What's new

Netgear router 50ms latency more than direct to modem.

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

jasoneo

New Around Here
Recently I noticed an increase of about 50ms in ping to WAN addresses. Connected direct to a cable modem ping rates are normal for my geographic area. Going through the router, basic configuration, no Qos, etc., the latency is 50ms more. This just started happening. I have changed routers, and changed firmware, but the 2 routers I tested both remain consistently about 50ms over a direct connection to the modem. I thought maybe there was a problem between the router and the modem, however, I can ping the modem local address from the router utility and there is no delay.

Below are the MTR results to a public address (google I think). The first trace is from the router. The second trace is directly connect to the modem. NO WIRELESS. This is all Cat cable. Is the ISP routing this based on the source? I noticed that on the router trace, it is similar until it hit "66-76-31-129.suddenlink.net" (6th down from top). This address is NOT in the trace of the direct modem trace. That is the turning point for times. From there it jumps up 50ms.

Any ideas are appreciated.

|------------------------------------------------------------------------------------------|
| WinMTR statistics | (router)
| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
| www.routerlogin.com - 0 | 42 | 42 | 0 | 0 | 0 | 0 |
| No response from host - 100 | 8 | 0 | 0 | 0 | 0 | 0 |
| 173-219-232-22.suddenlink.net - 0 | 42 | 42 | 9 | 11 | 18 | 10 |
|173-219-218-106.nlsncmtc01.suddenlink.net - 0 | 42 | 42 | 10 | 12 | 26 | 11 |
| 173-219-218-202.suddenlink.net - 0 | 42 | 42 | 16 | 19 | 26 | 18 |
| 66-76-31-129.suddenlink.net - 0 | 42 | 42 | 16 | 19 | 33 | 17 |
| 66-76-229-162.suddenlink.net - 0 | 42 | 42 | 83 | 88 | 97 | 87 |
| 108.170.240.98 - 0 | 42 | 42 | 82 | 84 | 93 | 83 |
| 142.251.49.194 - 0 | 42 | 42 | 83 | 87 | 94 | 84 |
| 142.251.49.207 - 0 | 42 | 42 | 76 | 80 | 112 | 100 |
| 72.14.233.11 - 0 | 42 | 42 | 77 | 81 | 94 | 78 |
| 108.170.243.193 - 0 | 42 | 42 | 77 | 79 | 86 | 78 |
| 209.85.255.173 - 0 | 42 | 42 | 76 | 79 | 92 | 76 |
| ord38s19-in-f4.1e100.net - 0 | 42 | 42 | 75 | 78 | 84 | 77 |
|________________________________________________|______|______|______|______|______|______|
WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

|------------------------------------------------------------------------------------------|
| WinMTR statistics | (modem)
| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
| No response from host - 100 | 7 | 0 | 0 | 0 | 0 | 0 |
| 173-219-232-22.suddenlink.net - 0 | 34 | 34 | 8 | 9 | 16 | 9 |
|173-219-218-104.nlsncmtc01.suddenlink.net - 0 | 34 | 34 | 9 | 10 | 14 | 10 |
| 173-219-218-200.suddenlink.net - 0 | 34 | 34 | 16 | 18 | 24 | 17 |
| 66-76-229-162.suddenlink.net - 0 | 34 | 34 | 51 | 59 | 175 | 54 |
| 108.170.246.34 - 0 | 34 | 34 | 28 | 30 | 36 | 29 |
| 108.170.232.199 - 0 | 34 | 34 | 27 | 30 | 37 | 31 |
| 209.85.252.46 - 0 | 34 | 34 | 27 | 29 | 37 | 29 |
| 209.85.249.137 - 0 | 34 | 34 | 27 | 28 | 36 | 28 |
| 108.170.243.174 - 0 | 34 | 34 | 28 | 30 | 37 | 28 |
| 209.85.255.145 - 0 | 34 | 34 | 28 | 29 | 37 | 29 |
| lga15s49-in-f4.1e100.net - 0 | 34 | 34 | 27 | 27 | 34 | 27 |
|________________________________________________|______|______|______|______|______|______|
WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider
 
The tracerts show two very different end points. What IP address are you pinging for your test?
 
Google. Both pointed at 172.217.5.4.

To be honest, I haven't duplicated this test to see the routes multiple times connected to modem and connected to router.

ISPGoogle LLC
Usage TypeData Center/Web Hosting/Transit
Hostname(s)lga15s49-in-f4.1e100.net
lga15s49-in-f4.1e100.net
ord38s19-in-f4.1e100.net
Domain Namegoogle.com
Country
blank.gif
United States of America
CityMountain View, California
 
You really should ping an IP address if you want consistent results,not a domain name.
Sites with huge traffic route traffic to any number of servers.

Your experiments have already shown this is a network, not a router issue.
 
You really should ping an IP address if you want consistent results,not a domain name.
Sites with huge traffic route traffic to any number of servers.

Your experiments have already shown this is a network, not a router issue.

I have tried ping ip addresses but haven't documented the results.

I tried a Tplink router with the same "bad" results. Getting 50ms over the direct modem connection.

It is frustrating that I'm not sure why this is happening but I agree, I don't believe the issue is in my router.
 
I've tracert one of google's IP's (not the domain name). It appears my ISP has implemented a routing policy that is treating my consumer routers differently than when I have a PC directly connected. This is most frustrating. There is an extra suddenlink hop on the PC direct compared to the router connected and that is when the route gets handed off to Google and the times get consistently higher. But I don't know much about the details of how modern ISP's handle data.

I noticed that the IP's being assigned DHCP from the modem (ISP) to the direct connected PC's are very different than to the WAN port on my router. I'm assuming must be based on the MAC of the physical port? I could set up a PC based router and see if that works?

|------------------------------------------------------------------------------------------|
| WinMTR statistics (router connected) |
| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
| ARCHER_C50 - 0 | 48 | 48 | 0 | 0 | 1 | 1 |
| No response from host - 100 | 9 | 0 | 0 | 0 | 0 | 0 |
| 173-219-232-20.suddenlink.net - 3 | 45 | 44 | 0 | 10 | 18 | 10 |
|173-219-218-104.nlsncmtc01.suddenlink.net - 0 | 48 | 48 | 10 | 11 | 28 | 11 |
| 173-219-218-200.suddenlink.net - 3 | 45 | 44 | 0 | 18 | 27 | 17 |
| 66-76-229-162.suddenlink.net - 3 | 45 | 44 | 0 | 85 | 91 | 85 |
| 108.170.246.34 - 3 | 45 | 44 | 0 | 85 | 92 | 85 |
| 142.251.49.77 - 3 | 45 | 44 | 0 | 84 | 91 | 85 |
| 142.251.49.91 - 3 | 45 | 44 | 0 | 80 | 90 | 78 |
| 72.14.232.108 - 3 | 45 | 44 | 0 | 78 | 86 | 84 |
| 108.170.243.174 - 3 | 45 | 44 | 0 | 79 | 91 | 83 |
| 209.85.255.145 - 3 | 45 | 44 | 0 | 79 | 88 | 79 |
| ord38s19-in-f4.1e100.net - 3 | 45 | 44 | 0 | 79 | 87 | 77 |
|________________________________________________|______|______|______|______|______|______|
WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider


|------------------------------------------------------------------------------------------|
| WinMTR statistics (PC direct to modem) |
| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
| No response from host - 100 | 6 | 0 | 0 | 0 | 0 | 0 |
| 173-219-232-20.suddenlink.net - 0 | 33 | 33 | 8 | 10 | 17 | 11 |
|173-219-218-106.nlsncmtc01.suddenlink.net - 0 | 33 | 33 | 9 | 10 | 18 | 11 |
| 173-219-218-202.suddenlink.net - 0 | 33 | 33 | 16 | 18 | 25 | 19 |
| 66-76-31-129.suddenlink.net - 0 | 33 | 33 | 16 | 18 | 25 | 20 |
| 66-76-229-162.suddenlink.net - 0 | 33 | 33 | 54 | 58 | 104 | 58 |
| 108.170.246.66 - 0 | 33 | 33 | 27 | 29 | 36 | 31 |
| 142.251.49.157 - 0 | 33 | 33 | 27 | 29 | 37 | 29 |
| 142.251.49.169 - 86 | 7 | 1 | 0 | 27 | 27 | 27 |
| 209.85.254.95 - 0 | 33 | 33 | 27 | 30 | 52 | 30 |
| 108.170.243.174 - 0 | 33 | 33 | 28 | 29 | 36 | 29 |
| 209.85.255.145 - 0 | 33 | 33 | 28 | 29 | 36 | 28 |
| ord38s19-in-f4.1e100.net - 0 | 33 | 33 | 26 | 28 | 35 | 29 |
|________________________________________________|______|______|______|______|______|______|
WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider
 
Check the DNS setting in the router and pc. Are you using something that is different than the ISP DNS?

If you change DNS servers when you're running traces or winmtr tests, you can end up with different routes going to the same endpoint, and you can end up with slightly different endpoints. I would interpret this as the result of using Domain names instead of IP addresses for the trace or winmtr test. Its interesting to see that your tests with an IP address as the destination were not identical. I still wonder if this might be due to different DNS addresses used for the router connected test versus the direct connected test.

When you connect a pc directly to the modem (in bridge mode, which I don't recommend) the pc will use the ISP DNS, as supplied by the CMTS, unless you have DNS addresses hard set in the pc's ethernet adapter settings to something other than the ISP DNS.

When you use a router, the pc will use the router DNS, unless you have DNS addresses hard set in the pc's ethernet adapter settings to something other than the router's DNS. If the router DNS setting is empty, the router should use the DNS address supplied by the CMTS. The follow-on from that is that the connected devices will use the same DNS addresses supplied by the CMTS, thru the router to the connected device.

If you're using an Asus router with DNSFilter set, the router will override the DNS setting in the pc.

As for the WAN IP address assignment for the directly connected pc or router, the WAN IP address that is assigned to any device will come out of the address pool available at the CMTS that you're connected to. Some ISPs run segregated address pools, so that a gateway modem, operating in Gateway mode instead of Bridge mode will end up with a WAN IP address from Pool A, or whatever the ISP uses as a name for that address pool. Flip the modem into Bridge mode, and the connected devices could end up with a WAN IP address from Pool B, or whatever the address pool is called. The WAN IP address is most likely not related specifically to the MAC address of the connected device. The CMTS would check to determine if the MAC address is valid and if so, assign an available WAN IP address to the device.
 
Last edited:
Check the DNS setting in the router and pc. Are you using something that is different than the ISP DNS?

If you change DNS servers when you're running traces or winmtr tests, you can end up with different routes going to the same endpoint, and you can end up with slightly different endpoints. I would interpret this as the result of using Domain names instead of IP addresses for the trace or winmtr test. Its interesting to see that your tests with an IP address as the destination were not identical. I still wonder if this might be due to different DNS addresses used for the router connected test versus the direct connected test.

When you connect a pc directly to the modem (in bridge mode, which I don't recommend) the pc will use the ISP DNS, as supplied by the CMTS, unless you have DNS addresses hard set in the pc's ethernet adapter settings to something other than the ISP DNS.

When you use a router, the pc will use the router DNS, unless you have DNS addresses hard set in the pc's ethernet adapter settings to something other than the router's DNS. If the router DNS setting is empty, the router should use the DNS address supplied by the CMTS. The follow-on from that is that the connected devices will use the same DNS addresses supplied by the CMTS, thru the router to the connected device.

If you're using an Asus router with DNSFilter set, the router will override the DNS setting in the pc.

As for the WAN IP address assignment for the directly connected pc or router, the WAN IP address that is assigned to any device will come out of the address pool available at the CMTS that you're connected to. Some ISPs run segregated address pools, so that a gateway modem, operating in Gateway mode instead of Bridge mode will end up with a WAN IP address from Pool A, or whatever the ISP uses as a name for that address pool. Flip the modem into Bridge mode, and the connected devices could end up with a WAN IP address from Pool B, or whatever the address pool is called. The WAN IP address is most likely not related specifically to the MAC address of the connected device. The CMTS would check to determine if the MAC address is valid and if so, assign an available WAN IP address to the device.
This is great feedback. Thank You. I agree it is very interesting there is an extra hop in the "PC mode". I'm trying to gather up parts I need to make a PC router to test.

I believe assigned DNS was identical in the 2 test modes. But I didn't think to record this information. I'll go back when I get the chance and verify and report back.
 
Check the DNS setting in the router and pc. Are you using something that is different than the ISP DNS?

If you change DNS servers when you're running traces or winmtr tests, you can end up with different routes going to the same endpoint, and you can end up with slightly different endpoints. I would interpret this as the result of using Domain names instead of IP addresses for the trace or winmtr test. Its interesting to see that your tests with an IP address as the destination were not identical. I still wonder if this might be due to different DNS addresses used for the router connected test versus the direct connected test.

When you connect a pc directly to the modem (in bridge mode, which I don't recommend) the pc will use the ISP DNS, as supplied by the CMTS, unless you have DNS addresses hard set in the pc's ethernet adapter settings to something other than the ISP DNS.

When you use a router, the pc will use the router DNS, unless you have DNS addresses hard set in the pc's ethernet adapter settings to something other than the router's DNS. If the router DNS setting is empty, the router should use the DNS address supplied by the CMTS. The follow-on from that is that the connected devices will use the same DNS addresses supplied by the CMTS, thru the router to the connected device.

If you're using an Asus router with DNSFilter set, the router will override the DNS setting in the pc.

As for the WAN IP address assignment for the directly connected pc or router, the WAN IP address that is assigned to any device will come out of the address pool available at the CMTS that you're connected to. Some ISPs run segregated address pools, so that a gateway modem, operating in Gateway mode instead of Bridge mode will end up with a WAN IP address from Pool A, or whatever the ISP uses as a name for that address pool. Flip the modem into Bridge mode, and the connected devices could end up with a WAN IP address from Pool B, or whatever the address pool is called. The WAN IP address is most likely not related specifically to the MAC address of the connected device. The CMTS would check to determine if the MAC address is valid and if so, assign an available WAN IP address to the device.
I used an old 2011 mac mini I had laying around and bare metal installed Pfsense on it. Works very well so far. A core i5 with a samsung SSD is a bit overkill for a router. lol.

repurposed my old netgear router into an AP for wireless. My pings are now back to normal or better. I can only suppose that my ISP is prioritizing data based on the router source. Unless I can find an insider I guess I'll never really know.
 
I think you don't know what you are doing. You need to do everything by IP. DNS should even factor into the test. DNS speed is a different topic.

If you changed QOS for ping or anything like it will have an impact on ping.

Factory reset router and use ping with an IP address. And don't run double NAT when you test.
 
I think you don't know what you are doing. You need to do everything by IP. DNS should even factor into the test. DNS speed is a different topic.

If you changed QOS for ping or anything like it will have an impact on ping.

Factory reset router and use ping with an IP address. And don't run double NAT when you test.
Appreciate you're feedback.

I agree that DNS isn't normally considered a factor in ping results. However, there is a repeatable and recordable difference in the trace route based on the type of device connection being made with the modem. Consumer grade routers all had a consistent different route (along the ISP's servers right before the handoff to Google's servers) and much higher ping from that point forward than when directly connecting a PC to the modem. Furthermore, this wasn't always this way - this just started in the last few weeks after having this connection for years. Nothing changed on my end.

1. I've used domain name as well as direct IP address to that domain and repeated the tests. Same results with both.
2. I have used multiple brand routers with various firmware (ddwrt, freshtomato, factory), cleaned and reset to basic configuration. Same results. NVRAM cleared.
3. I have never used any filtering or QOS, etc. during the tests.
4. Plugging in a computer directly to the modem resulted consistent results.

I don't know what is going on exactly. But I've been careful to use advice like yours and have posted the results.
 

Similar threads

Latest threads

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top