7-11 Slow Download Speeds

[rextilleon]

Very frustrating.  Every night between 7-11 my download speeds on my 75/75 plan slow to a crawl.  After several calls to Verizon tech, still the same problem.  We have determined  that there are no issues on my end, so I can only assume its on Verizons.  The interesting thing is that my upload speed remain perfect during the evening hours. 

[#2]

wow @headhunter, that's pathetic. file a complaint with the FCC. this is a known issue and it's verizon's infastructure. they're providing bad peering/congested routes and keep signing customers onto services that they can't even support for them because of the upgrades that are needed to support the mass amount of people already using the service. it's the same thing that happened with netflix last year but on a higher more nation wide scale that's affecting all services.

[tns2]

---

@Anonymous wrote:

@wow @Headhunter, that's pathetic. file a complaint with the FCC. this is a known issue and it's verizon's infastructure. they're providing bad peering/congested routes and keep signing customers onto services that they can't even support for them because of the upgrades that are needed to support the mass amount of people already using the service. it's the same thing that happened with netflix last year but on a higher more nation wide scale that's affecting all services.

---

The posted is probably not related to any peering congested problem.  It instead looks like a problem that Verizon definitely will have to look at.  Perhaps his ont (several have had that problem) or the equipment at the local verizon location.

[#4]

there's over 25,000 reports of this nation wide already and that's only the people that have noticed and taken initiative in running tracert's/speedtests/linequalitytests, who knows how many more average internet consumers that have no idea something is wrong let alone run those tests at peak times. verzion responded to the FCC and is taking blame for this issue but they're doing so in the dark because they don't want the bad press after what happened with netflix last year.

[Headhunter06]

A little test to prove that the issue is not in my house:  Here is my speed at approx 6:45 PM on 4 FEB (below). I'll repost with what my speed is a little later on during prime time hours. If the pattern holds (as it has since late 2013), we'll be down to somewhere between 1 and 5 Mbps at around 9PM.

[meisterMD]

You guys are having problems getting credit? I haven't had any issues now that I've had validation from multiple techs that it is a city-wide problem in my area (Inland Empire, CA), which is probably the least they can do.

I go from 50/50 from midnight to around 630pm where it gradually descends and tanks at around 730pm to 5 Mbps download max. Upload is almost always preserved. Very frustrating. I've spent 8+ hours on the phone with tech support and have had four techs out.

[Headhunter06]

So we go from way more than 75 just prior to prime time to this (below) at 8:45PM. (Note that this is double the speed I was getting at this time last night. Wheeee!!!

Network Diagnostic Test

The Network Diagnostic Tool results below will be used by our Verizon Technical Support Team to review your computer settings.


Analysis information:

WEB100 Enabled Statistics:
Checking for Middleboxes . . . . . . . . . . . . . . . . . .  Done
checking for firewalls . . . . . . . . . . . . . . . . . . .  Done
running 10s outbound test (client-to-server [C2S]) . . . . . 35.65Mb/s
running 10s inbound test (server-to-client [S2C]) . . . . . . 4.22Mb/s

	------  Client System Details  ------
OS data: Name = Windows 7, Architecture = x86, Version = 6.1
Java data: Vendor = Oracle Corporation, Version = 1.8.0_31

	------  Web100 Detailed Analysis  ------
100 Mbps FastEthernet link found.
Link set to Full Duplex mode
Information: throughput is limited by other network traffic.
Good network cable(s) found
Normal duplex operation found.

Web100 reports the Round trip time = 24.85 msec; the Packet size = 1452 Bytes; and 
There were 101 packets retransmitted, 801 duplicate acks received, and 865 SACK blocks received
The connection was idle 0 seconds (0%) of the time
S2C throughput test: Packet queuing detected: 81.69%
This connection is sender limited 25.75% of the time.
This connection is network limited 72.81% of the time.

[CMDuser]

new user here.

Started seeing this issue in Manhattan, NY as of 1/26, exactly a week ago.

My service , for a bit last week, was interrupted daylong. I attributed it to the NE blizzard storm.  It gradually got better, but never really recovered. What made me think it was more than broken fiber optic cables is that \service sucks after 4pm until midnight BUT only for connecting with certain servers , for example APP.BILL.COM

(I run a business , and we make most payments with Bill.com, a wonderful SaaS plaftorm for making e-checks)

In other words, most internet services work fine (yes, including NETFLIX and YOUTUBE)  but they ones that are not like bill.com ..... are literally broken. I am not sure if Verizon blacklisted certain servers / connections, but here is what I get when I try to ping the Bill.com server in CA anytime between 4 and 1am EST (See below image)

I am one of a handful of customers reporting this type of issue to bill.com out of NYC.

I have also run speedtests and every time I set my location to a SF bay area server, I get download speeds in the low 4's for downloads, and 20+ for uploads, even though i am on the 50/50 plan.

THIS IS BEYOND THROTTLING, THIS IS OUTRIGHT SABOTAGE!

[CMDuser]

point to be made:  The likely reason youtube works as well as it does for me (@ NYC) it because GOOGLE has a gigantic fiber optic cable on the corner of its offices at 14st and 8ave, where youtube is also located.  One of my friends is an engineer at GOOG and he said 33% of city internet traffic goes thru the fiber optic cable at the old port authority building (now owned by google).

This story seems backed up by the route of traffic per trace below. Note how verizon server times out, but signal is "rescued" so to speak by google servers.

[#10]

just because it's timing out it doesn't mean something is wrong, some nodes automatically decline ping requests by default, as long as you're getting to your destination without bad latency and or packet loss you're fine.

p.s that alter.net hop is beyond troublesome and is a nationwide issue for the people routing through it..

[CMDuser]

I don't necessarily disagree, although routing thru most of that network is giving out pings in the high 100s and over 200 in some cases.

[bensel3]

@Anonymous wrote:

just because it's timing out it doesn't mean something is wrong, some nodes automatically decline ping requests by default, as long as you're getting to your destination without bad latency and or packet loss you're fine.

p.s that alter.net hop is beyond troublesome and is a nationwide issue for the people routing through it..

---

p.p.s. ALTER.NET is still Verizon. They bought MCI (which owned/merged with Worldcom, that bought MFS, that bought UUNet which built the ALTER.NET backbone). It's just not out on the edge, but a backbone link.

[#13]

stuck on the same boat guys they won't let me void my contract even though im not even recieving 1% of what im paying for after 7pm this is beyond fraustrating.

[Headhunter06]

Hmmm.... maybe Verizon has fixed things?  Have had several tests on my iPad (while propped up watching TV) and got 75+ for download speeds. This on a Friday night, which is often one of the worst internet speed nights. Fingers crossed that this sudden improvement in speed is true and not a mirage.

Speeds below were done with two kids also watching Netflix on an Apple TV, so 35-40 for this Network Test seem fairly reasonable. I sure hope this is a good sign.

WEB100 Enabled Statistics:
Checking for Middleboxes . . . . . . . . . . . . . . . . . .  Done
checking for firewalls . . . . . . . . . . . . . . . . . . .  Done
running 10s outbound test (client-to-server [C2S]) . . . . . 38.48Mb/s
running 10s inbound test (server-to-client [S2C]) . . . . . . 39.31Mb/s

	------  Client System Details  ------
OS data: Name = Windows 7, Architecture = x86, Version = 6.1
Java data: Vendor = Oracle Corporation, Version = 1.8.0_31

	------  Web100 Detailed Analysis  ------
100 Mbps FastEthernet link found.
Link set to Full Duplex mode
No network congestion discovered.
Good network cable(s) found
Normal duplex operation found.

Web100 reports the Round trip time = 49.68 msec; the Packet size = 1452 Bytes; and 
No packet loss - but packets arrived out-of-order 2.74% of the time
S2C throughput test: Packet queuing detected: 30.59%
This connection is receiver limited 12.91% of the time.
  Increasing the the client's receive buffer (255.0 KB) will improve performance
This connection is sender limited 86.15% of the time.
This connection is network limited 0.94% of the time.

[Headhunter06]

Well, we now have two nights in a row with good (indeed, great) speed here in Northern Virginia. The below is even in the midst of a mighty wind storm, over the house Wifi connection, AND with the kids using Netflix on their iPods.! This is hoping that my issue is now resolved. I'll keep checking over the coming days.

[Headhunter06]

Still doing great here in Northern Virginia. Continuosly good speeds at all hours. Issue resolved. Thanks to all in the community who have been trying to make Verizon aware of the issue. Thanks to Verizon for fixing it. Much appreciated. It took a while to resolve, but it got done. And that is what matters!

[Headhunter06]

So far so good on Verizon fixingh things, eh?  Friday night at 10:35PM on a 75/75 that I am paying for:

Analysis information:

WEB100 Enabled Statistics:
Checking for Middleboxes . . . . . . . . . . . . . . . . . .  Done
checking for firewalls . . . . . . . . . . . . . . . . . . .  Done
running 10s outbound test (client-to-server [C2S]) . . . . . 38.56Mb/s
running 10s inbound test (server-to-client [S2C]) . . . . . . 1.59Mb/s

	------  Client System Details  ------
OS data: Name = Windows 7, Architecture = x86, Version = 6.1
Java data: Vendor = Oracle Corporation, Version = 1.8.0_31

	------  Web100 Detailed Analysis  ------
100 Mbps FastEthernet link found.
Link set to Full Duplex mode
No network congestion discovered.
Good network cable(s) found
Normal duplex operation found.

Web100 reports the Round trip time = 28.46 msec; the Packet size = 1452 Bytes; and 
There were 67 packets retransmitted, 267 duplicate acks received, and 307 SACK blocks received
The connection was idle 0 seconds (0%) of the time
S2C throughput test: Packet queuing detected: 93.08%
This connection is receiver limited 2.90% of the time.
  Increasing the the client's receive buffer (17.0 KB) will improve performance
This connection is sender limited 11.34% of the time.
This connection is network limited 85.77% of the time.
Excessive packet loss is impacting your performance, check the auto-negotiate function on your local PC and network switch

[Cfitzy84]

So I have had this problem starting on 01/01/15. I have had a few techs come out to my home finally around mid January a tech finally called me and told me that he would not even come out to my house (I live in Northern VA) he knows the problem is with Verizon itself but they will NOT fix it because the problem costs to much money and his advise to me was to make as much complaints as I can until they fix it. It has now been over a month and still to no avail. So hopefully someone on corportate actually reads this and knows how much their business practices sucks. I wish I could cancel but I'm under contract 😞

[#19]

file complaints with the FCC, that seems to be the only thing getting their attention.

look at that speed matching from last night 

WEB100 Enabled Statistics:
Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done
checking for firewalls . . . . . . . . . . . . . . . . . . . Done
running 10s outbound test (client-to-server [C2S]) . . . . . 94.56Mb/s
running 10s inbound test (server-to-client [S2C]) . . . . . . 1.59Mb/s

[Cfitzy84]

I just finished filing my FCC Complaint and all the screenshots of speedtests, traceroutes, etc. Hopefully this will help them get this problem fixed... I just want working internet when I come home from work 😞

[Headhunter06]

Again to prove my point that this is a Verizon network issue and NOT an issue here in my house, see below. Its 1:40AM on Sunday morning (yeah, up a bit late). Note that download spped is just fine. But look a couple of comments above to my last post and you'll see that during prime time we are down to less than 5mb/s most of the time and often less than 2 mb/s. I just don't get that Verizon cannot seem to understand there is a problem that needs to be fixed ASAP. Its just mind boggeling.

WEB100 Enabled Statistics:
Checking for Middleboxes . . . . . . . . . . . . . . . . . .  Done
checking for firewalls . . . . . . . . . . . . . . . . . . .  Done
running 10s outbound test (client-to-server [C2S]) . . . . . 51.61Mb/s
running 10s inbound test (server-to-client [S2C]) . . . . . . 75.97Mb/s

	------  Client System Details  ------
OS data: Name = Windows 8, Architecture = x86, Version = 6.2
Java data: Vendor = Oracle Corporation, Version = 1.7.0_21

	------  Web100 Detailed Analysis  ------
100 Mbps FastEthernet link found.
Link set to Full Duplex mode
No network congestion discovered.
Good network cable(s) found
Normal duplex operation found.

Web100 reports the Round trip time = 28.21 msec; the Packet size = 1460 Bytes; and 
There were 1 packets retransmitted, 203 duplicate acks received, and 201 SACK blocks received
The connection was idle 0 seconds (0%) of the time
S2C throughput test: Packet queuing detected: 17.61%
This connection is receiver limited 15.98% of the time.
This connection is sender limited 82.74% of the time.
This connection is network limited 1.26% of the time.

[ike71]

I am getting pretty good download speeds in Union, NJ.......right now.