Hacker News: dave_taht

New comment by dave_taht in "Ask HN: What are your predictions for 2023?"

dave_taht — Mon, 26 Dec 2022 00:41:30 +0000

hackernews gets ipv6 support.

Geoff Huston – On “Going Dark”

dave_taht — Sat, 26 Nov 2022 23:56:34 +0000

Article URL: https://blog.apnic.net/2022/11/04/opinion-going-dark/

Comments URL: https://news.ycombinator.com/item?id=33758346

Points: 1

# Comments: 0

New comment by dave_taht in "Ask HN: What weird technical scene are you fond/part of?"

dave_taht — Tue, 22 Nov 2022 14:46:58 +0000

I keep hoping everyone will have that Aha! moment one day, and we'll all have cake. https://libreqos.io

New comment by dave_taht in "Blip: A tool for seeing your internet latency"

dave_taht — Thu, 03 Nov 2022 06:05:38 +0000

so nice to see so many using cake.

New comment by dave_taht in "Blip: A tool for seeing your internet latency"

dave_taht — Thu, 03 Nov 2022 06:04:12 +0000

So happy to see so many sqm and cake users here.

New comment by dave_taht in "Blip: A tool for seeing your internet latency"

dave_taht — Thu, 03 Nov 2022 06:03:05 +0000

apple's location services have cost me MUCH grief in developing newer versions of fq_codel for wifi, until we figured it out.

New comment by dave_taht in "Blip: A tool for seeing your internet latency"

dave_taht — Thu, 03 Nov 2022 06:02:04 +0000

I'm always interested in packet captures of observed misbehavior.

New comment by dave_taht in "Blip: A tool for seeing your internet latency"

dave_taht — Thu, 03 Nov 2022 06:01:23 +0000

that is a dismal ratio. I would set cake to turn on the ack-filter in this case on the upload. Also try comparing cake besteffort with your fq_codel implementation. MOST likely, from your report, you have some higher priority (via diffserv) traffic than blip going, and cake is actually doing the "right thing" by prioritizing that over the test.

New comment by dave_taht in "Blip: A tool for seeing your internet latency"

dave_taht — Thu, 03 Nov 2022 05:59:01 +0000

So nice to see gfblip and cake together again at long last.

New comment by dave_taht in "Taildrop"

dave_taht — Thu, 13 Oct 2022 17:37:18 +0000

The openwrt folk have released a similar vpn idea, using the kernel wireguard version and a DHT: https://forum.openwrt.org/t/new-wireguard-based-openwrt-vpn-... works on std linux as well. Very raw.

Home internet connections are unfair

dave_taht — Sun, 09 Oct 2022 14:49:32 +0000

Article URL: https://www.youtube.com/watch?v=UICh3ScfNWI

Comments URL: https://news.ycombinator.com/item?id=33141691

Points: 7

# Comments: 1

New comment by dave_taht in "Ask HN: What did Linux not do right?"

dave_taht — Sun, 02 Oct 2022 16:19:10 +0000

NSF finally has a program that makes sense to support better open source orgs: https://www.nsf.gov/pubs/2022/nsf22062/nsf22062.jsp

Don't everyone apply at once!

New comment by dave_taht in "Show HN: An ultra-light-weight tool to quickly test your ping"

dave_taht — Sun, 02 Oct 2022 14:24:30 +0000

flent.org has some heavy duty, wonderful graphic tests.

ESR – Holding Up the Sky

dave_taht — Mon, 19 Sep 2022 15:48:19 +0000

Article URL: http://esr.ibiblio.org/?p=4196

Comments URL: https://news.ycombinator.com/item?id=32899953

Points: 2

# Comments: 0

New comment by dave_taht in "Framework Laptop 2.5Gbps Ethernet Expansion Card"

dave_taht — Mon, 12 Sep 2022 22:40:38 +0000

Mikrotik RouterOS7 supports fq_codel running native. Does that not work on 2.5Gbit -> 1Gbit?

Bufferbloat, the Internet, and how to fix it

dave_taht — Thu, 18 Aug 2022 22:26:37 +0000

Article URL: https://hackaday.com/2022/08/18/bufferbloat-the-internet-and-how-to-fix-it/

Comments URL: https://news.ycombinator.com/item?id=32515590

Points: 3

# Comments: 0

New comment by dave_taht in "Testing LEO and GEO satellite services in Australia"

dave_taht — Mon, 09 May 2022 18:46:42 +0000

Remarkable piece of work that didn't mention #bufferbloat once.

https://docs.google.com/document/d/1puRjUVxJ6cCv-rgQ_zn-jWZU...

New comment by dave_taht in "The state of fq_codel and cake in 2022"

dave_taht — Mon, 25 Apr 2022 10:46:52 +0000

At least some hardware provides the timestamp for you. Ideally if that's carried along with the packet, it measures the load on the whole system, so when you start dropping packets it's not just a function of network congestion but running out of cpu.

This would be a plus for a userspace router implementation, as presently in linux we timestamp on entry to the exit queue, not on all the overheads of routing lookup, and other processing from the read from the rx ring.

New comment by dave_taht in "The state of fq_codel and cake in 2022"

dave_taht — Mon, 25 Apr 2022 03:23:05 +0000

The principal use case for any AQM is when stepping down from one speed to another, in the case of wireless, wildly variable rates, or in the case of 10Gbit to 1Gbit on a port (rfc7567), or shaping to customer rates via middlebox, like libreQos does (using XDP, currently).

In the non-shaping case, say going two 10Gbit ports into one...

what I had expected (in the future I saw in 2015) was that the FQ/AQM would be offloaded to hardware, and that vpp/ddpk would also get a timestamp and 5 tuple hash from the receive path's HW.

The state of fq_codel and cake in 2022

dave_taht — Sun, 24 Apr 2022 22:00:08 +0000

Article URL: https://blog.cerowrt.org/post/state_of_fq_codel/

Comments URL: https://news.ycombinator.com/item?id=31148728

Points: 30

# Comments: 7