Since I upgraded to the 150 Hybrid plan in November of 2014, I have been experiencing high ping and packet loss at random times throughout the day. The high pings ususally start occuring after 6pm, and range from 12ms all the up to 500+ ms. I have a long history since Nov 2014 with Rogers tech support. Let me summarize it for you all:
Nov 2014: I complain about low upload speeds. Was getting around 1.8 Mbps upload speeds on testmy.net. Called Rogers tech support. Ticket was escalated to level 2 support, and upload speeds were fixed. I now get 7-8 Mbps stable.
Feb 2015: Start playing video games as well as watching netflix. Noticed lag while gaming as well as buffering issues when streaming videos on the internet. Starting using the cmd prompt and start pinging www.google.ca and get pings ranging from 12ms all the up to 500+ ms. I'm also on the CGN3 modem at this point. Call rogers. They tell me to switch the modem. I swap it for another CGN3, high ping and jitter still exists.
March 2015: Rogers sends a tech to my house and the tech is all confused. Tells me if I'm trying to set up a new service. I advise him that should of been a service call and that I'm not setting up a new connection. He tells me he doesn't have his laptop and that he will check the signal levels at the tower and come back if anything is wrong. I do not hear from him or rogers that day.
What I have tried:
It's weird because the high pings are not constant and occur at random times throughout the day. Even when I do a ping test to www.google.ca, I will get low pings of 13ms, but pings of 300ms+ as well as 5 - 10% packet loss when doing a trial of 1000 tests.
I was wondering if any of you in this awesome community can help me troubleshoot this issue while I wait a week to hear back from rogers. It will be greatly appreciated. I will provide any information that is needed and this could be a great case study for individuals in the future.
Here is some information:
Downstream Signal Screenshot:
Upstream Signal Screenshot:
I look forward to hearing what you guys have to say as well as I promise to keep you all updated with my progress.
Thanks so much and take care!
The CGN3 and CGN3ACR have high, unstable ping times. That is a known problem with the CGN3 and it looks like it has been carried into the CGN3ACR. I was hoping that the last firmware update in Dec 2014 would solve the issue, but unfortunately it hasn't. If you ping the modem a thousand time, you will probably end up with a ping time of 4 to 5 ms with a max ping time around 20 ms. Firmware version 184.108.40.206 has improved the max ping time from 100 ms down to approx 20, but that is still a far from a router ping time of 1 ms or less.
From your description it looks like you have packet losses as well, which points to a Rogers network problem.
Your signal levels and signal to noise ratios are ok, so they should not be the source of any problem.
Can you check the back of the CGN3 and confirm that all connected port LEDs are showing amber which means that the port is connected to the far port at 1 Gb/s. Green indicates that a 100 Mb/s connection is running.
You indicated that the DLINK DAP 1522 is connected as an access point. Is the hardware switch at the back set for "Access Point"? With that setting are you running a wireless network from the 1522 or is it just for wired purposes?
Thanks for the response Data Link!
Computer 1 (my computer) and computer 2 show up as amber. Computer 3 (my brother's computer) is green.
The Dap 1522 hardware switch is set to AP as well as there is one wired ethernet connection connected to it (network hard drive). The DAP 1522 is running a wireless network as well. I set the SSID to be the same as the 2.4 GHZ name as the CGN3, as well as the same password. This allows a user to not have to switch between two routers when connected to the 2.4ghz signal in my house.
Hopefully this answers all your questions correctly.
EDIT: Sorry if I don't respond as fast as I should. I don't receive email notifications when someone posts a message as this account is set up with my mother's email.
EDIT #2: My hardware version is 1A, and my software version is 220.127.116.11
While not changing the SSID name.. in theory DOES allow for not having to change the wireless, etc..
It doesnt always work that well.
I have set up a few enterprise level stuff.. and even in the case of enterprise level AP's.. the handoff DOESNT always work/play nice.
A ) You could give it a try with different names.. and SEE if there is any performance difference. It could be that its fighting for WHICH to connect to at the time, and causing issues.
B) What are the CHANNELS you have set on the CGN3, and your AP? Just making sure of no interfearance there.
All in all.. as most have found.. just some minor ping issues to the ROUTER portion itself... using a 3rd party router overall with the unit in bridge mode, is usually recomended.
Thanks for the reply Gdkitty!
I'm just at work at the moment, but do you really think having the same SSID name for both the CGN3 and AP will cause high ping issues to devices connected by ethernet to the CGN3ACR?
I will also check what channels I have set up on the CGN3ACR and the AP when I get home.
For wireless yes... but wired no.
Having a SMALL ping spike is pretty well known with them in gateway mode.. but your talking a 30ms max spike.. not 300 some odd.
I am GUESSING its something more down the line.. weither street, area, or the closest rogers node, etc.
If its CONSISTANTLY enough high, you could try and do some tracert's to it, and find out WHERE it is.
But if its only ocassionaly an then back to normal quickly.. might not get time for the tracert to find it.
Thanks for the reply Gdkitty.
I'm not even going to worry about maintaining low pings on the wireless clients right now. I just want to focus on getting low pings for clients connected by ethernet.
Rogers tech who came this Tuesday did not even file a report for a senior tech to come in. This is extremely frusturating. I called tech support again today and my pings were really bad, and they will send another level 1 tech to my house to do the same tests... This time, I will make sure the tech creates a report and provides me with an ID once he tells me that it's an area issue for the third time...
Here is a screenshot of the packet loss and high pings I have been receiving (pinged www.google.ca):
The struggle is real!
I can remember the good times when that test would of resulted in low 10's stable. Oh those were good times!
For now, I continue to play the waiting game. Tech scheduled for Monday which will again test my line and tell the exact same thing. Hopefully the tech actually writes a report this time...
Wish me luck! =(
Interesting that you should bring up this topic. Don't know if you were around for yesterdays session with Rogers Internet Product Manager, if not, have a look at the following thread:
There are a couple of interesting items in there. One is the following Samknows report for jan-feb, detailing the results of internet testing of Rogers network:
The second is the statment from the Shaw DSL Reports forum which reads:
The Hitron deprioritizes ICMP traffic causing any ICMP echo (ping) testing to be highly inaccurate.
Ookla speed tests (speedtest.net, speed.shaw.ca) use TCP based latency testing. You will noticed those tests show an accurately low latency.
So the ICMP numbers you see are artificially high, the actual latency for TCP data flow is typical 8ms - 12ms for the first Shaw hops. Unfortunately so many things (network performance testing, online gaming) depend on the ICMP latency figures, so even though the TCP latency is nice and low, the poor handling of ICMP (and maybe UDP) traffic needs to be solved.
Shaw engineering is aware and the vendor is working for a firmware resolution.
So, that is an interesting statement. I asked RogersShane about the latency with the modem and this statement and he indicated that he would follow up with Hitron.
This brings up the whole issue of latency testing via Ping (ICMP echo request) versus UDP and TCP testing. If the Hitron does deprioritize ICMP traffic causing ping testing to be highly inaccurate then we need to find another method of obtaining accurate test results until a firmware fix is implemented. The modems might be working perfectly well, but the ping test results might not indicate that. Digging thru a Samknows doc, it would appear that their latency testing is done via UDP packets, instead of ICMP Echo Requests. So, that alone might account for their results versus what we see via ping tests. I'm just starting to figure out the issue of testing via UDP so I don't have any ideas of what might help in terms of running an accurate test. This might be a starting point, which is PsPing. This is available from Microsoft's Technet;
PsPing is used for latency and bandwidth measurements, so, maybe it might be of some use. Gkitty is probalby the best one here to respond to this, so hopefully he will have a few ideas up his sleeve. If there is a network guru on the forum, please join the discussion. Your knowledge and ideas would be most welcome.
Appreciate the reply Datalink!
I'm going to have to take some time to learn PSPING and to give you some results. Although ICMP pings might be inaccurate, I do notice the poor quality of my connection when I'm getting ping spikes of 300 ms +. For example, I will get buffering issues when trying to watch an online video. As well, I will play a game such as starcraft II and will notice that my actions are taking 2 seconds + to register. Once I notice this behaviour, I quickly go to my cmd prompt and ping google.ca and notice pings that was attached in my previous post. Therefore although ICMP ping tests might be inflated due to the CGN3ACR modem, there still an issue with my connection.
It's weird because I notice the packet loss and high latency issues are severe around 6 - 10pm. After 12am the pings seems to drop a bit, and after 3am, they are even lower. For example here is a snapshot of the result of a 100 trial test performed at 12:50AM EST:
Again I appreciate the information, and hopefully I can give you some results with UDP TCP results soon. I want to do these tests when my connection is practically unusable around 6 - 10pm EST