https://www.bufferbloat.net/projects/codel/wiki/ actually points to RFC8289 twice.

Boy have a lot of issues accumulated here. I wish we were not stuck on an ancient hugo.

Please add reasons for why we need inbound shaping with CAKE if we are to use 'besteffort' + 'wash'.

Thanks.

I believe that you should make a kickstarter/similar website campaign for being able to create an even faster wifi,and to remunerate you because you have merit!
thank you for making a better world.
<3

For kernel >= 4.9 maybe it should be suggested to add this in sysctl:
net.ipv4.tcp_congestion_control = bbr
And this in /etc/modules:
tcp_bbr
Maybe I am wrong though... but if thats the case, then it could be mentioned in the webpage why I am wrong so others don't ask themselves this same question.
Thanks a lot!

The import scripts try to resolve links between wiki pages, and it mostly works, but there are a bunch that didn't resolve. These probably need to be fixed up manually. They appear marked with <link> tags in the markdown source, so grepping for these and fixing them should be fairly straight forward.

https://www.bufferbloat.net/projects/codel/wiki/RRUL_Rogues_Gallery/

and

https://www.bufferbloat.net/projects/bloat/wiki/RRUL_Chart_Explanation/

The live bufferbloat.net site seems fine, but I get broken links everywhere when trying to run locally with hugo serve. I set up the following as an experiment:

[Introduction]({{< relref "bloat/wiki/Introduction.md" >}}) 
[Introduction](Introduction)
[Introduction](Introduction.md)
[Introduction](./Introduction.md)

The first and second links work as expected, but the third and fourth (with ".md" suffix) gives the plain/unstyled 404 page not found page. Any clues (beyond dropping all the .md suffixes?)

The Van Jacobson links to the video and slides on the Bloat Videos page are both dead. The video times out, and the slides were sadly shared via Google+, which was unceremoniously dumped in a ditch this April.

Given it is the primary source of the "fountain" analogy alluded to in the introduction page, I think its important to address this.

If someone has a backup of the video, could we look at getting it on a hosting site like YouTube or perhaps Peertube? Failing that, a written description of the fountain model would be a nice idea. If someone can explain it to me (I was trying to learn about it when I discovered this linkrot) I'd be happy to draft something.

I was just referred to https://www.bufferbloat.net/projects/cerowrt/wiki/Setting_up_SQM_for_CeroWrt_310/ as an example for a confusing recommendation how to set up cake, and when I saw the Cerowrt 3.10 in the title I realized that maybe we should put a disclaimer into that page that the information contained does not reflect the current state of the art anymore.

lists.bufferbloat.net isn't answering on ports 80 or 443.

bufferbloar.net usually points the router as the source of buffer-bloat, but that's not always the case. It can be any network device.

For example my computer suffered from it on the Ethernet cable, but when I tried with a second computer the phenomenon didn't appear. So the Ethernet adapter was the cause itself, not the router.

I think it would be good clarifying that it could be due to any device. And mention that usually the router is to blame, just not always.

This is the first time I've looked at the git tree in forever.

languageCode: "en-us" # why not utf-8?

By what magic operation is a table of contents generated?

I see I've been using markdown wrong - I would use # title of section ## subtitle, where here its

WHATEVER

The rss.xml stuff kind of needs love, I think.

I was about to try and finish a few things on blog.cerowrt.org - notably "found with flent" and "flaws in flent" and to start a tutorial for lede users, and perhaps those might belong here rather than there. I feel compelled to write in a a highly opinionated bloggy style - at least at first, over there - that perhaps isn't the right thing here, though.

Otherwise the usability, look and feel are all great, the cdn behavior is marvellous (god, something like 1.5M wordpress sites compromised last week), and I should really, really, really go and make an edit pass here like I've been promising to do for a year now.

Greetings!

I m new to bufferbloat control... i m using a simple HFSC script to control my bufferbloat... i m on 30/30 fiber limited to 22 on both with veth system... problem that i m facing is that my upload goes very high on 5ghz wireless(mt7610E)...

What happens on my 5ghz is that it goes beyond 22 mbps and while it does i see bufferbloat... it eventually comes to settle at 22 mbps... but this thing doesnt happen to me on my 2.4 ghz as well as on my wired...

https://www.webpagetest.org/result/170210_61_1A5T/

I'm not sure if bufferbloat.net is being properly pipelined.... I was under the impression that

https://www.bufferbloat.net/ could be elided throughout the site and used with more relative urls and no domain name.

see also: https://news.slashdot.org/story/17/02/10/1255245/most-of-the-web-really-sucks-if-you-have-a-slow-connection

On https://www.bufferbloat.net/projects/make-wifi-fast/wiki/Hardware_Plan/, under Hardware Options, all the links except Supermicro lead to 404 pages or redirect to the company's home page.

and complains a lot when generating the site.

Images don't load on https://www.bufferbloat.net/projects/codel/wiki/RRUL_Rogues_Gallery/

I updated the bufferbloat.net repo to scratch a couple itches...

• Fix the "What Can I Do About Bufferbloat?" page to be more explicit and less duplicative of other pages.
• Remove/Update references to LEDE in favor of "OpenWrt"
• Make a pass through the CeroWrt index.md page to note that it's an archive of older work
• Fix a couple broken links

How does the official bufferbloat.net site get updated with the new content? Thanks.

commit fceb6435e852 ("netlink: pass extended ACK struct to parsing functions") in linux next-20170418+

causes this error.

I just arrived at the bufferbloat website, I've read a few introductory articles, now I'm reading 'what can I do about bufferbloat', and in Step 1 the author directs me to the 'getting sqm running right' to see an example of SQM tuning.

This article is very informative but I feel like I'm going off the deep end here. For example the first plot, is that before or after the tuning? From the text it's not clear to me.

And then the article is about SQM tuning, but it looks to me like they're tuning the router's bandwith caps to avoid filling a buffer at the ISP side and not the queue management. Is that right? It's not in the text as far as I can see, maybe I missed something.

hi,

Windows 10 Creators now has a CUBIC implementation and many changes maybe some are relevant to windows tips other i dont know:
netsh int tcp set supplemental template=internet congestionprovider=cubic netsh int tcp set supplemental template=internet congestionprovider=compound

Other changes here: https://blogs.technet.microsoft.com/networking/2017/07/13/core-network-stack-features-in-the-creators-update-for-windows-10/

I'd like to be doing "before - after" plots side by side. It doesn't appear hugo by default lets you do anything fancy with graphics like this. I figure writing a new macro {{%figure2 src1= src2= %}} to do it right would be good - but in this modern era do you use tables? div tags? Apply some extra css. what?

also would like to "flow" text along the side of a graph (image align left or right)

(yes, I'm willing to learn enough go/hugo/templating to do this, but being too ancient to even know how to apply styles in modern is a PITA)

@tohojo @richb-hanover @heistp @chromi (et al - normally I'd use the bloat list or a google doc for this, I don't know all the aliases folk use on github)

I have no idea what will do for a next project, if we do one. 7 years of debloating enough for y'all?

Whats left worth fixing?

Personally I have an hankering to go look hard at a "sane ecn", which in part involves looking at tcps and in part improving codel-ly responses to it. Other stuff might be fixing bloat in things like openvpn or wireguard. Am still in search of another wifi chipset worth fixing. Nobody seems very interested in speeding up cake, ebpf, or doing bobbie. As I have an early deployment of "fast" cake, running at 100mbit on wndr3800s, I personally need a faster shaper, so I took an old fork of that ( https://github.com/dtaht/sch_tart ) and will try to get it working at some point.

I could see trying to add better network stats awareness to various core applications, like netperf,
apache, ssh, and freeswitch, to get them to notice things like retransmits. "smartperf". Babel could also use some love here in how it blasts traffic. I'd like to be fiddling with audio at 2.7ms intervals.

I put a bunch of possibly worthwhile things for flent over here: tohojo/flent#148

On the tcp front, I've been forced of late to think about ecn issues hard ( systemd/systemd#9880 ) . I've done my best to ignore the whole issue for more than a few years now. Part of my problem is that few share my deep skepticism about it or my level of testing and the other is what it does to my BP when discussed. redteaming it would be reassuring.

Stuff I'd like to try includes:

fq

• can we make it faster?
• researching van's timer queue idea and virtualclock based fq and the etx qdisc

aqm

• Improving the bulk dropper (see: https://lists.bufferbloat.net/pipermail/codel/2018-August/002367.html )
• increase codel's marking frequency faster
• try to figure out of the l4s folk are entirely or only partially insane.

tcps (see also: systemd/systemd#9725 )

• if we are stuck at cwnd 2 and still getting loss, reduce the initial window, constrain growth for a while,
  pretend we are re-ordering (?).
• make cwnd 1 possible again
• decrease the mss dynamically as torrent used to do
• evolve tsq + pacing so that pacing is independent is independent of the cwnd
• an ect(0) - should at the very least constrain cwnd growth as it happens in the case of loss or reordering, I don't think it does. I'm also unsure if we have rfc compliant behavior on retransmits
• both ecn and loss on a RTT - cut back by more than half
• Fiddle with bbr
• try to incorporate some bbr ideas into cubic
• fiddle with quic

cake

• anything that could make it faster
• ISP version
• cake-mq

fq_codel

• Make sure the bsd version is right

BQL

• I'd like to see this scale better with multiple hardware queues
• AIMD version

website

• could use some love in general
• lartc finally died we can replace it
• hugo is at v44, and we are stuck at 18
• doing a better educational video
• pursuing publicity, picking another fight with the fcc
• more talks

videoconferencing

• try breaking the audio and video into different tuples, use the audio as a clock for the video
• try improving encoders to have better changes in network conditions

???
...

Or we could all declare victory and stay home merely enjoying our vastly better networks. I'm thinking of taking up starcraft 2 again. Anybody want to take on me as a Protoss? :)

The command "set-nettcpsetting -CongestionProvider CTCP" that is suggested on the page returns a Read-only error in and elevated PowerShell. It doesn't work in the current version of Win10.

When I use hugo serve on the master branch, I get hundreds of these errors:

ERROR: 2016/06/14 error processing shortcode shortcodes/issue_journal.html
 ERR: template: shortcodes/issue_journal.html:5:14: executing "shortcodes/issue_journal.html" at <markdownify>: wrong type for value; expected string; got template.HTML

I don't know enough about hugo to know how to fix them. Thoughts? Thanks!

Is there any development ongoing for handling bufferbloat with variable bandwidths?

For a lot of volks LTE/5G is the only option for their internet access and therefore have to deal with variable bandwidth.
In my case the download bandwidth varies from 30 to 130 Mbit/s (Download) 30 to 45 Mbit/s (Upload) which makes every algorithm with fixed bandwith unuseable, if you do not want to go with the minimum of 30 Mbit/s.

best regards

On this web page - https://www.bufferbloat.net/projects/bloat/wiki/What_can_I_do_about_Bufferbloat/

This link - https://lede-project.org/docs/howto/sqm

Should be updated to - https://lede-project.org/docs/user-guide/sqm

Currently, the list of wiki pages for each project can be accessed at (for the bloat project): https://www.bufferbloat.net/projects/bloat

However, this is not visible anywhere, and it may not be the best place to have it either. Something should be done about that.