FlexQoS FlexQoS 1.3.2 - Flexible QoS Enhancement Script for Adaptive QoS

[dave14305]

Weird issue on flexqos. Noticed it after I switched to gt-ax6000.
Most of streaming traffic (Netflix out of the issue) is shown on net control packets on the graphic. However, in the detailed list there is no device using net control packets.

Note - screenshot taken while streaming a movie on HBO MAX
Screenshot attached and debug code here:

Thoughts:

• You should switch back to stable channel
• You have 3 iptables rules putting traffic in Net Control. One seems to be related to NextDNS traffic, and one seems to cover the whole LAN 192.168.1.0/24 (why?).
• What is the AppDB mark for HBO? I don’t have an Asus router to check myself anymore.
• What LAN IP did the streaming?

 

[maghuro]

Thoughts:

• You should switch back to stable channel
• You have 3 iptables rules putting traffic in Net Control. One seems to be related to NextDNS traffic, and one seems to cover the whole LAN 192.168.1.0/24 (why?).
• What is the AppDB mark for HBO? I don’t have an Asus router to check myself anymore.
• What LAN IP did the streaming?

Hello Dave! Nice to see you

- Done. Same symptom
- My Lan is on the 10.0.0.0/24 subnet. That 192.168.1.0/24 is another Lan I sometimes need to access when I'm behind a double nat. Not the case right now, but I removed that rule just in case, but the symptom persists.
- 04009C - however, it happens with most of the streaming traffic. It shows on streaming on the list, but the graph shows is as net control.
- 10.0.0.65

My final question - why have most of you guys have switched from Asus routers? Is there something good I'm missing?

Thanks!

 

[dave14305]

- 04009C - however, it happens with most of the streaming traffic. It shows on streaming on the list, but the graph shows is as net control.

The graph is correct, but it’s unclear why the traffic ends up there. Capture the output of tc filter show dev br0 and tc qdisc.

Run tc monitor at the command line and then start some streaming. Post any output that appears in the window (there may be none).

Last, show the contents of /tmp/bwdpi/qosd.conf.

why have most of you guys have switched from Asus routers? Is there something good I'm missing?

Eventually the limitations of Asus routers become too great with increasing ISP speeds. You’re not missing anything, but when you’re ready to move on, it will hit you suddenly.

 

[maghuro]

The graph is correct, but it’s unclear why the traffic ends up there. Capture the output of tc filter show dev br0 and tc qdisc.

Run tc monitor at the command line and then start some streaming. Post any output that appears in the window (there may be none).

Last, show the contents of /tmp/bwdpi/qosd.conf.

Eventually the limitations of Asus routers become too great with increasing ISP speeds. You’re not missing anything, but when you’re ready to move on, it will hit you suddenly.

Here it goes:

adminGT-AX6000/tmp/home/root tc filter show dev br0filter parent 1: protoc - Pastebin.com

Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.

pastebin.com

Regarding ISP Speeds - in that case I have a router for the next 20 years, as in Portugal we don't have more than 1000mbps and no signs of future improvements...

 

[L&LD]

I'd bet you'd be wrong on that, over the next 20 years.

 

[dave14305]

Here it goes:

adminGT-AX6000/tmp/home/root tc filter show dev br0filter parent 1: protoc - Pastebin.com

Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.

pastebin.com

It looks like FlexQoS isn’t starting properly. Any messages in syslog?

 

[maghuro]

It looks like FlexQoS isn’t starting properly. Any messages in syslog?

Nothing...

It is properly starting, at least it's doing it's job, except for some streaming...

 

[dave14305]

Nothing...

It is properly starting, at least it's doing it's job, except for some streaming...

The second pastebin link suggests it is not starting properly. It would have been broken on the development snapshot on a GX6000, but should work on the stable branch.

 

[maghuro]

The second pastebin link suggests it is not starting properly. It would have been broken on the development snapshot on a GX6000, but should work on the stable branch.

Just forced update on the stable branch and the issue seems solved

What are the differences between branches?

 

[dave14305]

Just forced update on the stable branch and the issue seems solved

What are the differences between branches?

develop had test code to apply limits on burst and cburst, but it didn’t work out too well. And it was missing a fix from master related to the 4.19 kernel models. I just backported this fix to develop, but stable is the place to be.

Still waiting for a volunteer to take over the project.

 

[SomeWhereOverTheRainBow]

Still waiting for a volunteer to take over the project.

They would have some pretty big shoes to fill. I don't know of too many on the @dave14305 caliber.

 

[thiggins]

This thread closed by request. Post in new thread

FlexQoS - FlexQoS issues with 388.4 HND5.04 models

New thread opened by request. Old thread has been closed. https://www.snbforums.com/threads/flexqos-1-3-2-flexible-qos-enhancement-script-for-adaptive-qos.75620

www.snbforums.com