Re: [Bloat] A Transport Protocol's View of Starlink

[Kenneth Porter via Bloat]

The Register came out with this summary today:

https://www.theregister.com/2024/05/22/starlink_tcp_performance_evaluation/

Excerpt:

Using PING, he found "minimum latency changes regularly every 15 
seconds" and surmised "It appears that this change correlates to the 
Starlink user's terminal being assigned to a different satellite. That 
implies that the user equipment 'tracks' each satellite for a 
15-second interval, which corresponds to a tracking angle of 11 
degrees of arc."


During those handovers, Huston observed some packet loss – and a 
significant increase in latency. "The worst case in this data set is a 
shift from 30ms to 80ms," he wrote. Further: "Within each 15-second 
satellite tracking interval, the latency variation is relatively high. 
The average variation of jitter between successive RTT intervals is 
6.7ms. The latency spikes at handover impose an additional 30ms to 
50ms indicating the presence of deep buffers in the system to 
accommodate the transient issues associated with satellite handover."


Overall, Huston believes Starlink has "a very high jitter rate, a 
packet drop rate of around one percent to two percent that is 
unrelated to network congestion, and a latency profile that jumps 
regularly every 15 seconds."


[No need to cc me in replies, I'll read them on the mailiing list.]


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] A Transport Protocol's View of Starlink

[Kenneth Porter via Bloat]

This technical paper on Starlink by the chief scientist at APNIC crossed my 
feed this week. [I thought I'd share it to the Starlink list here but my 
application to join that list seems to have gotten stuck so I'll share it 
here for now.]





From the end of the paper:



While earlier TCP control protocols, such as Reno, have been observed to
perform poorly on Starlink connections, more recent TCP counterparts,
such as CUBIC, perform more efficiently. The major TCP feature that makes
these protocols viable in Starlink contexts is the use of Selective
Acknowledgement [11], that allows the TCP control algorithm to
distinguish between isolated packet loss and loss-inducing levels of
network congestion.

TCP control protocols that attempt to detect the onset of network queue
formation can do so using end-to-end techniques by detecting changes in
end-to-end latency during intermittent periods of burst, such as BBR.
These protocols need to operate with a careful implementation of their
sensitivity to latency, as the highly unstable short-term latency seen on
Starlink connections, coupled with the 15-second coarse level latency
shifts have the potential to confuse the queue onset detection algorithm.

It would be interesting to observe the behaviour of an ECN-aware TCP
protocol behaviour if ECN were to enabled on Starlink routing devices.
ECN has the potential to provide a clear signal to the endpoints about
the onset of network-level queue formation, as distinct from latency
variation.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] mDNS

[Kenneth Porter via Bloat]

On 2/27/2024 6:31 PM, Matt Taggart via Bloat wrote:
Also... starting back in the old CeroWRT days I switched to using the 
172.16 rfc1918 ranges when I realized that nobody else uses them, and 
that has been a good way to avoid collisions (but wouldn't work as an 
openwrt default). 


A customer of my company uses 172.16 subnets at their various locations, 
so it was good that I used a /16 from 10/8 for ours when we need to VPN 
in to work on a customer machine.


At home I use 172.16 subnets for the LAN side of my ISP CPEs with my own 
router in front of them. So my home LAN can be in a 10/8 subnet in spite 
of ATT's restriction.


I reserve 192.168/16 for when I need to plug in some new piece of 
hardware that defaults to that address block.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] mDNS

[Kenneth Porter via Bloat]

The ISP shouldn't be using an RFC 1918 address for the WAN port, anyway. 
For the management net, one would hope they'd use an IPv6 private link. 
Which should leave all of RFC 1918 available for the customer.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] mDNS (was Disappointment on "Best Newcomer Router" front)

[Kenneth Porter via Bloat]

On 2/26/2024 6:28 AM, Rich Brown via Bloat wrote:
- Avoid the WAN port's DHCP assigned subnet (what if the ISP uses 
192.168.1.0/24?)


I recently got ATT fiber and its modem won't let me assign from 
10.0.0.0/8! So I put a Raspberry Pi 4 in front of it.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] RED 13 years later

[Kenneth Porter via Bloat]

I didn't know what RED was and had to look it up. For others new to the 
lingo:


https://en.wikipedia.org/wiki/Random_early_detection


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] cloud gaming on a comeback?

[Kenneth Porter via Bloat]

I miss the days when teams could operate their own game servers on 
colo'd equipment. I ran several Tribes 2 servers for my team 20 years 
ago on a Linux server I shipped to a game hosting company in Kansas 
City. We also ran some of the early Battlefield servers. There were a 
lot of game servers like that. But the game publishers switched to 
operating their own servers to prevent teams from running servers that 
unlocked all the gated content, such as bonus weapons that a player had 
to earn with months of play. That was when I stopped playing those games.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] The Register; How TCP's congestion control saved the internet

[Kenneth Porter via Bloat]

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Does SQM always require you to tell it your ISP up/down speed?

[Kenneth Porter via Bloat]

Didn't you just ask this question?



There were several replies including links to auto-tuning scripts.

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] SQM tuning question

[Kenneth Porter via Bloat]

--On Saturday, June 03, 2023 7:44 PM +0300 Jonathan Morton via Bloat 
 wrote:



When your available bandwidth varies over time, that can be inconvenient.
There are methods, however, of observing how available capacity tends to
change over time (typically on diurnal and weekly patterns, if the
variations are due to congestion in the ISP backhaul or peering) and
scheduling adjustments on that basis.  If you have more information on
your situation, we might be able to give more detailed advice.


Are there any good solutions for regularly recomputing the bandwidth on a 
consumer link? I'm running the Linux SQM scripts to launch cake on a recent 
Debian system and CoDel on a CentOS 7 gateway. I set the bandwidth manually 
based on periodic manual speed tests. I'd love to automate that.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Open Source Needs You to Act - Help Fight Proposed Patent Rules

[Kenneth Porter via Bloat]

On 5/31/2023 10:57 AM, Rich Brown via Bloat wrote:
The letter you forwarded has really long URLs that made it hard to see 
the essentials of the argument.


Also, beware that when you share these "action needed!" emails, they 
include your personal unsubscribe link at the end, and possibly a 
subscriber ID in all the internal links, so readers who click those 
could unsubscribe you or register your interest in something you don't 
care about. This isn't unique to this particular email. I see them in 
all kinds of political action mailing lists, including some big name 
ones. They all use the same mailing list vendors. A proper solution 
would be to provide a link to a website for sharing the critical 
information without giving away your own subscriber information.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] speedtest-cli on multihomed gateway

[Kenneth Porter via Bloat]

On 2/3/2023 4:54 AM, Michael Richardson via Bloat wrote:

A new network namespace would certainly work, but it may be unmanageable
overkill.

What you probably need are policy-based routes, which you can establish
statically and then --source ought to work.


That's exactly what turned out to be the issue. I'd already planned to 
implement it to make some clients use the alternate ISP, but didn't 
realize it applies to packets with source address set to that interface. 
I'd read through a few tutorials and tested that and it worked!



I put these into "up" statements into my /etc/network/interfaes, but you say
you are running RHEL... I'm sure that there is a netplan way.
This also means that if you have a monitoring system elsewhere (smokeping or
something), and you ping each interface, then it will reply on that
interface.


That turned out to be harder to figure out. RHEL 8 is a transition from 
traditional network scripts in /etc/sysconfig/network-scripts to 
NetworkManager connections. So the RH NM has to be able to understand 
the old files. I'm used to editing files with a text editor to make 
changes, but the RHEL docs don't expose the NM files and require one to 
use a manager program (nmcli) to make changes. I have no idea where my 
rules went when I changed them with nmcli. I do see the new 
per-interface route tables in the old scripts location. RHEL 9 moves 
over completely to NM, so I'll need to figure out how to migrate the 
config when I upgrade.


https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/8/html/configuring_and_managing_networking/configuring-policy-based-routing-to-define-alternative-routes_configuring-and-managing-networking

BTW, I noticed that the numeric routing table IDs are 32-bit, but the 
reserved IDs are from 0-255. Some online tutorials specify that custom 
tables need to be in the 8-bit range. I suspect that the earliest 
implementation used a single byte for the table ID and it widened in 
later kernels. The example number used in the above documentation is 
5000. So I'm using 2000 and 4000 for eno2 and eno4. (eno3 is the default 
link and eno1 is the LAN with an explicit global rule.)




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] speedtest-cli on multihomed gateway

[Kenneth Porter via Bloat]

I'm trying to get my head around how to run speedtest-cli on a Rocky 8
box (RHEL8 respin) with LAN and three WAN connections. I want to run
speedtest-cli on all the WAN links to test each, and run the tests
from a timer to watch the speeds throughout the day (eg. a symlink to
a test script in /etc/cron.hourly or a systemd timer unit).
speedtest-cli accepts a --source option but it wants to always route
through the WAN link set to the default route when I specify the
addresses of the other two. From googling around it looks like I want
to run it in a "network namespace" but I haven't figured out how to
make that work.

I found this stackexchange answer on the namespace basics but I think
I need a few more commands to actually make an app usable in the
namespace by ifup'ing the interface, routing, and DNS.

https://unix.stackexchange.com/questions/234583/routing-on-per-application-basis

Hardware setup:

eno1: LAN, 10.96.0.64, default route for LAN
eno2: WAN1, 172.24.96.xxx, to ATT fiber gateway at 172.24.96.1
eno3: WAN2, 172.24.69.xxx, to Xfinity gateway at 172.24.69.1
eno4: WAN3, 172.24.0.xxx, to Xfinity gateway at 172.24.0.1

What I'm trying:

# ip netns add comcast-1
# ip link set eno4 netns comcast-1
# ip netns exec comcast-1 speedtest-cli
Retrieving speedtest.net configuration...
Cannot retrieve speedtest configuration
ERROR: 

At this point I'm not sure what I need to do to make the network
namespace usable.

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] Precision Time Protocol (PTP)

[Kenneth Porter via Bloat]

A nanosecond-scale replacement for NTP? This looks useful for measuring 
latency. Lots of technical implementation details here, down to the 
hardware level.




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] ZDNet gets caught re-posting year-old "best speed test" article

[Kenneth Porter]

Check the comments on this:



It's a year-old article with the publication date bumped to make it seem 
new. But older comments are a year old.


There's minor mention of latency in the article and no mention of 
bufferbloat. So it's still convincing readers (and, by extension, their 
representatives in the legislature) that the solution to their problems is 
more bandwidth, not better handling of bloat.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] ZDNet ruoter review

[Kenneth Porter]

Another router review with no mention of bufferbloat. So I nagged them 
about it in the comments.




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] tp-link request for SQM

[Kenneth Porter]

--On Friday, December 03, 2021 12:12 PM + "Wheelock, Ian" 
 wrote:



280Mbps service with Comcast is likely the 300/10 package offering… In
that case the US is limited to 10Mbps

I understand there might be some issue if 280Mbps was being processed –
but in the US direction, we are not talking >>100Mbps – its about
10Mbps US, I would have thought running cake on 10Mbps US cake would not
have triggered a 50% loss in performance even on this platform. Now if
cake is applied to both inbound and outbound traffic then having to deal
with ~280Mbps might be tough. In the case of DOCSIS AQM, PIE runs in the
GW only on outbound traffic.


That's indeed the package. I was seeing 270-280 down and 12 up.

"Good" tests without cake:



It got slammed down to 49 here:


I disabled cake and it slowly recovered:



(I wish the tests had a timestamp so I could be sure I'm ordering them 
correctly.)



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] tp-link request for SQM

[Kenneth Porter]

--On Friday, December 03, 2021 11:40 AM +0100 Sebastian Moeller 
 wrote:



No idea whether an archer 20 will do (not even sure what model that is,
here in Germany I see either an C20 or an AX20 but no plain unadorned
20). If you should try OpenWrt on that thing, the OpenWrt forum is a good
place to ask for configuration advice for specific models (will obviously
not help if you stick to the manufacturer's firmware).


Sorry, that was the AX 20. I'd misremembered it. (I find router model 
numbers quite confusing.)




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] tp-link request for SQM

[Kenneth Porter]

--On Thursday, December 02, 2021 10:48 AM -0800 Dave Taht 
 wrote:



tp-link, is, so far as I know, the last major home router vendor NOT
shipping a SQM system. Perhaps this could be modded up with someones
with accounts?

https://community.tp-link.com/us/home/forum/topic/511156


I just signed up an account and will add my vote.

I just bought an Archer 20 to replace my old 2016 Zyxel running OpenWrt. 
I'd found it by looking at various reviews of "best OpenWrt router for 
2021". I just updated my Zyxel firmware from v18 to v20 firmware. I get 
about 280 Mbps from Xfinity. I turned on cake and it dropped by 50%! So I 
think the old router's CPU isn't up to it. I'll be swapping in the TP-Link 
soon so I can turn on cake without the big performance hit.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] From OpenWrt forum: Bufferbloat, it's not just for WAN connections anymore

[Kenneth Porter]

Now that 500Mbps+ fiber and DOCSIS connections are becoming more common,
the issues with bufferbloat and uncontrolled latency are no longer just
affecting people's internet connections. Now the slowest point in the
link between your device and a server on the internet can easily be one
of your LAN links.


Read more:



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] ipv6 with telephone number mapping

[Kenneth Porter]

On 11/25/2021 9:26 AM, Dave Taht wrote:

What's the real barrier to doing this nowadays? We use signal in this
way, sort of:

https://twitter.com/ID_AA_Carmack/status/1463740452260171776

"There should be an automatic mapping from cellular numbers to IPV6
addresses so you can run a directly accessible server on any cell
phone."


Without root, phones make really frustrating servers.

Do/can phones have provider-independent addresses? Do we need root to 
assign one?



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] fq_codel wikipedia page in progress

[Kenneth Porter]

On 11/15/2021 2:15 PM, Kenneth Porter wrote:


I'd also suggest changing the page title to match the capitalization 
and hyphen of the RFC.


https://datatracker.ietf.org/doc/html/rfc8290


Following my own advice to mimic other protocol pages, I suggest that 
the title be the full, unabbreviated name of the protocol:


Flow Queue Controlled Delay Packet Scheduler and Active Queue Management 
Algorithm


Or maybe some subset of that. Use the acronym FQ-CoDel (taken from the 
RFC) in the body.


Obviously steal material from the bufferbloat site. ;)

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


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] fq_codel wikipedia page in progress

[Kenneth Porter]

On 11/15/2021 8:40 AM, Dave Taht wrote:

OK, I finally bit the bullet and started editing wikipedia  for the
first time, seriously. That said, i have very little patience for it,
especially cite hunting. It's taken me, what? 8 years to attempt this.
There's an awful lot on wikipedia that needs improvement elsewhere
also.

https://en.wikipedia.org/wiki/FQ_CODEL


When editing material about an RFC, I first look at another article that 
covers some RFC, to see how to organize it and how the links should 
look. Some you might use as models:



https://en.wikipedia.org/wiki/Address_Resolution_Protocol

https://en.wikipedia.org/wiki/User_Datagram_Protocol

https://en.wikipedia.org/wiki/Internet_Message_Access_Protocol


I'd also suggest changing the page title to match the capitalization and 
hyphen of the RFC.


https://datatracker.ietf.org/doc/html/rfc8290


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Cerowrt-devel] Little's Law mea culpa, but not invalidating my main point

[Kenneth Porter]

On 9/19/2021 9:00 PM, Valdis Klētnieks wrote:

I have a nice Chrome extension called IPvFoo that actually tracks the IP
addresses contacted during the load of the displayed page. I'll let you make
a guess as to how many unique IP addresses were contacted during a load
ofhttps://www.cnn.com

...


...


...


145, at least half of which appeared to be analytics.  And that's only the
hosts that were contacted by my laptop for HTTP, and doesn't count DNS, or
load-balancing front ends, or all the back-end boxes.  As I commented over on
NANOG, we've gotten to a point similar to that of AT long distance, where 60%
of the effort of connecting a long distance phone call was the cost of
accounting and billing for the call.


Should we be trying to block those, the way we do with ads? (I heartily 
recommend uBlock Origin for that.) Would it break the pages or make them 
faster?


The primary reason I block ads is to block malware. The second reason is 
to block video ads that insist on playing in the background when the 
browser is hidden, eating CPU and playing annoying sound. I really don't 
object to viewing commercial material if it's not shoving itself in my 
face or trying to hurt my machine. But saving bandwidth and speeding 
things up are worthy goals.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] DSLReports Speed Test doesn't like Remote Desktop

[Kenneth Porter]

Mea culpa! PEBCAK.

I forgot that I had one PC manually configured to use a different router 
and modem. I've got a cheap service on the other one as a backup in case my 
high speed Comcast goes down. (Usually because someone hits a pole between 
here and the CMTS.) I switched to the good router and now I'm seeing good 
numbers from DSLReports:






This is through an XB3 modem and Zyxel running OpenWrt with cake.

With SQM disabled (and hence no speed cap):





Both grant an A for bufferbloat. It looks like my service improved since 
the last time I tuned cake, when I set the cap to 170 Mbps. The new modem, 
an XB7, should make the SQM in the router unnecessary. (I might still need 
the router to shield my LAN from the modem's DHCP server, if it can't be 
disabled. I want my clients to use my own DNS server and static DHCP 
configurations.)


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] DSLReports Speed Test doesn't like Remote Desktop

[Kenneth Porter]

--On Friday, August 27, 2021 2:18 PM +0300 Jonathan Morton 
 wrote:



A browser-based speed test like DSLreports depends heavily on the
responsiveness of the browser itself.  It would appear that RDP
interferes with that quite spectacularly, although I'm unsure exactly
why.  The only advice I can give is "don't do that, then".


I've gotten good numbers from Xfinity's own speed test, and the Waveform 
test shows what I expect. I'm pretty sure I've gotten reasonable numbers 
from DSLReports in the past year, so this week's DSLReports test was quite 
surprising in sucking so badly. I did check the "low-fi" option to disable 
browser animations (and yet still got animations) so maybe DSLReports broke 
that and is trying to do browser animations in the same thread as the speed 
test.


I almost never have a problem with app responsiveness over the RDP 
connection, including Youtube on my 4k screen at 60 Hz. The one thing I 
don't virtualize is 3D stuff, notably gaming. It's actually impressive just 
how much stuff "just works" over RDP. (The two boxes ARE connected by only 
a 1 Gbit switch.)


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] DSLReports Speed Test doesn't like Remote Desktop

[Kenneth Porter]

The DSLReports speed test gives me <5 Mbps download speed when used over a 
Remote Desktop connection. The same test gives me around 200 Mbps when run 
on my machine connected to my display. The Waveform test shows 200 Mbps 
from the remote machine. All are done with Chrome. Bunch of tabs open on 
both, similar sets of extensions.


I'm testing my Comcast XB3 modem + OpenWrt router before upgrading it to 
XB7.


I use two computers, both Win10-x64. One's a half-height with a bit better 
CPU and memory that I use for development and web/mail, while the other has 
a full-height tower chassis so it has my good video card for gaming. I have 
my big 43" display hooked to the latter and I remote to the short machine 
for "business" use.


https://www.waveform.com/tools/bufferbloat?test-id=62b54f0c-eb3e-40c8-ab99-4f2105f39525

This one looks very poor, 4 Mbps:

http://www.dslreports.com/speedtest/69341504

Much better, direct instead of through RDP:

http://www.dslreports.com/speedtest/69341657

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Sidebar re illustrating quality (was: New Version Notification for draft-cpaasch-ippm-responsiveness-00.txt)

[Kenneth Porter]

On 8/19/2021 6:58 PM, Dave Collier-Brown wrote:
Look at the barrel link, in that case: I'll send you a sketch off-list 


Ok, the sketch is of a spoked wheel with 3 spokes, for throughput, 
latency, and RPM, and the spoke for throughput is much longer. The 
circle represents the spoke of smallest radius, indicating the worst 
rating of the service. The ISP will try to sell based on the longest 
spoke to make itself look better than the actual user experience.


It's like taking the barrel illustration and "exploding" the barrel so 
its staves lie flat, radiating from the base. In that case, the shortest 
stave is the worst rating.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Of interest: Comcast AQM Paper

[Kenneth Porter]

--On Sunday, August 01, 2021 9:28 PM + "Livingood, Jason via Bloat" 
 wrote:



The "XB3" definitely lacks it - as it's DOCSIS 3.0-based. You may be
eligible for a replacement, depending on your speed tier. Take a look at
https://customer.xfinity.com/#/devices and see what that says. If that
does not have anything that let's you replace it then send me an email
1:1 and I'll assist. Optimally you'll want an XB7 but an XB6 would also
work fine based on your config. (and I wish it could do more of what you
need on other areas)


How would I know which one I have? I can't find a model number on it and 
the web page "Gateway > Hardware > System Hardware" only shows the base 
model number of TG1682G, no suffix. Hardware Revision is 11.0.


I'd also consider buying my own, which might address my DHCP issue if the 
downloaded firmware doesn't take that away from me.




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Of interest: Comcast AQM Paper

[Kenneth Porter]

On 7/31/2021 3:55 PM, Neal Cardwell via Bloat wrote:

FWIW, from the paper it sounds like not all Comcast cable modems
had/have PIE, which enabled the A/B experiment:

"10. Latency Measurement Results
As explained earlier, for two variants of XB6 cable modem gateway,
upstream DOCSIS-PIE AQM was enabled on the CGM4140COM (experiment)
variant but was not available on the TG3482G (control) variant during
the measurement period. The TG3482G variant used a buffer control
configuration that predated AQM in DOCSIS."


For Comcast:

Mine is a TG1682G, which I'm betting lacks PIE. So how do we get an upgrade?

It's running in bridge mode so I can run my own DHCP server and override 
the default Comcast DNS servers, allowing me to use local names in my 
network. I run my own OpenWRT router behind it to do that. I'd love to 
eliminate the extra hop so how about giving us more control over the 
integrated DHCP(6) server? Let me override the options, both defaults 
and in fixed MAC-based assignments.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] rpm (was: on incorporating as an educational institution(s)?)

[Kenneth Porter]

--On Saturday, July 10, 2021 10:37 AM -0700 Dave Taht  
wrote:



I just finished up an insanely difficult (but profitable) stint with
apple working on (among other things)
a new "rpm" metric for "responsiveness under normal working
conditions", for which we just established a new mailing list here -

https://lists.bufferbloat.net/listinfo/rpm

and hope to march that concept towards plentiful open source code and
benchmarks in the ietf, and here, in the coming months.


What is "rpm"? I only know of the Redhat Package Manager and revolutions 
per minute. I don't see it explained on the mailing list page or in the 
mailing list postings.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Educate colleges on tcp vs udp

[Kenneth Porter]

On 5/27/2021 12:42 AM, Hal Murray wrote:

TCP is for things like FTP and web browsing - transferring large chunks of
data.  UDP is for simple things like NTP or DNS.  TCP is generally implemented
in the kernel.  UDP retransmissions are generally implemented by user code.


UDP is also for unicast where late is just as bad as never. I've got an 
IoT sensor that unicasts readings to a PC that uses UDP in a 2-host network.


Compare to FireWire's isochronous connections that send unicast packets 
at a regular rate, intended for streaming media and commonly used for 
cameras.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] fcc input

[Kenneth Porter]

--On Friday, March 26, 2021 1:09 PM -0700 Dave Taht  
wrote:



I have often thought hard about trying to explain things better via
animations, and on more than one occasion tried to find an animator
who could turn this old 8m talk from "people as packets" into "unruly
animals as packets".


I keep encouraging fellow space fans to go play with Kerbal Space Program. 
Perhaps there's a video game or engine out there that would be useful for 
rendering our concepts.


I'm not a particularly visual person so nothing comes to mind. But I 
suggest that a first step would be a story board, just some scribbled 
sketches to get started. With that we might get some input from hobbyist 
animators out there.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] fcc input

[Kenneth Porter]

--On Friday, March 26, 2021 10:58 AM -0700 Aaron Wood  
wrote:



I'm still surprised at how hard it is to get people to understand that the
problem they're having (especially with real-time video like Zoom) isn't
bandwidth, but jitter and bloat...


Speed is easy to understand. I see a similar issue with space travel fans, 
who have trouble grasping how acceleration, not velocity, is the important 
factor in that domain. And when velocity is a factor, it's the vector, not 
the scalar, that counts. If only people got decent math educations


Jitter and bloat aren't intuitive and need a lot more thinking and patience 
to understand. Perhaps we need a good animated cartoon to explain it.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] Meeting software continues to ignore the bloat issue

[Kenneth Porter]

No mention of bloat here!



Perhaps we can nudge this author to learn more?

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Updated Bufferbloat Test

[Kenneth Porter]

--On Wednesday, February 24, 2021 9:35 PM -0800 Dave Taht 
 wrote:



and with the sqmscripts off?




(I'd already gone home so this is using Firefox via remote desktop.)



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Updated Bufferbloat Test

[Kenneth Porter]

On 2/24/2021 2:15 PM, Kenneth Porter wrote:


I'll try to remember to run this from my office tonight when nobody's 
around. We've got a 50 Mbps fiber connection with AT and I'm using 
fq_codel on a CentOS 7 system for the SQM. That should be interesting.


https://www.waveform.com/tools/bufferbloat?test-id=494dbe95-5302-4e1c-84cd-fbb4c8871ea2

This is after restarting the sqm-scripts. I got initial bad results and 
looked at the debug log and found that the iptables commands were 
failing to get a lock during system boot to set up the mangle table. I 
think it's competing with firewalld and fail2ban and losing. I don't see 
any lock errors from the restart and now the test shows good bloat results.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Updated Bufferbloat Test

[Kenneth Porter]

My results:



My LAN is 100 Mbps. I'm using an OpenWRT-based router with Cake. My ISP is 
Xfinity.


I'll try to remember to run this from my office tonight when nobody's 
around. We've got a 50 Mbps fiber connection with AT and I'm using 
fq_codel on a CentOS 7 system for the SQM. That should be interesting.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] fq_codel on mikrotik, finally

[Kenneth Porter]

On 2/18/2021 8:44 PM, Dave Taht wrote:

That finishes covering the two biggest players in the wisp market, finally.

https://www.youtube.com/watch?v=wNT3CqmVFi4


He says he doesn't recommend this for production because it's a beta 
firmware load. He also avoids cake because he's heard it causes boot loops.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] Why you need at least 3Mbps upload to get good game performance with ~1500byte packets: Doing the math

[Kenneth Porter]

Upstream article:



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] Xfinity Flex streaming box starves on cake?

[Kenneth Porter]

I suspect that my Xfinity Flex box has too small an internal buffer and is 
starving when fed by my cake-enabled OpenWrt router.


For the last few weeks, I've been streaming Amazon Prime Video shows and 
movies and the box tends to go to black screen after 5-10 minutes of 
viewing, followed by a jump to the Flex main menu (which selects which 
streaming service to view). I thought perhaps the box was overheating or 
that there was a bug in the Amazon app that caused it to crash to the home 
screen. This week the same thing happened when I watched a show on HBO Max.


On a hunch, I disabled SQM on the router and was able to watch an hour of 
TV with no interruption, using a different service on successive nights. My 
next test will be to re-enable SQM and try viewing to see if I get a black 
screen and crash to home screen again.


Sometimes instead of a black screen I get a "problem with connection" 
message. So my theory is that the Flex box has too little buffer to handle 
hiccups in the feed, and is depending on a deep buffer in the Xfinity 
modem. Except that I have the modem set to bridge mode so I can disable its 
DHCP server and let my LAN server handle DHCP and DNS and other services.


My SQM config follows:

config queue 'eth1'
   option itarget 'auto'
   option etarget 'auto'
   option linklayer 'none'
   option interface 'eth1.2'
   option debug_logging '0'
   option verbosity '5'
   option qdisc 'cake'
   option qdisc_advanced '1'
   option squash_dscp '1'
   option squash_ingress '1'
   option qdisc_really_really_advanced '1'
   option script 'layer_cake.qos'
   option iqdisc_opts 'docsis wash besteffort nat ingress'
   option eqdisc_opts 'docsis ack-filter nat'
   option ingress_ecn 'NOECN'
   option egress_ecn 'NOECN'
   option download '17'
   option upload '9000'
   option enabled '1'


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] What is the ifb4 interface?

[Kenneth Porter]

I'd never understood what the ifb4 interface was for. Here's a nice 
diagram from an sqm-scripts GitHub issue.


Ken


 Forwarded Message 
Subject:Re: [tohojo/sqm-scripts] diagram needed (#125)
Date:   Tue, 17 Nov 2020 19:36:47 -0800
From:   taggart 
Reply-To: 	tohojo/sqm-scripts 


To: tohojo/sqm-scripts 
CC: Subscribed 



Well it's not great, maybe something like this
example 
 
Here's the dia file (gzip'd cause github)
example.dia.gz 



I was trying to wrap my head around why ifb4 existed and how it worked. 
Maybe it could somehow explain that qdisc's can only work on packets 
leaving a device and that's the reason for the ifb4.


—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub 
, 
or unsubscribe 
.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] Starlink packet routing details

[Kenneth Porter]

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] netperf server news

[Kenneth Porter]

--On Wednesday, October 07, 2020 3:23 PM -0400 Rich Brown 
 wrote:



I'm also aware of ipset, which I take to be an optimized alternative to
searching a long set of iptables rules (true?) I don't believe that my
OpenVZ VPS has kernel support for this, so as long as the
long-list-of-rules seems to work well, I'm going to leave it alone.


A quick google of "OpenVZ ipset" turned up a thread from 3 years ago 
suggesting it's in their kernel:




Note that ipset operates in addition to iptables. Other kernel subsystems 
can also use them. iptables has a module to query an ipset.


500 rules is a lot to search linearly. I'd think a hash table would give 
much superior performance. Note that every "good" packet has to check ALL 
the blocking rules to be approved.


I use ipsets to block probes to my mail servers from outside the country 
and from cloud services. I've managed to find a few sources of lists for 
those. I also use ipset with fail2ban.


The only complicated part is how to handle reboots or other service 
restarts. I use firewalld which does its own ipset management so I put the 
permanent lists there. (I have scripts to convert the cloud lists to a 
firewalld's XML format for its ipset storage.) fail2ban keeps its own block 
database in a sqlite file and tears down and recreates its ipsets on 
restart.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] netperf server news

[Kenneth Porter]

--On Tuesday, October 06, 2020 7:52 AM -0400 Rich Brown 
 wrote:



3) I would be pleased to get comments on the set of scripts. I'm a newbie
at iptables, so it wouldn't hurt to have someone else check the rules I
devised. See the README at https://github.com/richb-hanover/netperfclean


A couple of alternatives to custom scripts are fail2ban and the 
rate-limiting modules available for iptables such as hashlimit and recent. 
I haven't used fail2ban for rate-limiting so I'm not sure if it's the right 
tool for that but it monitors log files to add iptables rules for 
short-term banning. It's not hard to add your own log monitoring rule. I 
haven't used the iptables modules but they look like a natural solution for 
this.






Instead of using a unique iptables rule for each blocklist member, I 
suggest using an ipset. (I use firewalld as a front-end to iptables so I 
let it manage my ipsets, but you can also install ipset's service for use 
with raw iptables to save and restore the sets across boots.) Your block 
rule could be as simple as this:


iptables -I INPUT 1 -p tcp --dport netperf -m set --match-set 
NetPerfAbusers src -m conntrack --ctstate NEW -j DROP



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] cake + ipv6

[Kenneth Porter]

--On Monday, August 17, 2020 10:52 PM -0400 Daniel Sterling 
 wrote:



As I'm sure you know ipv6 addresses are essentially random on the
internal LAN as compared to v4 -- a box can grab as many v6 addresses
as it wants, and I don't believe my linux router can really know which
box is using which address, can it?


Is this the usual IPv6 allocation by autoconfig or is the Xbox grabbing 
extra addresses deliberately to break flow isolation? It should only 
advertise one public address for its updates.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] How about a topical LWN article on demonstrating the real-world goodness of CAKE?

[Kenneth Porter]

--On Tuesday, August 11, 2020 11:11 AM -0400 Daniel Sterling 
 wrote:



https://github.com/tohojo/sqm-scripts/blob/master/src/piece_of_cake.qos
looks to be more or less what I'm doing


I'm using the "layer cake" script on my OpenWrt router but I have nothing 
to dynamically change the bandwidth caps based on ISP "weather". I suspect 
that would be a useful thing to capture in a cron job or systemd timer unit 
as I still see Youtube stutter from time to time.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] How about a topical LWN article on demonstrating the real-world goodness of CAKE?

[Kenneth Porter]

--On Tuesday, August 11, 2020 9:43 AM -0400 Daniel Sterling 
 wrote:



as promised here is the script I run after rebooting my openwrt box,
to set up cake

https://gist.github.com/eqhmcow/c378c46a41aa5716767a0da811087dd4


How does this differ from the sqm-scripts available in Fedora and OpenWrt?

https://github.com/tohojo/sqm-scripts


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] dslreports is no longer free

[Kenneth Porter]

--On Friday, May 01, 2020 10:44 AM -0700 Dave Taht  
wrote:



https://www.reddit.com/r/HomeNetworking/comments/gbd6g0/dsl_reports_speed
_test_no_longer_free/

They ran out of bandwidth.

Message to users here:

http://www.dslreports.com/speedtest


Is there an open source speedtest of comparable quality and usability? I 
could run one on my Linode for friends and family.




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

--On Saturday, April 25, 2020 12:19 PM -0700 Dave Taht 
 wrote:



in /etc/config/sqm, with fq_codel, you need to uncomment or enter the
following lines

option ingress_ecn 'ECN'
option egress_ecn 'ECN'


Does this only affect flows that terminate on the box, or also those for 
which this is a (NAT) router in the middle? The tests I'm running are from 
an internal Win10 box, not from the router.




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

On 4/25/2020 8:55 AM, Dave Taht wrote:

You would be the first person in the history of the bufferbloat
effort, to attempt enabling ecn more fully on windows, and
observing the results on real traffic and real traffic types.

be prepared to take a lot of packet captures.

/me passes the vodka


I'm more a gin and tonic guy. I'll keep some ready. :D

Time to update Wireshark.

I realized I have two machines I can mess with at home (my game machine 
and my development machine) and a family member with some tech savvy can 
do this with his devel machine that he uses for a lot of video meetings 
lately. All Win10Pro. This would be behind our OpenWrt/Comcast router 
with cake.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

On 4/25/2020 9:00 AM, Dave Taht wrote:

Oh, I misread your report. I thought this was cake, not fq_codel. Care
to try that?


I'd love to if I knew how to add cake to CentOS 7. I've installed kernel 
modules for unsupported Ethernet interfaces before, so perhaps cake is 
available that way from a 3rd party repo? Or I could adapt a 3rd party 
driver RPM's source to add cake, instead. I really don't want to do a 
full custom kernel, though. That's hard to maintain over time as there's 
a new kernel in the updates every month or two. Some hints on how to add 
a qdisc to a kernel would be welcome.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

On 4/25/2020 8:43 AM, Dave Taht wrote:

and the last flaw of this test series is that ken took the dslreports
"fiber" setting for the dslreports test as "The right thing". the
"fiber" test is structured to stress test an asymmetric 1gbit/100mbit
connection, not a shaped fiber connection running at 50mbit symmetric.
The number of uploads is 4, downloads, 32 it's totally ok to pick
a given fiber/cable/whatever test, but it does help to apply the same
characteristics to more of the tests you do, if you are trying to
compare technologies.


I'd be happy to test with a different set of speed test settings. Would 
Corporate/Edu be a better choice?


In going through this exercise, I can imagine someone saying "where'd 
all the bandwidth I paid for go?" as we trade bandwidth for reduced 
latency. How do you sell that to end users? ISPs have trained people to 
think only in terms of raw speed (obviously that's best for an ISP's 
bottom line as they can upsell you the higher-priced package) and most 
people (outside of gamers) don't really grasp latency with the 
intuitiveness of speed. Maybe we need some Youtube videos showing end 
user experiences of the benefit of reduced latency.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

On 4/25/2020 8:32 AM, Dave Taht wrote:

It would be a useful test for an intrepid windows administrator to
actually enable ecn fully and see what breaks in their
vpn, smb, and rdp implementations, and across their remote workforce,
and observe any difference in QoE. I have long
tried to get one drunk enough to deploy ecn across their windows
infrastructure without any success.


I can easily do it for myself, now that I know where that setting hides. 
Is there reason to make others do it or is it sufficient that I be the 
only guinea pig here?




since (I think?) the proposed vpn traffic is client -> server (?) not,
clients -> router -> server, each individual using this link
over the vpn will tend to get their own queue, and a big upload or
download through their openvpn will tend to "do them in",
because openvpn has lousy queue management internally... and (when
last I looked), windows itself had not a lot of backpressure when
dealing with that device.


I'm using openvpn to avoid exposing raw RDP servers to the Internet. I 
still don't trust MS enough to allow a Windows box directly connected to 
the Internet. So it's Win10(@home) -> openvpn(client on Win10 @home) 
->openvpn(Linux router @office)->Win10(@office). I think another user 
may use RDP's file transfer capabilities, but I personally use a 
separate ssh (scp) connection for that. I have no problem with command 
lines.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

On 4/25/2020 7:19 AM, Jonathan Morton wrote:

In Linux and OSX, to make the setting persist across reboots, edit 
/etc/sysctl.conf.


For the lurkers, CentOS (and presumably other Red Hat distros) now has 
an /etc/sysctl.d for vendor-specific settings. The prevents modifying 
package-supplied config files so that package upgrades are easier. Much 
of Linux seems to be moving towards this model of a config subdirectory 
to place overrides.


I created 51-bufferbloat.conf in that directory to add the ecn setting 
and to select the default_qdisc.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

On 4/25/2020 7:19 AM, Jonathan Morton wrote:

No, not the qdisc (where ECN is enabled by default), but on the client.


I enabled it on both the router (Linux) and the client where I'm 
conducting the speed test (Win10):


http://www.dslreports.com/speedtest/62818220


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

--On Saturday, April 25, 2020 6:00 PM +0300 Jonathan Morton 
 wrote:



Looks like a definite improvement.  The Quality grade of C may indicate
that you haven't enabled ECN on your client; without it, Codel has to
drop packets to do congestion signalling.


I see "ecn" in the qdisc commands. Here's the log from running the script 
today:




I'm using sqm-scripts 1.3.0 from . I 
just checked and see he has a 1.4 tag. I'll have to go look at the 
changelog and consider upgrading.




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

before:

http://www.dslreports.com/speedtest/62767361

after:

http://www.dslreports.com/speedtest/62803997

Using simple.qos with:

UPLINK=45000
DOWNLINK=42500

(The link is supposed to be 50 Mbps symmetric and speed test does show it 
bursting that high sometimes.)


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

Alas, CentOS 7 lacks cake. It does have fq_codel so I used the 
simple.qos script from sqm-scripts, with uplink 5 and downlink 45000:


http://www.dslreports.com/speedtest/62797600
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

--On Friday, April 24, 2020 10:32 AM -0700 Dave Taht  
wrote:



That's miserable. 480ms latency on fiber?? You can do so much better.
But why centos? sure the sqm-scripts work with that but you should be
able to shape 45Mbits with even a wndr3800. openwrt works great on x86
hw, also. :)


The same box is providing a bunch of other public-facing services, so I 
need some moderately heavy iron. (Still a cheap server, though.) If it were 
JUST a NAT router, I'd consider a cheap OpenWrt-capable router like the one 
I'm using at home.


Note that this test was without any shaping parameters. I think CentOS 
(like Fedora) defaults to fq_codel, though.



do you get dedicated ipv6 with that AT service?


Yep, they give us a /56, which seems to be the default for "sites" unless 
you ask for something bigger. So I'm assigning a /64 to the link between 
our box and their gateway, and another to our LAN. That leaves 254 more for 
whatever. I need to assign a  to the public side and test. Haven't 
gotten to that, yet.


We also get some VOIP lines that their gateway deals with. So no SIP yet 
within the LAN. We do use the "WiFi calling" feature on our mobiles, 
though. Cellular coverage at our location is terrible.



What will be the vpn type? ipsec, terminating on the router, works
well with fq_codel because the hash is propagated to the tunnel,
wireguard and openvpn currently do not.


I'm using OpenVPN with proto udp and dev tun. Our main use is to run Remote 
Desktop from home to our office and lab PCs. If I need to move files, I 
usually use scp. Outbound, we use Cisco's VPN to connect to customers to 
check binaries into their Subversion repo over HTTPS.


For customers and vendors, we have secure FTP drops. Mostly used for CAD 
drawings.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

Here's what the fiber connection looks like with no SQM applied:



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

--On Thursday, April 23, 2020 7:20 PM -0700 Dave Taht  
wrote:



I used layer_cake with this overriding the defaults on my cable modems
in /etc/config/sqm

option iqdisc_opts 'docsis wash besteffort nat ingress '
option eqdisc_opts 'docsis ack-filter nat'


It turned out I didn't actually have the SQM enabled before. I hadn't 
spotted the checkbox to turn it on so I'd just loaded the bandwidth 
parameter but not enabled it.


Now with your settings I get this result:



My next project will be to enable cake on my CentOS 7 box that just got a 
new 45 Mbps symmetric fiber connection from AT ("Business in a Box"). We 
upgraded from 1.5Mbps/128kbps ADSL. Any hints on what settings to use? 
Right now everyone's giddy with the massive speedup. But I'm sure that'll 
change as people start using more bandwidth. (About a dozen users.) The big 
win will be for us moving large solid model files and binary software 
distributions between partners using either SMB over VPN or secure FTP. I 
expect transferring a few GB files could saturate the link and drop some 
SIP calls.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

On 4/24/2020 7:56 AM, Toke Høiland-Jørgensen via Bloat wrote:

`man tc-cake`


CentOS 7 lacks that man page but I found it online. Thanks!

http://man7.org/linux/man-pages/man8/tc-cake.8.html


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

On 4/23/2020 6:20 PM, Dave Taht wrote:

I used layer_cake with this overriding the defaults on my cable modems
in /etc/config/sqm

 option iqdisc_opts 'docsis wash besteffort nat ingress '
 option eqdisc_opts 'docsis ack-filter nat'


I think I see where to do this in the GUI. (I can't remember if I 
installed a mini-emacs and I can never remember how to drive vi, so raw 
file editing might be harder than a GUI field.)


Where can I find documentation on what those options do?


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

--On Friday, April 24, 2020 4:47 AM +0300 Jonathan Morton 
 wrote:



It looks like there was a websockets error during the test, so try it
again and it might work.


This time I got an A score.



I'm using OpenWrt 18.06.4 on a Zyxel NBG6716 with the piece_of_cake.qos 
queue setup script.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] this explains speedtest stuff

[Kenneth Porter]

--On Thursday, April 23, 2020 1:38 PM -0700 Dave Taht  
wrote:



dslreports.com is only on the third page of the search results.


What does it mean that my bloat indicator is a grey dot?





___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] Real-life bufferbloat while shopping

[Kenneth Porter]

I tried to go shopping this afternoon and Costco is enforcing social 
distancing by lining up customers outside the store in a very long line. 
They're allowing people to enter in batches of 50. Which means the line 
doesn't move for a very long time. I finally gave up and left. So now I've 
had the experience of being a dropped packet.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Bufferbloat glossary

[Kenneth Porter]

--On Sunday, March 29, 2020 10:23 PM +0300 Jonathan Morton 
 wrote:



I think the main distinction between online gaming and teleconferencing
is the volume of data involved.  Games demand low latency, but also
usually aren't throwing megabytes of data across the network at a time,
just little bundles of game state updates telling the server what actions
the player is taking, and telling the player's computer what enemies and
other effects the player needs to be able to see.  Teleconferencing, by
contrast, tends to involve multiple audio and video streams going
everywhere.


But most gamers DO use voice chat systems to coordinate their play with 
teammates. This might be built into the game or it might be a second 
program such as Mumble, Ventrilo, or TeamSpeak. Two-way headsets were 
popular with gamers long before one saw them used for office conferencing. 
And gamers care much more about latency than some office flunky who hears 
something a second or two later than transmitted. So their codecs tend to 
be a lot more network-friendly, trading off quality for low latency.


(Given the high bandwidth needs of video, I wonder if anyone's working on 
avatar-based meeting systems wherein one creates an avatar from one's photo 
(like Bitmoji) and uses pre-downloaded content (like video games) to 
construct low-bandwidth video streams?)



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] Bufferbloat glossary

[Kenneth Porter]

I get a bit lost with all the jargon and acronyms in use on this list. Is 
there a glossary somewhere? (Ideally a wiki so we can all expand it.) I've 
picked up a lot over my years running Linux servers but this list is a good 
example of how the more you know, the more you realize how ignorant you 
are. ;)


For example, in today's message from David P. Reed I find "EDF" and "ACID". 
The rest of his post reminds me a lot of the kinds of issues one runs into 
with real-time multiplayer video games. I don't write those but I've always 
been fascinated with how they work and their limitations. (Think about how 
many high-end consumer routers target gamers who are looking for a magic 
bullet to combat the lag that just got their character 
killed.)"Videoconferencing" is really just a special case of a massively 
multiplayer online role-playing game (MMORPG).


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] fcc's coronovirus guidelines

[Kenneth Porter]

--On Friday, March 27, 2020 3:41 PM -0700 Dave Taht  
wrote:



"put everyone on a schedule"... sigh

https://www.fcc.gov/home-network-tips-coronavirus-pandemic


How do we educate officials? It's not clear who we'd even address a 
correction to. Is there a bufferbloat page we can point them to?




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] Red Hat kernel: Backport TCP follow-up for small buffers (BZ#1739130)

[Kenneth Porter]

I saw that intriguing "fix" in the latest Red Hat kernel errata but the 
cited bugzilla isn't available to the public.




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Getting bloat tests into open source speedtest

[Kenneth Porter]

I just got an email newsletter from Ubiquiti advertising their speedtest. 
Unlike most, it graphs the speed over time as it runs.




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Configuring sqm-scripts on OpenWRT

[Kenneth Porter]

--On Tuesday, September 03, 2019 6:22 PM -0700 Etienne Champetier 
 wrote:



No idea if it's up to date, but my google foo gives me
https://openwrt.org/docs/guide-user/network/traffic-shaping/sqm


Thanks. Somehow I wasn't seeing the luci-app-sqm package in the downloaded 
list of available packages. I used the Find function to locate it and now 
get a very nice set of GUI knobs to set it up. I set it to piece of cake, 
which I can't do on my CentOS 7 box as its kernel is to old to have the 
module.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Rigorous Coffee Shop Bloat Testing

[Kenneth Porter]

On 9/3/2019 5:40 PM, Stephen Hemminger wrote:

There was a recent Wall Street Journal article that faster Internet doesn't 
mean anything.
https://www.wsj.com/graphics/faster-internet-not-worth-it/

I just thought "faster Internet just exposes your existing Bufferbloat"


I hit a paywall trying to read that so I looked up the article title and 
found some interesting commentary:


https://tech.slashdot.org/story/19/08/20/1450204/the-truth-about-faster-internet-its-not-worth-it

https://stopthecap.com/2019/08/20/wall-street-journal-says-faster-internet-not-worth-it-but-they-ignore-bottlenecks-and-data-caps/

Most people are streamers and won't fill a fat pipe. The big winners of 
fast Internet are people who want to download a huge game and play it 
quickly. But those are rare. (I'm a gamer but I'm patient and can wait a 
day to play so I'm happy to save money on a cheaper package that can be 
used for something else.)


As you say, when people report slow Internet, it's probably bloat, not 
the speed of the package. But faster packages make money for the ISPs.



___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] Configuring sqm-scripts on OpenWRT

[Kenneth Porter]

I've been using Linux on a small business server for 20 years and started 
messing with shaping back when Wondershaper was a thing, maybe 10 years 
ago. I've flashed open source builds into my home router to get better 
firmware but haven't really tuned it before.


I just installed the latest stable OpenWRT on my aging ZyXEL (the stock 
firmware was crap) and installed the sqm-scripts opkg. On my CentOS server 
I edited /etc/sqm/.conf and use systemd to launch the shaper. How do 
I do that on OpenWRT?


I'm guessing there's no pretty Luci web admin thing for it. Pointers to how 
to write one would be welcome. I'm sure others would love having GUI knobs 
for it.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Rigorous Coffee Shop Bloat Testing

[Kenneth Porter]

--On Tuesday, September 03, 2019 11:29 AM -0700 Dave Taht 
 wrote:



In many cases they share the wifi with their credit card reader and
when I say that fixing bufferbloat helps,
their eyes light up. That was a specific problem that at least one had
- demonstrable - I saw it take forever
to clear a transaction (and the bloat was 2+ seconds long at the time
- NOT triggered by me) once... he had a synology router, and "applying
QoS" "just worked", and we did other things like reposition the
antenna, also. got me lunch
that did and he punched a whole bunch of holes in my "repeat business"
card


My favorite cafe owner would just look like a deer in the headlights over 
this. She doesn't do online at all and just provides the wifi for 
customers. I'm not sure who set it up for her. Her CC reader is dial-up.


I'm pretty new myself to hacking a consumer router so I don't know what's 
even possible. Likely her router doesn't even have any settings to deal 
with bloat. What can you do without risking breaking things and ending up 
with an angry owner?




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] Does 5g have the bloat problems of WiFi?

[Kenneth Porter]

I just saw this article claiming that 5g is allowing brick-and-mortar 
automation providers to overcome limitations in WiFi, and I'm wondering if 
the technology is going to suffer from all the same problems previously 
seen in WiFi.




(The "googly-eyed robots" in the title are man-sized robots that wander 
grocery stores to visually track inventory. Someone put googly eyes on them 
to keep them from scaring customers.)


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] RPM for sqm-scripts

[Kenneth Porter]

--On Wednesday, July 31, 2019 10:37 PM +0200 Sebastian Moeller 
 wrote:



Pretty cool! (Caveat: due to lack of a centos system I will not be able
to actually test it)


It should work on any recent Red Hat system, including RHEL, Fedora, 
CentOS, and Scientific. It's using the systemd files so it can't start 
automatically with the older initscript-based versions. RH doesn't have the 
ifup hook for packages so it doesn't include that feature to enable/disable 
the SQM on interface up/down.


Note that this is just a packaging wrapper. I didn't change any of the 
scripts. This just drops all the files into the correct place and updates 
the package database so they can be easily removed or upgraded as 
necessary.


To install, just use "yum install sqm-scripts-1.3.0-1kp.noarch.rpm".

Start with "systemctl start sqm@eth0" (substituting your interface name for 
eth0). You need to first create /etc/sqm/eth0.iface.conf based on the 
provided example.



So maybe run the following to get some debug output:

SQM_DEBUG=1 SQM_VERBOSITY_MAX=10 /etc/init.d/sqm start ; tc -s qdisc ; tc
-d qdisc ; SQM_DEBUG=1 SQM_VERBOSITY_MAX=10 /etc/init.d/sqm stop

I would hope that this should undo its damage at the end but will also
capture the wedged state in between. Then again this still might lead to
an unusable interface


Thanks, I'll give that a try, probably tomorrow morning before everyone 
gets in. I was able to quickly get in to repair the damage and stop the 
service ("systemctl stop sqm@em2") by remoting in through a server I have 
in parallel and shelling to the router's internal interface (which isn't 
being shaped). The slowness didn't happen instantly but after a minute or 
more, when I got a call from the office that "the internet is down!". The 
external interface wasn't completely dead, just extremely slow, enough to 
eventually kill my ssh session. It recovered instantly with the stop 
command.


Ken


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] RPM for sqm-scripts

[Kenneth Porter]

I decided to try the sqm-scripts and created an RPM to make it easy to 
install/upgrade/uninstall on my CentOS 7 system. Here's the result:




An initial attempt at using simple.qos on my system killed performance and 
the interface became unusuable, but I was probably too optimistic in 
setting the bandwidth numbers based on best response from speedtest-cli. I 
temporarily disabled it until I can test during off-hours.


___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] How can I tell if fq_codel is running?

[Kenneth Porter]

--On Tuesday, July 30, 2019 4:58 PM -0700 Dave Taht  
wrote:



see also
https://www.bufferbloat.net/projects/bloat/wiki/Wondershaper_Must_Die/

and https://github.com/tohojo/sqm-scripts


Coincidentally, I found sgm-scripts right after posting. What prereqs are 
there to using that? I'm poking through the code and it looks like it 
should work, although I don't know if the CentOS 7 kernel has cake.




___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

[Bloat] How can I tell if fq_codel is running?

[Kenneth Porter]

I'm running CentOS 7 on a Dell R230 as a small office router. It's on a 
slow ADSL link, 1.5 Mbps down and 150 kbps up. The kernel supports 
fq_codel, as does the tg3 driver used by the interfaces.


In the past I've run the Wondershaper script and now I'm hoping the new 
codel improves on that. I think it's installed but I don't see its name 
in a "tc show". Should I see it there? Is there additional setup I need 
to do?


The system seems responsive but I've been tasked with prioritizing the 
network for certain users. Before, I set up some buckets in the 
wondershaper for that. I'm now trying to figure out how that's supposed 
to work with codel, which is ostensibly "knob-free".


[root@saruman ~]# tc -s class show dev em2
class mq :1 root
 Sent 127961591021 bytes 412946653 pkt (dropped 0, overlimits 0 
requeues 2726)

 backlog 0b 0p requeues 2726
class mq :2 root
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
class mq :3 root
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
class mq :4 root
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
class mq :5 root
 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
 backlog 0b 0p requeues 0
[root@saruman ~]# sysctl net.core.default_qdisc
net.core.default_qdisc = fq_codel
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat