 
comcastisbad01
@ameritech.net
| Terrible provider impacting my gaming and even browsing This is dang true. Yesterday I was gaming with nothing else using my connection. I lagged out of 3 games in about 2 hours for no reason. I was also checking my e-mail and got that server is not responding then magically just started responding. These trash service practices are terrible. I am really annoyed by how monopolistic ISPs are in an area. In Chicago there are 3 major ISPs(ATT, RCN, and Comcast). Depending on your area you have to be MEGA lucky to have 2 of these serving you. I would say about 90% of the areas have only one that serves them. Yeah, no monopoly in the state but they have "turfs" just like the gangs do. | |
|
 | 
FifthE1ement
Tech Nut
join:2005-03-16
Fort Lauderdale, FL
| Re: Terrible provider impacting my gaming and even browsing Actually this has been happening to me for a while now (approx last two months in my area) as now whenever I try and do something as simple as downloading a file I get issues. My connection will start strong and then about 20 seconds into the download it lags out but then starts again and lags out and starts again a cycle which repeats constantly.
This also affects my video viewing as I was watching Fox On Demand (fox.com) and during the video approx 20 seconds in the video becomes choppy and lags every ten seconds. This is ridiculous considering I have the 8mb/down 768kbps/up package and should not be experiencing these issues on a service I pay good money for! So basically in South Florida they are killing any connection which lasts more than 20 seconds (or at least slowing it down).
Below is my last speed test result which is the same no matter which server I use or site I test on. I constantly get error messages when downloading files or playing video feeds and yet my modem connection stats are better than average. Piss off Comcrap!
Speed Test #48922506 by dslreports.com
Run: 2008-04-07 19:38:55 EST
Download: 4021 (Kbps)
Upload: 1576 (Kbps)
In kilobytes per second: 490.9 down 192.4 up
Tested by server: 50 java
User: 2 @ dslreports.com
User's DNS: comcast.net
Compared to the average of 3400 tests from comcast.net:
* download is 45% worse, upload is 29% better
:::.. testmy.net test results ..:::
Download Connection is:: 4606 Kbps about 4.61 Mbps (tested with 5983 kB)
Download Speed is:: 562 kB/s
Upload Connection is:: 1004 Kbps about 1 Mbps (tested with 2992 kB)
Upload Speed is:: 123 kB/s
Tested From:: »testmy.net (Main)
Test Time:: 2008/04/07 - 5:42pm
D-Validation Link:: »testmy.net/stats/id-GFH13EU85
U-Validation Link:: »testmy.net/stats/id-4KXSR7NWM
User Agent:: XXXX [!]
--
»www.xphilez.com | |
|
TIGERON
join:2008-03-11
Pacifica, CA | Just a game what a joke. This is nothing more than a public-relations game. Comcast is not going to change. | |
|
luckystreich
join:2005-11-05
Hartford, CT
 ·NETPLEX
| does this affect me? If I use a subscription IPTV service, receiver plugged into my router....using Comcast HSI.... does this packet shaping affect me at all? Or does it just screw with upload traffic? I can get AT&T DSL @ 6.0 mbps, and will certainly switch if I notice problems. I've just had the IPTV service for 2 weeks now. No issues so far. | |
|
ANWDREW
join:2004-01-21
Sterling, VA
1 edit | I can tell the differance I can tell the difference when playing CS its like I am on an ICE rink strafing. the response time blows then your dead. I have 25/2 and it sucks. | |
|
impala
join:2008-03-08
Clemson, SC
·AT&T Southeast
| did we crash the cs.colorado database? 
CUSystems has a problem
Sorry! This site is experiencing technical difficulties.
Try waiting a few minutes and reloading.
(Can't contact the database server: Too many connections (127.0.0.1)) | |
|
| Ulmo
join:2005-09-22
San Jose, CA
 ·Comcast
·SONIC.NET
1 edit | Re: did we crash the cs.colorado database? Now it's saying:
quote:
A note regarding our findings: Further experiments have led us to believe that our initial conclusions that indicated Comcast's responsibility for dropping TCP SYN packets and forging TCP SYN, ACK and RST (reset) packets was incorrect. Our experiments were conducted from behind a network address translator (NAT). The anomalous packets were generated when the outbound TCP SYN packets exceeded the NAT's resources available in it's state table. In this case, TCP SYN, ACK and RST packets were sent. We would like to thank Don Bowman, Robb Topolski, Neal Krawetz, and Comcast engineers for bringing this to our attention. We sincerely apologize for any inconvenience that this posting may have caused.
| |
|
| | 
fruhead
join:2002-01-29
Montclair, NJ
| Re: did we crash the cs.colorado database? said by Ulmo  :Now it's saying: quote:
A note regarding our findings: Further experiments have led us to believe that our initial conclusions that indicated Comcast's responsibility for dropping TCP SYN packets and forging TCP SYN, ACK and RST (reset) packets was incorrect. Our experiments were conducted from behind a network address translator (NAT). The anomalous packets were generated when the outbound TCP SYN packets exceeded the NAT's resources available in it's state table. In this case, TCP SYN, ACK and RST packets were sent. We would like to thank Don Bowman, Robb Topolski, Neal Krawetz, and Comcast engineers for bringing this to our attention. We sincerely apologize for any inconvenience that this posting may have caused.
Sooooo, how long will this incorrect and now disproven thread title be left up without editing or correction by the original poster?
Karl? Are you there? | |
|
craptastic
| crap it will just to push people to fios or other providers. keep it up comcast! | |
|
Ulmo
join:2005-09-22
San Jose, CA | Is this impacting VPN too? Is this impacting VPN too? | |
|
| 
The WeaseL
Premium
join:2001-12-03
Minnesota
2 edits | Re: Is this impacting VPN too? No because they already said it was a problem with their NAT hardware, not Comcast sending reset packets.
Any bandwidth/performance issues you think you are having are not a result of reset packets (so far). I'm not a fan of Comcast but these researchers jumped the gun on proclaiming Comcast was forging packets for all types of traffic.
--
How lucky am I to have known someone who is so hard to say good-bye to. | |
|
| | 
dadkins
Can you do Blu?
Premium,MVM
join:2003-09-26
Hercules, CA
·Comcast
| Re: Is this impacting VPN too? said by The WeaseL :No because they already said it was a problem with their NAT hardware, not Comcast sending reset packets. Any bandwidth/performance issues you think you are having are not a result of reset packets (so far). I'm not a fan of Comcast but these researchers jumped the gun on proclaiming Comcast was forging packets for all types of traffic. Yeppers! 
--
Think outside the Fox... Opera | |
|
|
Comcastrated
@calpop.com
| It is for me see my post below. I have both Qwest and Comcast. When they started forging RST packets and denying that they were I had to get another provider. I have both hoping that they stop this foolishness.
I run OpenVPN and have my own rack with my own firewall etc. in a data center. When I run on Comcast my connection gets reset periodically and I drop ssh sessions. Sometimes they stay up through a reconnect most of the time they do not. I leave them up overnight and on comcast they always drop, on qwest they always stay up. They stay up for days...
I recently downloaded a large file (which by they way is really hard to do on Comcast without a download manager that will reconnect) I turned on load balancing on my home router so that the (MSDN) download manager which will use four connections could use both qwest and Comcast. I needed a large ISO and I didn't want to wait. Anyway I forgot to turn it off for my IP and I ended up with a VPN over comcast and I kept losing my ssh connections. I turned it off (so that it uses Qwest) and I have not dropped a connection since. | |
|
Comcastrated
@calpop.com
| Forged TCP packets. I'm lucky enough to have two providers. Comcast and Qwest. I wish I could get FIOS.
They are definitely forging TCP RST packets. I've been writing software for over twenty years... I'm working on a backup application. They have been forging these packets since about October '07 in my area and I see RST packets on every port. It does not appear to be as aggressive if it does not look like bit torrent. I've done captures on both sides via Wireshark. I've done this on Qwest and Comcast. I get about ten RST packets an hour on Comcast and about zero on Qwest. Unfortunately it just makes the Comcast service really unreliable.
You can't really write around the problem as the RST packet flag is handled at the TCP/IP driver level in the stack. Even if you could you would have to it on both the client and the server at the driver level... You can detect an RST packet and re-establish the connection though. Most applications like web browsers don't handle the "Connection Reset by peer" error in this way. So a lot of applications just break like games, ssh, web browsers, etc.
One thing I've noticed but I have not confirmed so I could be wrong. Is that it does not matter if you are tunneled. I still drop connections running over VPN on comcast. I'm not sure if it is because they are sending RST packets on my VPN socket and it is reconnecting or something else. My best guess is it is just the VPN being reset and that is the problem. I do know it never happens on Qwest just Comcast. I just don't have enough time to track it down further.
The only explanation I can imagine is that they really really want to have more regulation. They are generating way too much bad publicity. If the regulators don't have enough power to stop it now they will soon enough. I don't understand why they would want those handcuffs on their business. The cost of lobbying, fighting, and ultimately loosing will cost them far far more then taking another path. Their shareholders will suffer greatly. ATT (ma bell) thought they could win too. Regulate the industry and chop them up.
Comcast The fastest dropped connections. It's Comcastic! | |
|
funchords
Hello
Premium,MVM
join:2001-03-11
Washington, DC | Univ. of Colo Retraction -- I agree 100% Based on my email exchange with them yesterday, I can concur that their findings were in error.
Robb Topolski | |
|
dadkins
Can you do Blu?
Premium,MVM
join:2003-09-26
Hercules, CA
1 edit | Uh huh...Thanks again Robb for putting shit straight!
--
Think outside the Fox... Opera | |
|
MyDogHsFleas
Premium
join:2007-08-15
Austin, TX
 ·AT&T U-Verse
·AT&T Southwest
| Everyone who said "It's true" was WRONG and you should admit it.
There are a TON of posts on this story saying "Yes! I was playing Blast-O-Matic 3000 last night, and my computer froze! I use Comcast and this PROVES your story is ABSOLUTELY TRUE!!!"
Now... fess up that YOU WERE WRONG. | |
|
| |
funchords
Hello
Premium,MVM
join:2001-03-11
Washington, DC
·Verizon Online DSL
 ·Skype
| Live has delicious ironies... We would like to thank Don Bowman, Robb Topolski, Neal Krawetz, and Comcast engineers for bringing this to our attention. Don Bowman is a co-founder of Sandvine. Robb Topolski is me. Now that's funny!!!
(Dr. Neal Krawetz is a famous security guru.)
--
Robb Topolski -= funchords.com =- Hillsboro, Oregon
FCC Public Hearing on the Future of the Internet - Thursday, April 17th - Stanford Univ., Calif. | |
|
| SuperWISP
join:2007-04-17
Laramie, WY | Re: Live has delicious ironies... Robb, how do we know that your own claims about Comcast weren't exaggerated or bogus? I'm beginning to see an agenda here: a war on ISPs. | |
|
| |
funchords
Hello
Premium,MVM
join:2001-03-11
Washington, DC
·Verizon Online DSL
·Skype
1 edit | Re: Life has delicious ironies... Yeah, because a war on ISPs makes so much sense. Don't you think that, if I were in a war on ISPs (or Comcast in particular), that I would have kept my mouth shut about Colorado.edu's errant findings?
I'm not playing this very politically wise -- because I'm not a politician. That's why this was a "delicious irony." Politicians would never have let that happen. It is so not because they're dishonest, but because they're on a mission to get or prevent a certain policy.
You, Brett, clearly are a politician. I'm okay with that. A lot of your life has been involved in local, state, and even some national issues.
When I posted the original report, nobody really could know if the claims were exaggerated or bogus. So I took steps to, at the very least, encourage others to test my claims and either validate or invalidate them. My findings were made under my real name -- my own reputation was offered. They were made in a manner such that any competent person (such as yourself and other readers here) could reproduce them, provided that they had access to a Comcast account. And finally, because they eventually were reproduced and validated by both the Electronic Frontier Foundation and the Associated Press.
Thank you for trying to discredit me. I really appreciate that, Brett.
--
Robb Topolski -= funchords.com =- Hillsboro, Oregon
FCC Public Hearing on the Future of the Internet - Thursday, April 17th - Stanford Univ., Calif. | |
|
| | | SuperWISP
join:2007-04-17
Laramie, WY
1 edit | Re: Life has delicious ironies... said by funchords :Yeah, because a war on ISPs makes so much sense. Don't you think that, if I were in a war on ISPs (or Comcast in particular), that I would have kept my mouth shut about Colorado.edu's errant findings?
They don't say what role you played, Robb. Did you actually debunk what they were doing? Or did you egg them on?
said by funchords :I'm not playing this very politically wise -- because I'm not a politician. That's why this was a "delicious irony." Politicians would never have let that happen. It is so not because they're dishonest, but because they're on a mission to get or prevent a certain policy. You, Brett, clearly are a politician. I'm okay with that. A lot of your life has been involved in local, state, and even some national issues. Nonsense. I despise politics and only engage in them as a matter of self defense. You, on the other hand, seem to be trying to stretch the 15 minutes of fame you achieved for pointing out what many already knew: that Comcast was using Sandvine appliances, which inject RST packets to manage traffic. And you're doing that not only by getting political but by participating in Lauren Weinstein's "Network Vigilante Squad." You and others continue to spread FUD and advocate policies that would destroy small and independent ISPs -- the only alternative to the cable companies and telcos. You deserve to be discredited, because you deserved no credit for your malicious activities in the first place. | |
|
bent
not broken
Premium
join:2004-10-04
Loveland, CO
clubs: | Refer Boulder has the best weed in the western hemisphere, outside of Vancouver.
Coincidence? |) | |
|
| 
Topmounter
Sent By Grocery Clerks
join:2001-02-20
Evergreen, CO
·Cox HSI
| Re: Refer said by bent :Boulder has the best weed in the western hemisphere, outside of Vancouver. Coincidence? |) Didn't they ban smoking in the People's Republic of Boulder?
Oh wait... they forgot to include marijuana  | |
|
fruhead
join:2002-01-29
Montclair, NJ
| MVM, indeed! Robb Ruless.
Thanks, you famous personage, you!
(looks for some way to insert "I was talking to Robb Topolski just the other day, and he said..." into next conversation)
 | |
|
Jack2131
@comcast.net
| Researchers at the University of Colorado should be FIRED!! Although Comcast had gotten alot of bad press lately, these "researchers" must had done it on purpose. NAT table overloading is a bunch of crap, it's easily diagnosed, PHd researchers can't find out the problem? UoC should be sued, the researchers, and have their credentials revoked. UoC CS department has such low standards when dealing with ethics, and/or research. Is it so difficult to research one's sources before publishing? Journalists got fired and had their careers destroyed because of things like this! The names of the "researchers" should be published and the national press should get a hold of this story. | |
|
| 
Doctor Four
My other vehicle is a TARDIS
Premium
join:2000-09-05
Dallas, TX | Re: Researchers at the University of Colorado should be FIRED!! You have to wonder if those so-called researchers ever took
any IT classes, like the basics of NAT and so on. | |
|
| | SuperWISP
join:2007-04-17
Laramie, WY
| Re: Researchers at the University of Colorado should be FIRED!! My guess is that the "researchers" were really just a couple of computer science grad students who hoped to garner brownie points by jumping on the Comcast-bashing and ISP-bashing bandwagon. They posted their "results" online in a blog rather than in a refereed journal and submitted them to Slashdot. They were probably hoping that the campus papers would report record traffic to the site where they described their "brilliant" "research."
In the meantime, small ISPs such as the one I operate (and NO ONE ELSE has made an investment in bringing broadband to many of the areas I cover) are set to be strangled by rules stimulated by alarmism related to such bogus claims. The "Save the Internet" loonies are even claiming that throttling back bandwidth hogging P2Pers is somehow denying free speech -- as if preventing kids from downloading pirated pornographic videos would do that. The fact is that it's P2P mitigation and blocking that's saving the Internet from being crippled by these bandwidth hogs. | |
|
| 
gaforces
United We Stand, Divided We Fall
join:2002-04-07
Santa Cruz, CA
| And what about the legal and licensed content providers using BT that lost customers because they were not able to provide service because of forged packets, AYE?
--
There is no greater sign of a general decay of virtue in a nation, than a want of zeal in its inhabitants for the good of their country. ~ Joseph Addison | |
|
DataDoc
My avatar looks like me, if I was 2D.
Premium
join:2000-05-14
Greenville, NC | What ever happened to the Scientific Method? Or having separate verification of the results? When they finally got around to this, they were wrong.
They should have done this sooner. Now they've got egg on their face.
--
KEN LEEEEEEE... tulibu dibu douchoo! | |
|
Jack2131
@comcast.net
| Researcher's Names Does any one know the names of the "Professors" who can't tell the difference between NAT table overload and packet forging?
This REALLY HURTS the consumers in the future. Now when Comcast really forges other types of packets in the future, they can quote this case. Does anyone live near the University of Colorado and can go to CS department's first floor where they have pictures of Professors/Phd candidates, etc... and take a couple pictures of the researchers. Cellphone camera image would be fine.
Many would love to see the "mug" shots of these experts/"idiots". The PHd candidates should have their certificates deferred for a few more years as punishment, as these "Phd candidates" seriously needs more education before being unleashed onto the workforce to harm others with there so called "credentials". Lawyers gets disbarred, doctors have their licenses revoked, and journalists gets canned for serious mistakes/ethical violations, these people should be appropriately punished! Just because a company is crap, doesn't give these people the right to lie in order to further their own careers. | |
|
|
gaforces
United We Stand, Divided We Fall
join:2002-04-07
Santa Cruz, CA | Re: Researcher's Names They made a mistake and then admitted it in public, unlike some cable company on forging packets ... sue em! | |
|
jester121
Premium
join:2003-08-09
Lake Zurich, IL
 ·surpasshosting
·ViaTalk
1 edit | *snicker* I bet the Consumerist and their ilk are thanking their lucky stars they didn't start wailing about this latest "atrocity" too early. Wonder how many articles got trash canned when the truth came out?
Oh, and Karl, you might want to re-queue this in the front page news since it was on the "most viewed" news list for this week. It's only fair, after all.  | |
|
|
|
|
