Another ham-fisted attempt at traffic shaping

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
Highlighted
I'm Here A Lot
Posts: 6

Re: Another ham-fisted attempt at traffic shaping

Right, I was referring to the people, not the hardware. Smiley Wink

Highlighted
I've Been Around
Posts: 1

Re: Another ham-fisted attempt at traffic shaping

Dear Chris,

 

Would you mind then explain me why when I try to push and download a file to my NJ server using the default utorrent port I get a max transfer rate of 5-10 kbps, when I switchh the port to the same port used by windows update or ssh then the transfer rate jumps to 4mbps down and stick there for as long as I let it...???

 

More so... I tested a torrent download using default ports and it would jump to 4mbps then within 1 minute it to drop to 5-10kbps I just switched the port and watched it jump back to 4mbps to fall again 60 seconds later to 5-10 kbps so I used one of my US servers running on a OC4 to act as a proxy and encrypted the data between my machines and the server and proxies through it running 2 instance of utorrent one through the proxy and the other straight to the net...

 

The "undetectable" tunneled connection downloaded at 4mbps steady while the other couldn't get pass 10kbps after 10 minutes. DONT tell me it was the hosts I was downloading from they were both downloading from the same sources at the same time only one was undetectable to traffic shapping software and the other was... both were fired from the same machine using the same firewall settings on linux. I have 4 monitors running from the same machine and one is dedicated as security console I run SNORT IDS and real time traffic monitoring I see packets by packets what come in and out of my house network and machines I have no virus or anything that could cause interference beside and exterior source such as traffic shapping...

 

So I went and used various fun tools I have to detect shaping and here's something fun... What do you know first test I run...

My 30mbps connection is throttle of 32mbps for P2P traffic... While other traffic when through perfectly fine... DO NOT tell us your not throttling...

 

I work for a major canadian ISP myself and I speak to about 20 rogers customers a night that use hour VOIP services and I WILL have them ALL run this test and gather the results to log a complaint to the CRTC within 5 business days if now rectified.

 

 

DiffProbe release. January 2012. Build 1008.
Shaper Detection Module.

Connected to server 4.71.254.149.

Estimating capacity:
Upstream: 10153 Kbps.
Downstream: 198713 Kbps.

The measurement will take upto 3.0 minutes. Please wait.

Checking for traffic shapers:

Upstream: No shaper detected.
Median received rate: 9930 Kbps.

Downstream: Burst size: 18758-20638 KB;
Shaping rate: 32984 Kbps.

 

 

 

 

Highlighted
Resident Expert
Resident Expert
Posts: 14,233

Re: Another ham-fisted attempt at traffic shaping

I would guess if anything, it's not doing PACKET based shaping, as what they were doing before. Any port you used would have degraded/shaped results.
Likely if anything then based solely off the port.

Since day one of torrenting, I have never used the default ports... More for my own personal choice. For better performance, I fully allow them through port forwarding, but that's again one reason why I use a different #, so less prone to people probing the standard ports on my ip.

As soon as the throttling was 'removed' a year + ago I noticed a huge performance difference. Prior I only got the speeds your listing.. But I think the slowest I see now is around 500k, usually sits between 2-6mbps (depending on the torrent).

Curious as to what port you are using? I could test on my machine.

(Also, this is primarily a user to user forum, just Rogers moderated. Don't expect an answer from them, this isn't the proper venue :(. You can try via fb/twitter, or call in and ask to escalate it to a manager. )


Highlighted
Social Media Specialist
Social Media Specialist
Posts: 81

Re: Another ham-fisted attempt at traffic shaping

Hi jdunphy, we phased out traffic management on P2P filesharing in December 2012. If you want to provide more specifics on which ports you're using, I'll share this with our network team and ask them to take a look. I assume you're using our Rogers Hi Speed internet product, correct?

@Rogers_Chris
Highlighted
Resident Expert
Resident Expert
Posts: 14,233

Re: Another ham-fisted attempt at traffic shaping

jdunphy, i did run the tool (before the link was removed.. not sure why...)

 

Ran it on my connection here.

I am on the 25/2 internet connection

Showed upstream at 2.1 (so correct), and downstream burstable 9-101.

It DID show shaping.. but shaping to 27.. so shaping it down, to what my regular internet level to be from its max burstable level.



Highlighted
I Plan to Stick Around
Posts: 98

Re: Another ham-fisted attempt at traffic shaping

This seems to be the common theme here - anytime someone has any problem where their internet connection isn't working exactly as they think it should they assume Rogers is traffic shaping.

 

In the end these threads always end up in the same place - user or transient error.

Highlighted
Resident Expert
Resident Expert
Posts: 14,233

Re: Another ham-fisted attempt at traffic shaping

I mean, its not to say that rogers doesnt have issues.... people run into things like CONGESTION, etc.. where their connection can slow down to a CRAWL for even webpages, etc..

And while these are things that rogers definately should fix... there are people who have complained of traffic shaping, etc for it.

its VERY unlikely to be shaping.. and more usually ANOTHER rogers problem slowing it down.

 

(or as you said.. transiant error... a problem along the way.  I help out on the WoW forums sometimes.. and the number of time you see someone complaining of LAG in the game.. they blame BLIZZARD.  Analysing  tracerts, pathping, etc.. usually leads to be either ISP.. or a NODE somewhere inbetween.  There is nothing that the ISP or endpoint can really do in the end then, short of changing ALL the routing statements to avoid that node)



Highlighted
I'm a Reliable Contributor
Posts: 359

Re: Another ham-fisted attempt at traffic shaping

"P2P is the ONLY traffic shaping rogers has ever done."
I have experiencing something else which looks suspicious. While doing FTP transfers to a server in Europe after uploading about 5-6GB my system would not be able to sustain the FTP connection. This issue seemed to go away at midnight and I saw it happen at least twice. My theory is that after uploading about 5GB in a day someone is choking off my connection. Is it Rogers? I don't know but it sure looks suspicious.
Highlighted
I Plan to Stick Around
Posts: 98

Re: Another ham-fisted attempt at traffic shaping

Thing with europe is that there are so many hops between here and there that it really could be anyone.

Do you have the same issue to non-european servers?
Highlighted
I Plan to Stick Around
Posts: 152

Re: Another ham-fisted attempt at traffic shaping

Agreed - I would say the first thing to do is to test that same file between someone with a local FTP server - If that works, it would be something way upstream that's causing things to drop.