Help me configure my SQM for videoconferencing and gaming

Except IPv6; the rationale fear of the tracking involved created the pressure which begat IPv6 privacy extensions...

Nope ATM my ISP uses dual-stack with public IPv4 and IPv6 prefix, nice actually. Unlike the incumbent they have no publically known upgrade path into the future (but they are owned by Spain's incumbent so probbly have something in the works as well). Telekom is in the process of switchin their mobile network to IPv6 and 464XLAT. I would not be amazed if they would switch to that for their fixed line business (even though they championed the IMHO sane ds-lite extension lw4over6), but they also sit on a big stack of public IP-v4 space so will have time to wait for more traffic to switch to IPv6 before relegating IPv4 from native to as-service-over-IPv6. I would guess Telefonica will behave similarly (except their pile of IPv4 addresses used in Germany is considerably smaller, so they might have to move earlier).

6rd is just that, except that the ipv6 prefix is determined by shoving the ipv4 onto the end of some initial prefix, so they change in lock step. ATT was doing that when I first signed on, but these days they're doing full ipv6 native prefix.

BTW I ran IPv6 only on my lan for a full year, using NAT64, DNS64, an http proxy, and Tayga. The things that didn't work were a small number of computer games. Minecraft at the time for example had only partial support (these days I think it works fine with ipv6 ... mostly).

I'm convinced if we passed a law saying any consumer ISP which offers 100% IPv6 only with a /56 by default and /48 for anyone who requests it, by Dec 31 2022 would get paid $200 per customer signed up and connected... the switch would happen without incident.

@EXREYFOX you can't add ip adress with my settings for the moment because the rules don't work

I has volontary séparate TCP and UDP

like my screen

hello, how can i check if my isp is washing my priority packages?

just i have a last question @dlakelan

@EXREYFOX suggere disabled nat dual ?

like use only upload for wan and lan

i'm delete the nat dual how

Yes if you are doing DSCP tags then just delete nat dual-xxx you can leave ingress and ack-filter

thanks for fast response

like that
Capture d’écran 2021-02-04 à 11.53.11

It's surprisingly hard. You need to send packets and control the receiving end so you can check. But also there are multiple hops. Anyone in between could alter DSCP. The norm is that it doesn't survive end to end. In fact DSCP is not designed as end to end. It's designed as end to end within a single domain. Where a domain is something like a single ISP or a single business campus or a single university, or your wired LAN in your house.

No just do ingress and ack filter, delete dsthost and srchost.

1 Like
config queue
	option debug_logging '0'
	option verbosity '5'
	option interface 'eth1.100'
	option upload '85000'
	option qdisc 'cake'
	option enabled '1'
	option linklayer 'ethernet'
	option overhead '44'
	option download '0'
	option script 'piece_of_cake.qos'
	option linklayer_advanced '1'
	option tcMTU '2047'
	option tcTSIZE '128'
	option tcMPU '73'
	option qdisc_advanced '1'
	option ingress_ecn 'ECN'
	option egress_ecn 'NOECN'
	option qdisc_really_really_advanced '1'
	option eqdisc_opts 'nat dual-srchost'
	option linklayer_adaptation_mechanism 'default'
	option squash_dscp '0'
	option squash_ingress '0'

config queue
	option debug_logging '0'
	option verbosity '5'
	option interface 'eth0.1'
	option download '0'
	option upload '185000'
	option qdisc 'cake'
	option script 'piece_of_cake.qos'
	option linklayer 'ethernet'
	option overhead '44'
	option enabled '1'
	option linklayer_advanced '1'
	option tcMTU '2047'
	option tcTSIZE '128'
	option tcMPU '73'
	option qdisc_advanced '1'
	option ingress_ecn 'ECN'
	option egress_ecn 'NOECN'
	option qdisc_really_really_advanced '1'
	option eqdisc_opts 'nat dual-dsthost ingress'
	option linklayer_adaptation_mechanism 'default'
	option squash_dscp '0'
	option squash_ingress '0'


Im using dscp-tags with traffic-rules, in my case i have to delete the eqdisc_ops on my WAN and use only on eqdisc_ops "ingress".

The iqdisc_ops i dont use because i set ingress speed to "0"

1 Like

Thank you for all my friends, here is the connection at home in vdsl2 and at my brother in ADSL,

the gameplay is perfect :slight_smile:

http://www.dslreports.com/speedtest/67233446 vdsl2 in my home

http://www.dslreports.com/speedtest/67235144 adsl at my brother

config queue 'eth1'
        option interface 'eth1'
        option qdisc 'cake'
        option script 'piece_of_cake.qos'
        option ingress_ecn 'ECN'
        option enabled '1'
        option download '0'
        option upload '16000'
        option debug_logging '0'
        option verbosity '5'
        option qdisc_advanced '1'
        option qdisc_really_really_advanced '1'
        option linklayer 'ethernet'
        option overhead '44'
        option linklayer_advanced '1'
        option tcMTU '2047'
        option tcTSIZE '128'
        option tcMPU '0'
        option linklayer_adaptation_mechanism 'cake'
        option squash_dscp '0'
        option squash_ingress '0'
        option egress_ecn 'NOECN'
        option eqdisc_opts 'nat dual-srchost'

config queue
        option enabled '1'
        option download '0'
        option upload '56000'
        option debug_logging '0'
        option verbosity '5'
        option qdisc 'cake'
        option script 'piece_of_cake.qos'
        option qdisc_advanced '1'
        option ingress_ecn 'ECN'
        option egress_ecn 'NOECN'
        option qdisc_really_really_advanced '1'
        option linklayer 'ethernet'
        option overhead '44'
        option linklayer_advanced '1'
        option tcMTU '2047'
        option tcTSIZE '128'
        option tcMPU '0'
        option linklayer_adaptation_mechanism 'cake'
        option interface 'eth0'
        option squash_dscp '0'
        option squash_ingress '0'
        option iqdisc_opts 'nat dual-dsthost ingress'

config queue
        option enabled '1'
        option download '0'
        option upload '56000'
        option debug_logging '0'
        option verbosity '5'
        option qdisc 'cake'
        option script 'piece_of_cake.qos'
        option qdisc_advanced '1'
        option ingress_ecn 'ECN'
        option egress_ecn 'NOECN'
        option qdisc_really_really_advanced '1'
        option linklayer 'ethernet'
        option overhead '44'
        option linklayer_advanced '1'
        option tcMTU '2047'
        option tcTSIZE '128'
        option tcMPU '0'
        option linklayer_adaptation_mechanism 'cake'
        option interface 'wlan0'
        option squash_dscp '0'
        option squash_ingress '0'
        option iqdisc_opts 'nat dual-dsthost ingress'

I have à problem , all my output data is CS0 and all my input data is cs4.. so easy to kill me fast but so difficult to kill for me ...

When you capture on an interface anything recieved on that interface has not been processed by iptables yet. But things sent on the interface have been all the way through. So what you see is no tags on packets coming in and tags on packets going out. But if you capture on the "other side" of your router you will see the opposite

1 Like

So is normal ? I don’t understand why I am always so late, my bufferbloat is just perfect for play..

Yes it's normal there can't be tags on packets until after you tag them.

I think the game is just not well programmed.

1 Like

Are you using the same sqm and traffic-rules like @Dopam-IT_1987