07-15-2021 03:58 PM - last edited on 07-15-2021 05:09 PM by RogersMoin
I tried unplugging modem and 3rd party router.
This problem seems to happen more frequently. I am now entering the info into my phone (notes) - time/day of disconnection and duration.
It went down Tuesday and now today. I don't recall how long it was out last time. Maybe 30 min or 1 hr total?
I also use a wifi Smart TV.
It has reconnected a few times only to drop 5 mins (estimate) later. Sometimes the speed is normal but often, the connection is at a much lower speed suggesting a problem.
I am wondering if there's an issue.
*Added Labels*
02-01-2021 12:10 PM - edited 02-01-2021 12:15 PM
@doctor80 if you're on a congested node that requires splitting, it will probably take a few months for that to occur. Even if Rogers had ordered new equipment to split the node, it might take several months just to receive any additional equipment given the ongoing pandemic situation. Just pointing this out.
As for the approved modem list site: personal opinion, that site is garbage. Please don't refer to it or refer others to it. That site plagiarizes the work that I and many others did during the Puma 6 debacle, without accreditation to that work. You can really thank Rogers, @xymox1 and @mackey for pushing Intel to finally do something about the Puma 6 latency issues. The real question is, were those issues all finally resolved, including the Denial of Service potential for all Puma modems, which includes the Puma 7 Hitron CODA-4582. I haven't had the chance to test a Puma 6 modem since I moved on to the Puma 7 CODA-4582.
Fwiw, here's the Puma 6 modem list that I built, just over four years ago after trolling thru numerous website and documents.
https://www.dslreports.com/forum/r31134484-
You will recognize many of those modems as they are included in the approved modem list site without accreditation. That site also speaks in rather authoritative language without ever showing any test procedures or test results. In other words, it's all hot air.
If you're interested in reading about the whole debacle from the very start of the DSLReports threads, here they are:
https://www.dslreports.com/forum/r31122204-SB6190-Puma6-TCP-UDP-Network-Latency-Issue-Discussion
Keep in mind that there were complaints scattered throughout the DSLReports site since Jan 2015, but no one connected the dots until the first thread was started. By that time Rogers had already pushed Hitron and Intel in June 2016 to resolve the latency issues, which resulted in the first firmware update in Sept 2016 for the CGN3 (Puma 6) modems.
The current generation of Puma 7 modems do not suffer from the latency design defects exhibited by the Puma 6 modems so your statement that "FYI Coda modems are known to be faulty as they part of the massive PUMA chipset defect that is well known:" is not founded in fact.
While the Puma 6 and 7 modems might share designs to a certain extent, the Puma 7 modem does not exhibit the same latency issues as the Puma 6 modems in any way.
If you were to test an Intel Puma 7 modem such as the Hitron CODA-4582 modem and a Technicolor XB6 (CGM-4140COM) modem which uses a Broadcom BCM-3390 chipset, as an everyday user you would be very hard pressed to tell the difference between the two test results. We're really in the weeds at this point. Yes, you can drive both modems to exhibit higher than normal latency by running a simultaneous high speed ping test and a speed test, and that is one of the few ways that you can tell the two modem apart.
Fwiw, I've been running a CODA-4582 in bridge mode for the last four years, same modem from day one and I have no complaints other than the eventual possibility of a successful DOS attack. I haven't tested for that with version 7.x, one of these days I should do it just to see what happens. That 4582 has been rock solid running in Bridge mode and I've had no complaints from anyone in my family. I've been using an Asus RT-AC68U, followed by an RT-AC86U, soon to be replaced by an RT-AX86U. I've never had any issues running those routers behind the 4582 and I normally run two routers connected to the modem, one of which I use for test purposes.
The big question is whether I'll feel the same way when OFDMA is enabled for our CMTS, resulting in an upstream DOCSIS 3.1 channel. That appears to be the crux of the ongoing problems, in addition to any neighbourhood node and CMTS load issues.
Ok, nuff said. Please drop the approved modems site. It has no bearing on the current situation or Puma 7 modems in use today.
02-01-2021 12:15 PM - last edited on 02-01-2021 12:26 PM by RogersCorey
Hello there,
We have been having issues with our internet, everything from bad upload to short disconnects and degraded speeds and worst of all high ping. we have had couple of techs out and open tickets to monitor our internet, sometimes I get told its a bad signal and they send a tech out, he tests the cable puts some new connectors on and leaves.
Our most recent tech visit the guy was outside, replaced the connector at the box outside and left with remarks that, he "fixed it". Sure enough I was getting proper upload speed, however it does not seem to stay like this for very long, so after his visit I rebooted the modem and checked the signal levels and they has not changed, I believe it maybe an underground cable, the reason we are getting these issues.
I am having the same issue from November 2020 tech. support replaced modem twice with no improvement, and Rogers does not do anything about it, their solution is to restart the modem and refresh signal. Call support is a complete waste of time, as staying on phone for more than hours long and at the end rebooting/replacing modem, monitor for weeks and then advised to book a tech to visit. So far two tech. visited and mentioning that everything is ok at home, field team need to take a look and he will be going to open ticket for that, but no update to customer. Another tech. visit is scheduled today, let see what he is doing, but I'm pretty much sure, he is going to follow same steps as other tech. did in the past. And at the end issue is still there, it's very frustrating situation, when kids doing online schooling.
Any input/suggestion will appreciated. Below is the screenshot from modem DOCSIS WAN. Thanks.
02-01-2021 12:54 PM - edited 02-01-2021 12:56 PM
@JK6399 here's the plot of your upper DOCSIS 3.0 signal levels:
1. They should be a flat line at 0 dBmV
2. They should not exceed 10 dBmV despite the advertised operating range of - 15 to + 15 dBmV. Realistically, from what I've seen on the Rogers network, they should not exceed +/- 7 dBmV. Outside of that range, customers usually experience problems with their service.
3. Having said that, the modem is using an OFDM downstream channel which runs from approx 350 Mhz up to 500 Mhz. So, there is a signal rise in that range.
One of the moderators will have to approve the image so it can be seen publicly.
4. The bigger problem at the moment is probably the fact that an upstream OFDMA channel is enabled and running. That appears to be causing problems for a large and growing number of customers. That OFDMA channel runs below 50 Mhz, on the left hand side of that plot, where the normal QAM upstream channels are running. Have a look at the following posts:
https://communityforums.rogers.com/t5/Internet/Internet-Keeps-Dropping/m-p/475923#M65432
https://communityforums.rogers.com/t5/Internet/Intermittent-disconnects-CODA-4582U/m-p/475918#M65431
02-01-2021 12:55 PM
Update from my end. Had another technician come out and look at the cables in the box on the outside of the house, He apparently disconnected the cables and reconnected them so fast that my internet on the inside of the house didn't go out, but he said that the issue should be fixed. Queue 3 hours later the problem persists so I chatted with another agent for him to be incredibly rude to me. He factory reset my modem from his end and said everything should be up and going so I told him that theres no issue at this very second but it'll probably go out again and I had mentioned this forum post and his response was "I'm sure the forum is full of such issues, I mean nobody would go onto a service forum to ever say that everything is fine. If you want I can have another technician do a follow up for you however if no faults are found there would be a $50 charge for the visit."
This was the first time I had been informed of such a policy and asked to speak to a senior technician or his manager and he just like refused to transfer me and gave me another number to call so after arguing with him for like 10-15 more about how stupid this all is I called and the technician I talked to then had me on the phone for awhile and told me that there was "flapping" in the line and now they are sending another tech out tomorrow to fix that problem supposedly.
Really disappointed in Rogers costumer service and their service in general at this point.
02-01-2021 01:22 PM
If Rogers had charged me $50 every time a technician came out and found "no problem", I would be out $650 and still have broken internet.
Don't accept that.
02-01-2021 01:47 PM - edited 02-01-2021 01:48 PM
Just here to throw in my DOCSIS signal levels, which I assume are not where they should be.
Is there any way I could directly point this out to a tech?
Downstream Overview
Port ID | Frequency (MHz) | Modulation | Signal strength (dBmV) | Channel ID | Signal noise ratio (dB) |
1 | 591000000 | QAM256 | 8.000 | 7 | 40.366 |
2 | 597000000 | QAM256 | 8.099 | 8 | 40.366 |
3 | 603000000 | QAM256 | 7.900 | 9 | 40.366 |
4 | 579000000 | QAM256 | 7.400 | 5 | 38.983 |
5 | 585000000 | QAM256 | 7.800 | 6 | 38.983 |
6 | 279000000 | QAM256 | 11.199 | 1 | 40.366 |
7 | 609000000 | QAM256 | 7.400 | 10 | 40.366 |
8 | 615000000 | QAM256 | 7.300 | 11 | 40.366 |
9 | 621000000 | QAM256 | 7.300 | 12 | 40.366 |
10 | 633000000 | QAM256 | 7.300 | 13 | 40.366 |
11 | 639000000 | QAM256 | 7.199 | 14 | 40.366 |
12 | 645000000 | QAM256 | 7.199 | 15 | 38.983 |
13 | 651000000 | QAM256 | 6.900 | 16 | 40.366 |
14 | 657000000 | QAM256 | 6.500 | 17 | 38.983 |
15 | 663000000 | QAM256 | 6.000 | 18 | 38.983 |
16 | 669000000 | QAM256 | 5.800 | 19 | 38.983 |
17 | 675000000 | QAM256 | 5.699 | 20 | 38.983 |
18 | 681000000 | QAM256 | 5.800 | 21 | 38.983 |
19 | 687000000 | QAM256 | 5.900 | 22 | 38.983 |
20 | 693000000 | QAM256 | 5.900 | 23 | 38.983 |
21 | 699000000 | QAM256 | 5.900 | 24 | 38.983 |
22 | 705000000 | QAM256 | 6.000 | 25 | 38.983 |
23 | 711000000 | QAM256 | 6.099 | 26 | 38.605 |
24 | 717000000 | QAM256 | 5.900 | 27 | 38.983 |
25 | 723000000 | QAM256 | 5.800 | 28 | 38.605 |
26 | 825000000 | QAM256 | 5.599 | 29 | 38.983 |
27 | 831000000 | QAM256 | 5.500 | 30 | 38.605 |
28 | 837000000 | QAM256 | 5.400 | 31 | 38.605 |
29 | 843000000 | QAM256 | 5.400 | 32 | 38.605 |
30 | 849000000 | QAM256 | 5.699 | 2 | 38.983 |
31 | 855000000 | QAM256 | 5.400 | 3 | 38.605 |
32 | 861000000 | QAM256 | 5.000 | 4 | 38.605 |
Receiver | FFT type | Subcarr 0 Frequency(MHz) | PLC locked | NCP locked | MDC1 locked | PLC power(dBmv) |
0 | 4K | 275600000 | YES | YES | YES | 9.800003 |
1 | NA | NA | NO | NO | NO | NA |
Port ID | Frequency (MHz) | Modulation | Signal strength (dBmV) | Channel ID | Bandwidth |
1 | 32300000 | 64QAM | 34.770 | 3 | 6400000 |
2 | 38700000 | 64QAM | 35.520 | 4 | 6400000 |
3 | 21100000 | 64QAM | 33.010 | 1 | 3200000 |
4 | 25900000 | 64QAM | 33.020 | 2 | 6400000 |
5 | 0 | QAM_NONE | - | --- | 1600000 |
6 | 0 | QAM_NONE | - | --- | 1600000 |
7 | 0 | QAM_NONE | - | --- | 1600000 |
8 | 0 | QAM_NONE | - | --- | 1600000 |
Channel Index | State | lin Digital Att | Digital Att | BW (sc's*fft) | Report Power | Report Power1_6 | FFT Size |
0 | OPERATE | 0.2359 | 9.6377 | 9.6000 | 40.2815 | 32.5000 | 2K |
1 | DISABLED | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 2K |
02-01-2021 02:35 PM
02-01-2021 02:45 PM
@Datalink Thank you for replying back to me. Finally some valuable information.
For the node segmentation, I just learned that even though Node segmentation was set on my node, at the end Rogers decided to cancel it as it seems that the issue is not in high utilization reason.
It made completely sense for me, because I know how high utilization acts.
In my specific case I see ALWAYS the same pattern:
1. more than 95% the issue happens during Zoom call
2. The issue ALWAYS happens for the duration of 6 minutes
3. I never loose the actual internet connection, however it's enough to drop the Zoom call
4. I replaced already 4 or 5 CODA modems with each other being pretty much the same in behavior
5. My internal wiring, DMARC point checks, bridge mode to router, new router - everything was checked, RG 6 connectors checked and replaced.
6. Outside checked , Low SNR excluded , signals issue excluded.
We are barking here on the wrong tree, and this is ingoing at least from October.
Is CODA a good modem ? Maybe, to its specified needs and capabilities.
Is CODA capable to handle OFDM introduced in 3.1 DOCSIS rollout ?
I don't know.
I called Tech Support and asked them to turn it off for my modem. Of course they did not have a clue on what I'm talking about.
Bottom line, what we have is bad performing internet and no progress since Oct 2020.
I'm being tossed from left to the right and nothing is being done.
And now I'm getting offered to pay more for the XB modem. I disagree to that approach.
I want to get to the bottom of this.
Take your time, if you want I can assist with whatever you need to test.
But believe me the issue is wide spread.
I used to work as a Headend Technician from DOCSIS 2.0 era. I have experience.
Let's progress here.
If this is not the line, not the wiring, not the CMTS, not the router - than let's exclude the modem .
I'm ready to do whatever you guys need , because I'm sick and tired of this issue not moving anywhere.
Please do understand my side as well.
Thank you
02-02-2021 12:46 AM
02-02-2021 09:53 AM
Tested yesterday ping from laptop to CODA (192.168.0.1) & from CODA to 8.8.8.8. Could observe the issue at around 6:01 PM.
The ping to CODA was working while ping from CODA to 8.8.8.8 experienced delays.
02-02-2021 10:13 AM
Latest from my situation....
Called and spoke with a manager on support, he confirmed there was a overcrowded node and several timeouts were happening on my modem. Told me to ignore the OFDMA information as it's something new and is in spec. Although didn't know what the values for it to be in spec were when asked.
He did send in a new ticket to Maintenance which I got the text confirmation of the ticket number to investigate splitting the node?
Sadly, this morning I had a reply to the text saying the ticket had been closed as no issues had been found and if I needed help to call back.
"Thank you for reporting a problem with your Rogers Internet Service. We have been unable to identify an issue at this time. If you're still having an issue, please contact us and quote ticket number XXXXXXX."
Here we go again.
02-02-2021 10:52 AM - edited 02-02-2021 10:55 AM
Doubt it is node congestion. I've had my node monitored too and have been told my area is far from congested. Yet all of us in this thread have the same problem. And each of us just seems to keep going in circles with various levels of technicians and maintenance who keep closing tickets and denying any issues exist.
I ended January with 133 disconnects in the month (recorded via PingPlotter). That's pathetic.
02-02-2021 11:20 AM
@Datalink @RogersCorey @RogersTony @RogersAndy
Guys, as you can see, this isn't progressing nowhere.
We have here pretty much solid use case where node congestion, internal and external conditions that could affect the discos EXCLUDED.
Let's be more proactive here!
I and I believe many more of us will happily be able to cooperate with debugging this problem on our end.
Can we get some sort of debug? Get the INTERNAL Coda logs so this can be investigated in PROD?
Do you need to install a special debug version on my modem? Go ahead !
Talk to me. Talk to us.
Let's start from somewhere . Let's stop wasting time on apologies, wasted tickets, pseudo node congestions.
Let's progress.
02-02-2021 04:29 PM
3 network disruptions for me today. Around 9am, 1:10pm and 4:17pm.
02-02-2021 05:35 PM
Had another technician come out today after talking to someone on Sunday who said my connection was flapping. This technician today didn't even bother to call and tell me that they were here but called once he was done looking at the box outside and told me that they had to put in a work order for someone to come and fix it which would be done in the next 24 hours. I don't think the agents who schedule these technicians even bother telling them why they are being sent out....
First one when my modem died 3 weeks ago came and didn't even have a modem with him to replace my other one. The rest of them have called to ask me what the problem is. The guy today didn't but I can't say I understand why they wouldn't send someone to fix the problem since the agent I talked to on the phone seemed to know what the problem was. I'm getting really tired of having to deal with this.... Does Rogers just expect all the people having this issue to give up at some point with out a fix?
My internet has also went out around 10am, 1pm, and 4pm but the issue hasn't been fixed so I'm sure it'll continue to go out and get worse as time goes on.
02-03-2021 10:59 PM
Well this is a fun new symptom in addition to the 6 drops today:
02-03-2021 11:13 PM
@88dalejrfan was that during one of the dropouts?
02-03-2021 11:35 PM - edited 02-03-2021 11:37 PM
@Datalink- no, just three hours of wild latency causing me to have blips in a live YouTube eSports broadcast. When it "drops" (my normal problem), the whole PingPlotter graph consists of vertical red bars indicating 100% packet loss. This was just some massive slowdown (albeit there was one "drop" in that 3 hour window that lasted 15 seconds -- you might be able to make it out from the graph). Rebooting the modem made the latency issue go away... for now. But the daily drops of 5-15 times continue. I'm so close to giving up.
02-03-2021 11:57 PM - edited 02-04-2021 01:44 AM
Yup, this is incredibly frustrating to watch. Its the same type of situation seen five years ago when the Intel Puma 6 modem debacle started to blow up in Intel's face. It would be very sad to learn that there's yet another Puma hardware or firmware design issue afoot and this time around Rogers is in full denial, rather than being proactive as it was five years ago. The Puma 6 modems are the Hitron CGN3xxxx series modems. The Hitron CODA-4582 and XB6 (Arris TG-3482ER) are Intel Puma 7 modems.
02-04-2021 06:48 AM
Had the third technician today, he said everything is fine in the house, and the issue is with the Hitron modem, and that they will disable OFDMA for my modem within 24-48 hours.
Hopefully this will solve the problem (assuming they actually disable OFDMA).
02-04-2021 08:56 AM
Is the new modem included in the Ignite service immune to the issues discussed in this forum?
I don’t mind shelling out few extra dollars if it’s going to bring back a stable connection and I can work without interruptions.