Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

AT&T Wireless traffic shaping apparently making some websites unusable

adriano.fyi

Post reply on HN
Print view View original

191–200 of 305 posts

Re: AT&T Wireless traffic shaping apparently making some websites unusable

#191

Post by joecool1029 »

LoganDark wrote: post #137
baobrien wrote: post #133

Earlier quoted context omitted.

I bought my own 3rd party cable modem for Xfinity and it works fine.

Well, in my situation/ISP it did happen. Though it was a while ago so I don't remember too many specifics.

Two guesses:

1. You had a modem that wasn't on their approved device list. It needs to be on this list to receive firmware customized to their network parameters. https://www.xfinity.com/support/devices/

2. (the most likely scenario) You were switching the macaddr of your router's WAN port every time you connected. For IPV4 Comcast was always set up to only allow one macaddr to bind to a public address at a time, you needed to trigger a dhcpcd release on the old macaddr before reconnecting or clone the mac of the old device if switching.

Re: AT&T Wireless traffic shaping apparently making some websites unusable

#192

Post by bagels »

crazygringo wrote: post #34

But why would AT&T be throttling Cloudfront requests specifically? That's not traffic shaping, it's just dumb. Traffic shaping is slowing down types of traffic like torrents or video in order to prioritize regular sites. But this is degrading regular sites that are usually what benefit from traffic shaping if there's overall network congestion. I'm not questioning anything the author reports, it all seems extremely w…

They are probably targeting video content on CloudFront, and it's easier/cheaper to do by domain or ip vs content inspection (when even possible due to encryption).

I experienced the same thing when building a large scale video service, but with Comcast. I did extensive testing to isolate it to any direct connection to aws (CloudFront or other aws services).

Re: AT&T Wireless traffic shaping apparently making some websites unusable

#193

Post by tiedieconderoga »

dspillett wrote: post #120
neurostimulant wrote: post #75

Earlier quoted context omitted.

Now that my ISP bundles Netflix subscription into their internet plans, access to netflix and fast.com now practically saturate the fiber link, while before it was outright blocked. Hooray for no net neutrality I guess. Another fun part: when netflix IPs was blocked by this ISP, it's pretty much impossible to use netflix because the only way to get around the block was to use VPN, but netflix itself blocks VPN access…

> Now that my ISP bundles Netflix subscription into their internet plans, access to netflix and fast.com now practically saturate the fiber link They may have an ISP-local netflix cache, from Netflix themselves not some home-grown hack, so they can achieve that with some reliability without it costing as much as it otherwise would for bandwidth peering.

Netflix distributes "red boxes" to ISPs which are exactly that: edge content caches.

(PDF): https://openconnect.netflix.com/Open-Connect-Briefing-Paper....

Re: AT&T Wireless traffic shaping apparently making some websites unusable

#194

Post by kevin_nisbet »

acaloiar wrote: post #154
kevin_nisbet wrote: post #111

Taking a quick 5 minute look at the packet capture attached to the post, it looks to me like this is likely traffic shaping. While it's always difficult to be 100% sure from just the client side capture, the capture looks relatively clean of errors, and the amount of data in flight doesn't appear to even approach the advertised window. We're getting some merging of segments likely from a segment offload, but I doubt…

Thanks for the feedback. I was hoping to see more of this sort of analysis from the HN crowd :) There's one other person from HN doing some analysis with me via email. Maybe something will come of it. You both came to the same conclusion based on the pcap data. They're also analyzing the iphone tether pcap data I just provided in an update.

You're welcome.

I took a very quick look at the iPhone capture. I don't see any notable issue (which is expected since this is "working" behavour), and the main noticeable item in the comparison is an apparent lower latency to the server. This appears to match the speculation I saw in the updates as well.

One thing you might be able to try, if you're able to change the APN on the hotspot, is to see if you're able to connect with a different APN from the consumer side. This is pure speculation, but what some carriers do is use a different APN for their business services, that will route to enterprise only packet gateways. Some carriers may not necessarily disable access to the consumer APNs though that might go through a different P-GW that's closer to you. The iPhone looks like it's using a path about 120ms shorter than the hotspot. And just by chance, going through a different P-GW you might hit different traffic shaping equipment or configuration.

Edit: although be aware that the carrier could also charge different APNs separately. I have no idea what AT&T does.

Re: AT&T Wireless traffic shaping apparently making some websites unusable

#195

Post by washadjeffmad »

dboreham wrote: post #119
Animats wrote: post #96

Remember how 5G was supposed to support "the metaverse"?[1] I've been working on a high performance metaverse client. All Rust, all multi-threaded. Designed to max out a gamer PC. Supports Second Life and Open Simulator. Second Life had a reputation for being sluggish. Fixing that. I'm pulling content from the servers at 200Mb/s, sustained. 400Mb/s in tests, but don't need to go that fast. The servers (AWS front-ende…

Unlimited is a term of art that means "limited".

"Un limited." It's Spanish, probably.

Re: AT&T Wireless traffic shaping apparently making some websites unusable

#196

Post by Scalene2 »

alexpotato wrote: post #177

Back in the early 2000's, I worked for a small company that was hired by the law firm representing one of the big cable companies (let's call them Company A). Company A was suing another big cable company (Company X) b/c A claimed that X was throttling customers without telling them and advertising what were in effect unthrottled bandwidth numbers. In order to prove this, the company I worked for went out and hired p…

A big part of the reason that things didn't change is because that lawsuit and others like it ended in a settlement. If this behaviour and data was established as part of the judicial/public record, it would have been much easier to address.

Re: AT&T Wireless traffic shaping apparently making some websites unusable

#197

Post by blincoln »

kevin_nisbet wrote: post #111

Taking a quick 5 minute look at the packet capture attached to the post, it looks to me like this is likely traffic shaping. While it's always difficult to be 100% sure from just the client side capture, the capture looks relatively clean of errors, and the amount of data in flight doesn't appear to even approach the advertised window. We're getting some merging of segments likely from a segment offload, but I doubt…

Could it not also be the problematic traffic getting routed to a Cloudfront node that's overloaded or otherwise misbehaving?

Years ago, I had to troubleshoot a very mysterious issue where sometimes a huge percentage of the PCs in the company I worked for would have trouble loading the company's e-commerce site.

Everyone was going through the same array of 2-3 proxy servers, and the e-commerce site had Akamai caching in front of it. The proxy servers would hold onto a DNS lookup result or a TCP connection for longer than they were supposed to, and so they'd continue routing traffic to the old IP address even after Akamai had rotated a node out of service.

Re: AT&T Wireless traffic shaping apparently making some websites unusable

#198

Post by egberts1 »

anonymousnotme wrote: post #158
avianlyric wrote: post #58

Earlier quoted context omitted.

> Said the other way around: fast.com is great to see actual throughput from Netflix as opposed to some fake throughout from a dedicated speed test site for the exact same reason. Yup. This is pretty much the reason Netflix created fast.com. They wanted a speed test service that couldn’t be gamed by ISPs. Many ISPs will prioritise traffic to know speed test services (like Ookla’s speedtest.net), making their services…

It definitely looks like testing sites are prioritized. The fastest download speed that I have gotten is maybe 7 MB (bytes) per second; generally it is 2-5 MB per second. The speed test sites generally get 100 Mb per second dowload. In general the best I seem to get is about half the speed of the speed test sites. To me the real speed of the ISP is how fast one can download something one wants, not the result of a te…

MB = megabyte

Mb = megabits

1 byte = 8-bit

Re: AT&T Wireless traffic shaping apparently making some websites unusable

#199

Post by albert_e »

So not just traffic shaping but also pretty a heavy moderation of discussion on that topic? Interested to see what justification AT&T comes up with for suppressing these posts on their forum and marking them as 'private' without any explanation to the original poster at least.

Re: AT&T Wireless traffic shaping apparently making some websites unusable

#200

Post by acaloiar »

kevin_nisbet wrote: post #194
acaloiar wrote: post #154

Earlier quoted context omitted.

Thanks for the feedback. I was hoping to see more of this sort of analysis from the HN crowd :) There's one other person from HN doing some analysis with me via email. Maybe something will come of it. You both came to the same conclusion based on the pcap data. They're also analyzing the iphone tether pcap data I just provided in an update.

You're welcome. I took a very quick look at the iPhone capture. I don't see any notable issue (which is expected since this is "working" behavour), and the main noticeable item in the comparison is an apparent lower latency to the server. This appears to match the speculation I saw in the updates as well. One thing you might be able to try, if you're able to change the APN on the hotspot, is to see if you're able to…

I considered switching APNs, but didn't want to run afoul of any AT&T rules.

I went ahead and gave it a try anyway, switching from `broadband` to `NXTGENPHONE`. I wasn't able to authenticate with `NXTGENPHONE`, so I switched it back to `broadband`.

After re-authenticating, I'm no longer being throttled. I'm both happy and deeply dissatisfied with this outcome because I think now I'm less likely to get to the bottom of it.

I'll add this to my updates and keep monitoring the situation.

Thanks, I think!

Post reply on HN