Question Slow download speeds across multiple devices on the network, wired and wireless. No difference after a new gateway.

[korvisus]

Had great download speeds, great speed tests from google, ookla, fast, etc... then download speeds on all networked devices began to drop significantly. multiple wired computers, phones, laptops have been tried. noticed this, did speedtests, now all testing sites are wildly different. some say 10, some say 50, some say 150, and xfinity's says 200+ like nothing is different. they sent out a new gateway because we were due for an upgrade. got it set up, and now its the same. slow DL speeds, all test sites saying the same speeds as before, consistently.

the way i see it, its either a comcast problem, or a wiring problem. they want to charge me for a tech to come out, but i told them i shouldnt have to pay for it if it isnt a problem i could have prevented or resolved myself. i pay wildly inflated costs for internet alone.

i just wanna make sure im not missing anything. the symptoms seem pretty specific, and it would be nice to be able to tell them exactly what it is, because they do not diagnose at all. they look at their workflow and charge you to send out a tech if all else fails.

 

[Ralston18]

Make and model modem and router? Or combo device if combined?

Connectivity being something like the following line diagram:

Comcast ---coax--> Modem ---Ethernet cable --->[WAN]Router[LANs] ---Ethernet cable--> Wired devices and ~~wireless ~> Wireless devices.

Edit and correct as necessary.

On one of your wired computers, run and post the results of "ipconfig /all" (without quotes) via the command prompt.

Also: are you familiar with "ping", "pathping", and "tracert".

 

[SamirD]

I've had to deal with problems like this. Sounds like either there is a routing issue once your packets get into their network or an issue with packets between your modem and their equipment.

So here's how I go about getting them to do their job.

First, call support and ask them to run an internal ping to your equipment and see if there's any packet loss. If there is, that's the culprit, they will send a tech at no charge as it is a line issue.

If there is no issue, thank them and hang up. Now you will need to determine at what hop these issues start. You will need to run several trace routes to different sites multiple times a day to figure this out. If the hop is within their network, this is when it gets hard because you'll need to somehow get this to their net ops and the front line support agents won't let you do this. Generally, I've found that this type of problem resolves itself in a week or so, and I'll call in for credits and generally get them.

If you find that the hop is beyond your isp, then this is where it gets really hard as you basically are needing to tell an isp that they are having an issue with their provider. This can take weeks to years to resolve. At that point you either live with it or need to find another form of internet access that doesn't suffer the same issue. Oh, but I forgot about one way around such an issue--vpn. VPN services many times will take a completely different route and bypass a bad hop so that can be a way around the issue. And vpngate.net is a great free way to test this or even run it permanently if need be.

Good luck and keep us posted.

 

[korvisus]

tg02dcw3482p2ct (model number of the modem/router combo)

Spoiler: cmd ipconfig /all

Connection-specific DNS Suffix . : hsd1.sc.comcast.net
Description . . . . . . . . . . . : Realtek PCIe GbE Family Controller
Physical Address. . . . . . . . . : 70-85-C2-FF-73-71
DHCP Enabled. . . . . . . . . . . : Yes
Autoconfiguration Enabled . . . . : Yes
IPv6 Address. . . . . . . . . . . : 2601:743:280:5eb0::aabd(Preferred)
Lease Obtained. . . . . . . . . . : Sunday, May 17, 2020 3:31:51 PM
Lease Expires . . . . . . . . . . : Saturday, May 23, 2020 2:05:33 PM
IPv6 Address. . . . . . . . . . . : 2601:743:280:5eb0:18d5:a4f2:b209:1ce9(Preferred)
Temporary IPv6 Address. . . . . . : 2601:743:280:5eb0:9187:622d:5dd7:695e(Preferred)
Link-local IPv6 Address . . . . . : fe80::18d5:a4f2:b209:1ce9%6(Preferred)
IPv4 Address. . . . . . . . . . . : 10.0.0.46(Preferred)
Subnet Mask . . . . . . . . . . . : 255.255.255.0
Lease Obtained. . . . . . . . . . : Sunday, May 17, 2020 3:31:50 PM
Lease Expires . . . . . . . . . . : Sunday, May 24, 2020 4:03:14 PM
Default Gateway . . . . . . . . . : fe80::ca63:fcff:fe0e:ca79%6
10.0.0.1
DHCP Server . . . . . . . . . . . : 10.0.0.1
DHCPv6 IAID . . . . . . . . . . . : 108037570
DHCPv6 Client DUID. . . . . . . . : 00-01-00-01-26-36-74-A3-70-85-C2-FF-73-71
DNS Servers . . . . . . . . . . . : 2001:558:feed::1
2001:558:feed::2
8.8.8.8
8.8.4.4
2001:558:feed::1
2001:558:feed::2
NetBIOS over Tcpip. . . . . . . . : Enabled

Spoiler: cmd ping (several targets)

Pinging www.google.com [2607:f8b0:4008:811::2004] with 32 bytes of data:
Reply from 2607:f8b0:4008:811::2004: time=25ms
Reply from 2607:f8b0:4008:811::2004: time=21ms
Reply from 2607:f8b0:4008:811::2004: time=22ms
Reply from 2607:f8b0:4008:811::2004: time=21ms

Ping statistics for 2607:f8b0:4008:811::2004:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 21ms, Maximum = 25ms, Average = 22ms

C:\Windows\system32>ping www.yahoo.com

Pinging new-fp-shed.wg1.b.yahoo.com [2001:4998:44:41d::4] with 32 bytes of data:
Reply from 2001:4998:44:41d::4: time=71ms
Reply from 2001:4998:44:41d::4: time=67ms
Reply from 2001:4998:44:41d::4: time=66ms
Reply from 2001:4998:44:41d::4: time=67ms

Ping statistics for 2001:4998:44:41d::4:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 66ms, Maximum = 71ms, Average = 67ms

C:\Windows\system32>ping www.speedtest.net

Pinging zd.map.fastly.net [151.101.206.219] with 32 bytes of data:
Reply from 151.101.206.219: bytes=32 time=41ms TTL=56
Reply from 151.101.206.219: bytes=32 time=38ms TTL=56
Reply from 151.101.206.219: bytes=32 time=37ms TTL=56
Reply from 151.101.206.219: bytes=32 time=37ms TTL=56

Ping statistics for 151.101.206.219:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 37ms, Maximum = 41ms, Average = 38ms

Spoiler: cmd tracert (several targets)

Tracing route to www.google.com [2607:f8b0:4008:811::2004]
over a maximum of 30 hops:

1 1 ms <1 ms 1 ms 2601:743:280:5eb0:ca63:fcff:fe0e:ca79
2 11 ms 11 ms 7 ms 2001:558:4041:7::1
3 8 ms 8 ms 8 ms 2001:558:152:4b::1
4 16 ms 9 ms 11 ms 2001:558:150:55::1
5 18 ms 16 ms 15 ms 2001:558:190:244::1
6 25 ms * * be-33489-cr02.miami.fl.ibone.comcast.net [2001:558:0:f7cb::1]
7 * * * Request timed out.
8 25 ms 26 ms 24 ms 2001:559::7ba
9 23 ms 23 ms 22 ms 2607:f8b0:80f3::1
10 * * 26 ms 2001:4860:0:1161::1
11 23 ms 23 ms 23 ms 2001:4860:0:1161::2
12 23 ms 25 ms 23 ms 2001:4860::c:4000:fd8c
13 23 ms 32 ms 23 ms 2001:4860::1:4000:fd89
14 22 ms 22 ms 24 ms 2001:4860:0:12e2::1
15 24 ms 24 ms 24 ms 2001:4860:0:1::295b
16 24 ms 23 ms 21 ms mia09s18-in-x04.1e100.net [2607:f8b0:4008:811::2004]

Trace complete.

C:\Windows\system32>tracert www.yahoo.com

Tracing route to new-fp-shed.wg1.b.yahoo.com [2001:4998:44:41d::4]
over a maximum of 30 hops:

1 1 ms 1 ms 1 ms 2601:743:280:5eb0:ca63:fcff:fe0e:ca79
2 10 ms 12 ms 8 ms 2001:558:4041:7::1
3 8 ms 9 ms 9 ms 2001:558:152:4b::1
4 11 ms 9 ms 10 ms 2001:558:150:55::1
5 15 ms 15 ms 15 ms 2001:558:190:244::1
6 23 ms * * be-33489-cr02.miami.fl.ibone.comcast.net [2001:558:0:f7cb::1]
7 * * * Request timed out.
8 22 ms 21 ms 22 ms 10gigabitethernet9-1.core1.mia1.he.net [2001:470:0:32d::1]
9 30 ms 30 ms 28 ms 2001:478:124::115
10 40 ms 41 ms 42 ms 2001:4998:f02d:206::
11 59 ms 60 ms 59 ms ae-13.pat1.che.yahoo.com [2001:4998:f00d🅰:]
12 69 ms 70 ms 69 ms ae-5.pat1.nez.yahoo.com [2001:4998:f000:3::]
13 72 ms 72 ms 71 ms 2001:4998:f000:9::1
14 74 ms 73 ms 72 ms 2001:4998:44:fe1b::1
15 69 ms 68 ms 68 ms 2001:4998:44:fa12::1
16 70 ms 68 ms 67 ms media-router-fp2.prod1.media.vip.ne1.yahoo.com [2001:4998:44:41d::4]

Trace complete.

C:\Windows\system32>tracert www.speedtest.net

Tracing route to zd.map.fastly.net [151.101.206.219]
over a maximum of 30 hops:

1 <1 ms 1 ms 1 ms 10.0.0.1
2 10 ms 11 ms 8 ms 96.120.3.13
3 8 ms 8 ms 8 ms 68.86.144.141
4 9 ms 9 ms 8 ms 68.86.130.205
5 15 ms 15 ms 17 ms 68.87.164.125
6 24 ms 24 ms 25 ms be-33489-cr02.miami.fl.ibone.comcast.net [68.86.95.45]
7 40 ms 40 ms 41 ms be-302-cr12.doraville.ga.ibone.comcast.net [96.110.38.61]
8 40 ms 41 ms 47 ms be-1312-cs03.doraville.ga.ibone.comcast.net [96.110.34.185]
9 39 ms 40 ms 40 ms be-1311-cr11.doraville.ga.ibone.comcast.net [96.110.34.170]
10 40 ms 39 ms 40 ms be-302-cr13.56marietta.ga.ibone.comcast.net [96.110.39.49]
11 41 ms 39 ms 41 ms be-1113-cs01.56marietta.ga.ibone.comcast.net [96.110.34.225]
12 39 ms 41 ms 40 ms be-2101-pe01.56marietta.ga.ibone.comcast.net [96.110.32.86]
13 41 ms 39 ms 40 ms 66.208.228.110
14 38 ms 40 ms 39 ms 151.101.206.219

I am not familiar with the pingpath.

 

[SamirD]

Thank you for the traceroute. You can clearly see in the first two traces that there is a problem at hop 7, and with the results of the 3rd trace that resolves inside the comcast network, that hop 7 is only having problems when hitting servers outside the comcast network.

If someone higher up at comcast sees these, they will investigate. You will need to call in and let them know what you have found (what I wrote above) and see if they will transfer you to someone who will look at your traces to start fixing the problem. If not, you can call back and try again. Sometimes the social media route or dslreports or other places are a better way to reach someone, but this is either going to fix itself in a week or a month or you'll have to live with it.

 

[korvisus]

Thanks so much! I appreciate your help. Ive been around the block with comcast many times before. i hate the company but have to admit that their tier 1 customer service has improved alot over the last few years. we will see @.@

 

[bill001g]

Try to disable ipv6 support in the router to test. IPv6 even though they have been claiming it is the future for 20yrs still is not fully supported. It many times follows a different path and it does not seem to be as stable.

Even if the problem is not IPv6 it makes the troubleshooting easier because there is one less variable to content with.

From what you have posted you likely have some intermittent issue which is going to make it harder to find. The traces you have could be a issues or is could also be that the intermediate routers were busy doing more important things like passing user data instead of responding to ping/trace.

None of the ping or trace you have actually show any issues of high latency or loss to the end node. This means you have no problems

You are going to have to do more data collection until some form of pattern exists to show where the problem really is. Your best case is going to be if you see issues in hop 1 an hop 2 since those the level 1 tech know how to resolve. Hops farther up the trace the level 1 techs tend to not have the skills or the authorizations to do anything about and it is hard to get them to escalate to a higher level person unless you have very compelling data to back it up.

 

[SamirD]

Thanks so much! I appreciate your help. Ive been around the block with comcast many times before. i hate the company but have to admit that their tier 1 customer service has improved alot over the last few years. we will see @.@

You're welcome! Bill001g also has some great information as well especially about ipv6.

 

[korvisus]

So, it boils down to, are there problems with the hops in the trace or not? if so, that is something i can focus on. if not, then im back at square one.

 

[bill001g]

This is the problem with issues that come and go. It takes a lot of testing to get correct data on the issue.

What I would do first is run tracert to some IP address like 8.8.8.8. Next leave constant ping to hop 1 (your router) and hop 2 (isp first router) run in the background. What you are hoping to see is packet loss or large delays in these.

You can do the same with hops farther in to the trace but what can you possible do if the problem is in say google or level3 or any other ISP. Even you ISP can not fix things that are messed up in other ISP.

Still I would first disable the ipv6 support. This is a extremely common cause of different performance to different servers. Some use ipv4 and some use ipv6. IPv6 many times does not have as good a path for some unknown reason in many cases.

 

[korvisus]

I did disable ipv6, in windows, as unfortunately comcast believes that their customers should not have control over their router software with one of their main gateway models. It did not make a difference in several speed tests.

in regard to the constant ping:

Spoiler: IPV6 Enabled

Spoiler: First Hop

Ping statistics for 10.0.0.1:
Packets: Sent = 194, Received = 194, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 0ms, Maximum = 2ms, Average = 0ms

Spoiler: Second Hop

Ping statistics for 96.120.3.13:
Packets: Sent = 189, Received = 189, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 6ms, Maximum = 23ms, Average = 6ms

Spoiler: IPV6 Disabled

Spoiler: First Hop

Ping statistics for 10.0.0.1:
Packets: Sent = 327, Received = 327, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 0ms, Maximum = 2ms, Average = 0ms

Spoiler: Second Hop

Ping statistics for 96.120.3.13:
Packets: Sent = 349, Received = 349, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 6ms, Maximum = 27ms, Average = 6ms

These tests were taken at 3:00pm

I have not noticed this issue being intermittent. It seems to have changed at some point and this is the new normal. It has been like this for about 2 weeks.


Ill reserve the following space for another series of tests taken at 7:00pm.

Spoiler: IPV6 Enabled

Spoiler: First Hop

Ping statistics for 10.0.0.1:
Packets: Sent = 122, Received = 122, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 0ms, Maximum = 2ms, Average = 0ms

Spoiler: Second Hop

Ping statistics for 96.120.3.13:
Packets: Sent = 117, Received = 117, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 6ms, Maximum = 21ms, Average = 7ms

Spoiler: IPV6 Disabled

Spoiler: First Hop

Ping statistics for 10.0.0.1:
Packets: Sent = 93, Received = 93, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 0ms, Maximum = 2ms, Average = 0ms

Spoiler: Second Hop

Ping statistics for 96.120.3.13:
Packets: Sent = 93, Received = 93, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 6ms, Maximum = 15ms, Average = 7ms

 

[SamirD]

Some of the tools I've used to provide data for packet loss related issues are the freeola line test (not speedtest) and the thinkbroadband monitor. There's also a new one that I recently discovered called packetlosstest. These 3 help provide a lot of data that you can share with the isp (especially the thinkbroadband one--I actually keep this running on all the accounts I manage).

 

[korvisus]

Hey, found the ultimate solution to this issue, thought somebody might want or need it.
CMD (elevated)
netsh int tcp set global autotuninglevel=normal

went from 10mbps to almost 400 instantly.

windows update is continually disabling this feature, which is why reinstalling windows resolved it, and then it disabled it and the problem was present again. happened multiple times on both mine and my roommate's computers. the differences between devices of course vary due to hardware.. but for the computers, this is what resolved it.

 

[SamirD]

Glad you found it and thank you for posting the solution!