CODA-4582 - Open Issues for Investigation

Need Help?

That's what we're here for! The goal of the Rogers Community is to help you find answers on everything Rogers. Can't find what you're looking for? Just ask!
cancel
Showing results for 
Search instead for 
Did you mean: 
Reply
I Plan to Stick Around
Posts: 207

Re: CODA-4582 - Open Issues for Investigation


@Datalink wrote:

Here is some food for thought on this.  For anyone running a pc based game, download and install Wireshark. Then, when you are in a game fire up Wireshark and run a short data capture while you're in game, actually doing some activity within the game.  The more activity on your part while engaging another external party, the better.  The goal is to cause/generate UDP packet transfer in both directions.  You probably only need somewhere between 2 to 5 min to capture enough data. Then stop the capture and save the file.  You don't want to run long captures just yet as the file size can grow pretty fast.  I'm sure there is a way to filter data and save that to a file but I havent' looked at Wireshark closely enough to determine exactly how to do that.  I would suggest starting with small files for now and work our way up if necessary.  

 

You don't necessarily have to be a Wireshark guru to read the data, although, it does help, and I'm not a Wireshark guru by any means.  But, you should be able to identify inbound and outbound UDP packets from the display and determine if any of the inbound packets indicate packet corruption, fragmentation or anything else.  

 

You could do a screen capture and send that to Dave via pm as it will contain your IP address and various IP addresses that your pc is communicating with.  Possibly Dave will request the file via pm, don't know at this point.

 

It would be interesting to see what turns up on the data capture.   

 

https://www.wireshark.org/

 

Edit:  Don't post a Wireshark screen capture to this or any other public forum due to the type and amount of IP addresses contained in the capture.  Or, in other words, for your own online security, keep your own data secure. 

 


The only problem with this is, unles you have a managed switch with a monitoring port you're only going to see one side of the traffic, even in permiscuous mode. An alternitive is to man in the middle the connection but then you have to log the connection one way at a time so it isnt exactly matching in and out transactions, and it also causes you to only duplex at 10 base t when doing so.

 

So unless an end user has the equipment to do it, the data collected might not be very useful to @RogersDave

 

 

Resident Expert
Resident Expert
Posts: 6,435

Re: CODA-4582 - Open Issues for Investigation

Yup, absolutely agree, passive tap to log both inbound and outbound would be ideal.  Too bad this never occurs under ideal conditions 😞  You're correct in that it would only show the effects on one side of the traffic which is the inbound traffic.  But, its better than nothing and it would serve as a check to determine if there are issues with inbound UDP traffic for gaming or any other application.  Ideally, you would monitor both sides of the modem for both inbound and outbound traffic to see the before and after effects that are imposed by the modem.  But, that requires a much higher level of equipment and ability to match the packet frames.  Seems to me that the chipset manufacturers should be the ones doing this on a regular basis in order to ensure that the modem operates as it should.  It shouldn't be left to the tail end of the manufacturing and distribution chain to come up with methods to accomplish this. Personal opinion of course 😞



I Plan to Stick Around
Posts: 207

Re: CODA-4582 - Open Issues for Investigation


@Datalink wrote:

Yup, absolutely agree, passive tap to log both inbound and outbound would be ideal.  Too bad this never occurs under ideal conditions 😞  You're correct in that it would only show the effects on one side of the traffic which is the inbound traffic.  But, its better than nothing and it would serve as a check to determine if there are issues with inbound UDP traffic for gaming or any other application.  Ideally, you would monitor both sides of the modem for both inbound and outbound traffic to see the before and after effects that are imposed by the modem.  But, that requires a much higher level of equipment and ability to match the packet frames.  Seems to me that the chipset manufactures should be the ones doing this on a regular basis in order to ensure that the modem operates as it should.  It shouldn't be left to the tail end of the manufacturing and distribution chain to come up with methods to accomplish this. Personal opinion of course 😞


I agree with you, this should be Hitron soing tests on Reported gaming infrastructure and other reported service issues with UDP.

 

I have no problems letting Dave set up some sort of monitor on my Wan connection on his end but i dont even know if thats possible for him to track my I/O in that aspect.

One thing ive noticed pinging download.xbox.com which hits their CDN, is the thing flip flops back and forth between IPV6 and IPV4 and whn it does i get a latency spike.  I wonder if this is also salting the wound so to speak on the issues with gaming CDN's and UDP.

Resident Expert
Resident Expert
Posts: 6,435

Re: CODA-4582 - Open Issues for Investigation

Wonder why it doesn't operate solely on IPV4 or IPV6 instead of switching back and forth?  You should pm Dave with the details and logs if you have any that show that.



I Plan to Stick Around
Posts: 207

Re: CODA-4582 - Open Issues for Investigation

Trying to ponder on a way to send this data to @RogersDave in some sort of meaningful format as my trial of PinPlotter has expired so it only shows real time data and ive just happened to catch it in the ipv4 ipv6 flip flop act a few times today. 

Resident Expert
Resident Expert
Posts: 6,435

Re: CODA-4582 - Open Issues for Investigation

You can see red arrows on the bottom? And if you hover the mouse over the arrow it displays the IP address change?

 

I've seen that over the last couple of days, running a ping test to google.ca where the IP address for google changes, and then changes back.  The first change results in a slightly higher ping time and remains that way until it changes back.  This is all with IPV4 as I have IPV6 turned off for now.   The flip flop from IPV4 to IPV6 and back is a different twist on this.  

 

With the freebie pingplotter, the data will still be there, but you won't be able to access it.  Anyone with the Standard or Pro version will be able to access the file if you export the sample set.  I run the standard version.

 

 

Edit:  If you go Edit .... Copy as Image and then paste that into something like MS Paint, you will see the notes at the bottom of the image which includes the details of the events that are marked by the markers or arrows.  At least that's what the Standard version does.  I would expect the freebie version to do the same, unless they cut that off as well.   So, it should be a matter of waiting until one of these appears on the display, then run the Copy as Image function and dump it into something to see if the notes show up at the bottom left of the image. 



I Plan to Stick Around
Posts: 207

Re: CODA-4582 - Open Issues for Investigation


@Datalink wrote:

You can see red arrows on the bottom? And if you hover the mouse over the arrow it displays the IP address change?

 

I've seen that over the last couple of days, running a ping test to google.ca where the IP address for google changes, and then changes back.  The first change results in a slightly higher ping time and remains that way until it changes back.  This is all with IPV4 as I have IPV6 turned off for now.   The flip flop from IPV4 to IPV6 and back is a different twist on this.  

 

With the freebie pingplotter, the data will still be there, but you won't be able to access it.  Anyone with the Standard or Pro version will be able to access the file if you export the sample set.  I run the standard version.

 

What are you pinging that you see the shift from IPV4 to IPV6?

 

Edit:  If you go Edit .... Copy as Image and then paste that into something like MS Paint, you will see the notes at the bottom of the image which includes the details of the events that are marked by the markers or arrows.  At least that's what the Standard version does.  I would expect the freebie version to do the same, unless they cut that off as well. 


Yep seeing the red arrow and seeing the change from ipv4 to 6.  when it happens it flips and then 5 min later flops. I'll start up the test again and get some data and take a image.

download.xbox.com

 

Further more i am seeing it switch me from local CDN servers and then to the netherlands as well some times, and some tiems seeing my route go through Verizons layer in the states and soem times it doesnt. its realy unstable. 

Resident Expert
Resident Expert
Posts: 6,435

Re: CODA-4582 - Open Issues for Investigation

More than a little strange considering that Pingplotter should be using an IP address.  But, if you are using download.xbox.com as the target, I wonder if pingplotter goes out with a DNS request and gets a different response at some point?  That would show up on a Wireshark log perhaps?  What do you use for DNS servers?  I'm using OpenDNS and see and address changes which are all within the Rogers IP address ranges.  If you're using the Rogers DNS and seeing an address shift from IPV6 to IPV4 and back, thats a little interesting as well.  

 

Just thinking about this I'm using Google.ca as a target.  I'll have to give this a try with an IP address to see what happens.  Maybe this is nothing more than a DNS problem at the DNS servers?  

 

 



I Plan to Stick Around
Posts: 207

Re: CODA-4582 - Open Issues for Investigation


@Datalink wrote:

More than a little strange considering that Pingplotter should be using an IP address.  But, if you are using download.xbox.com as the target, I wonder if pingplotter goes out with a DNS request and gets a different response at some point?  That would show up on a Wireshark log perhaps?  What do you use for DNS servers?  I'm using OpenDNS and see and address changes which are all within the Rogers IP address ranges.  If you're using the Rogers DNS and seeing an address shift from IPV6 to IPV4 and back, thats a little interesting as well.  

 

Just thinking about this I'm using Google.ca as a target.  I'll have to give this a try with an IP address to see what happens.  Maybe this is nothing more than a DNS problem at the DNS servers?  

 

 


I am using Rogers DNS. I decided to stick with it for testing, i typically use googles DNS. 

 

 

THis could very well be a dns issue I agree, I just find it curious as once pingplotter resolves a name, and gets its target it should stick to it unless its forced to relookup?

I'm Here A Lot
Posts: 6

Re: CODA-4582 - Open Issues for Investigation

Hi, 

 

When I found this forum I was happy that I'm not the only one having issues with Gig Internet.

I followed all the steps here:

 

-Changed modem to one with 2 black dots.

-reboot and tested with and without bridge mode

-asked tech to come and check my signal 

-tested with one device, multi devide, waited to have the latest firmware and anything else that I know of

 

but still having issues with download speed.

SpeedCheck.png

my Modem firmware

Hardware Version

1A

Software Version

2.0.10.24

 

also this is what my modem shows in the Docsis wan

Downstream Overview

 

Port ID

Frequency (MHz)

Modulation

Signal strength (dBmV)

Channel ID

Signal noise ratio (dB)

 

1

615000000

256QAM

5.500

11

38.983

 

2

561000000

256QAM

5.800

2

38.605

 

3

567000000

256QAM

5.900

3

37.636

 

4

573000000

256QAM

6.700

4

37.356

 

5

579000000

256QAM

6.400

5

38.605

 

6

585000000

256QAM

6.400

6

38.605

 

7

591000000

256QAM

6.000

7

38.605

 

8

597000000

256QAM

6.100

8

38.605

 

9

603000000

256QAM

5.700

9

38.605

 

10

609000000

256QAM

5.700

10

38.983

 

11

555000000

256QAM

6.300

1

38.605

 

12

621000000

256QAM

5.200

12

38.605

 

13

633000000

256QAM

4.800

13

38.605

 

14

639000000

256QAM

5.800

14

38.983

 

15

645000000

256QAM

4.800

15

38.605

 

16

651000000

256QAM

4.400

16

38.605

 

17

657000000

256QAM

5.700

17

38.605

 

18

663000000

256QAM

6.500

18

38.983

 

19

669000000

256QAM

5.200

19

38.605

 

20

675000000

256QAM

4.800

20

38.605

 

21

681000000

256QAM

5.800

21

38.605

 

22

687000000

256QAM

6.400

22

38.605

 

23

693000000

256QAM

5.100

23

37.356

 

24

699000000

256QAM

5.500

24

37.636

 

25

705000000

256QAM

5.700

25

37.636

 

26

711000000

256QAM

5.700

26

37.356

 

27

717000000

256QAM

5.600

27

37.356

 

28

723000000

256QAM

5.200

28

37.356

 

29

825000000

256QAM

2.500

29

36.387

 

30

831000000

256QAM

2.500

30

36.610

 

31

837000000

256QAM

2.200

31

36.610

 

32

843000000

256QAM

1.600

32

35.780

 

OFDM Downstream Overview

Receiver

FFT type

Subcarr 0 Frequency(MHz)

PLC locked

NCP locked

MDC1 locked

PLC power(dBmv)

0

NA

NA

NO

NO

NO

NA

1

4K

275600000

YES

YES

YES

6.000000

          

 

Upstream Overview

 

Port ID

Frequency (MHz)

Modulation

Signal strength (dBmV)

Channel ID

Bandwidth

 

1

38595609

ATDMA - 64QAM

45.750

3

3200000

 

2

30596000

ATDMA - 64QAM

41.500

1

6400000

 

3

23700000

ATDMA - 64QAM

40.750

2

6400000

 

OFDM/OFDMA Overview

 

Channel Index

State

lin Digital Att

Digital Att

BW (sc's*fft)

Report Power

Report Power1_6

FFT Size

0

DISABLED

0.5000

0.0000

0.0000

-inf

-1.0000

4K

1

DISABLED

0.5000

0.0000

0.0000

-inf

-1.0000

4K

             

 

 anyone have any idea what to try next?

@RogersDave I really appreciate if you can take a look and help me to figure this out. I'm dealing with this since Oct last year!