to be fixed

  10. Avatar photo James says:

    Wonder if this is more of a widespread issue.

    KCOM user here and been having similar ‘weird’ experiences since Friday of last week!

    Packet loss During gaming, bizarre inconsistent speed tests, unusual pings.

  11. Avatar photo El Guapo says:

    I had noticed recently that I would type in a webpage, and it would just think about it for a moment or two. Whereas previously it would load UK pages in a second, it was taking 5 to 10 second to just start rendering pages.

    Of course VM told me that nothing was wrong and there were no known faults. I don’t know what I was thinking when I called them, I guess I thought they might be a real ISP and admit when there was a fault but it’s always a surprise to them when you call them and tell them it’s not working.

    1. Avatar photo NE555 says:

      > it would just think about it for a moment or two. Whereas previously it would load UK pages in a second, it was taking 5 to 10 second to just start rendering pages.

      Although there are lots of possible causes, this one sounds like it could be DNS.

    2. Avatar photo El Guapo says:

      Hi. Thanks for the reply. That was my initial though too, but a DNS query to cloudflare’s 1.1.1.1 is answered instantly like the results return as soon as I press enter. I should have tried to tcpdump it but the problem is gone now.

  12. Avatar photo Hurray says:

    Seems they solved it for a day so this article could be published 😉

    1. Avatar photo Jonathan says:

      I second that, we need this resolved it’s strange how it can stop over the weekend when weekend you would expect higher usage. They must know the issues, we just want it sorting

  13. Avatar photo Jonathan says:

    This is still an issue and like clockwork 8pm it starts till about 10pm then seems to have the weekend off and show its ugly head, all vmo2 say is your connection is in spec and not issues, however evidence is mounting up on their community forums and not one word from any mods.

  14. Avatar photo justapasserby says:

    Confirmed the issue persists. It doesn’t manifest itself really to users unless they’re actually doing something in real time where the fluctuations would be noticed. What gets me is they’re allowed to advertise this as a gaming service, when in reality during peak times its awful, if the ping remained consistant it wouldn’t be so bad, its the swings from 18ms – 60ms constantly on an EU gaming server, you’re moving forward but the only way I can explain what its like is that you take 2 steps then its like you’re falling over but you then teleport a few steps forward….its extremely jarring, and thankfully CityFibre have indicated they’re going to run fibre down the hill that I live, the moment that happens I am off of VM, speed is great, but anything else is pot luck with VM. That this website or many other major broadband news websites aren’t destroying VM over this is kind of sad really….chance to make a proper name for themselves and collect a whole bunch of readers and advocates. This website and tbb may be ‘big’ broadband sites but they’re a long long way from being a ‘household’ name….this could go a decent way toward that. The out of date late 90’s aesthetic to the website could do with an update too 😉

  15. Avatar photo Richard says:

    This has returned for me too. Started showing up again on the 2nd of May for my BQM graphs.

  16. Avatar photo Jonathan says:

    This has returned again a few days told it was fixed now back and issues all throughout the day, also they have closed the main thread that had over 55 pages today

Comments are closed

Cheap BIG ISPs for 100Mbps+
Community Fibre UK ISP Logo
150Mbps
Gift: None
Virgin Media UK ISP Logo
Virgin Media £26.00
132Mbps
Gift: None
Shell Energy UK ISP Logo
Shell Energy £26.99
109Mbps
Gift: None
Sky Broadband UK ISP Logo
100Mbps
Gift: None
Plusnet UK ISP Logo
Plusnet £27.99
145Mbps
Gift: None
Large Availability | View All
Cheapest ISPs for 100Mbps+
Gigaclear UK ISP Logo
Gigaclear £17.00
200Mbps
Gift: None
YouFibre UK ISP Logo
YouFibre £19.99
150Mbps
Gift: None
Community Fibre UK ISP Logo
150Mbps
Gift: None
BeFibre UK ISP Logo
BeFibre £21.00
150Mbps
Gift: £25 Love2Shop Card
Hey! Broadband UK ISP Logo
150Mbps
Gift: None
Large Availability | View All
The Top 15 Category Tags
  1. FTTP (5534)
  2. BT (3518)
  3. Politics (2542)
  4. Openreach (2299)
  5. Business (2267)
  6. Building Digital UK (2247)
  7. FTTC (2045)
  8. Mobile Broadband (1978)
  9. Statistics (1790)
  10. 4G (1669)
  11. Virgin Media (1624)
  12. Ofcom Regulation (1466)
  13. Fibre Optic (1396)
  14. Wireless Internet (1392)
  15. FTTH (1382)

Helpful ISP Guides and Tips

Promotion
Sponsored

Copyright © 1999 to Present - ISPreview.co.uk - All Rights Reserved - Terms , Privacy and Cookie Policy , Links , Website Rules , Contact
Mastodon