If I was good enough to make my own script it would be all about D prioritizing everything but gaming only given them what they need to work... UPnP already prioritises the gaming in my humble opinion also it prioritizes every game it's not selective if it's working it's working... If you look at that screenshot above it has Ubisoft and Activision ports both being used and prioritized at the same time on different playstations it prioritizes all gaming traffic!
1 Like
Wish I could hear h i s h a m speak on this topic everything I've ever read he posted is badass learned a lot from just reading his post... Solidus1983 badass too just saying!!!
Your rule is telling UPnP to use the whole LAN subnet @ 192.168.1.0/24, My rule was JUST allowing one console (192.168.2.250) to use UPnP, There many devices i have that look for UPnP so that why i limit mine to that one IP.
This is where gaming traffic needs port forwarding for various games.
The script has updated 7 hours ago,
But i dont understand the new system traffic for priorize the console and port
1 Like
Lmao you have wan 1024-65535 rule deny... 
How do you think its working if 0.0.0.0/0 is deny on those ports upnp is not working!#!!
Finally at my laptop...
config upnpd 'config'
option download '1024'
option upload '512'
option external_iface 'wan'
option internal_iface 'lan' # Firewall Rules
option port '5000'
option uuid 'xxxxxxxx'
option enabled '1'
option use_stun '1'
option stun_host 'stun.stunprotocol.org'
option notify_interval '30'
option clean_ruleset_threshold '2'
option clean_ruleset_interval '300'
option igdv1 '1'
option log_output '1'
option force_forwarding '1'
config perm_rule
option comment 'Allow high Ports'
option ext_ports '1024-65535'
option int_addr '0.0.0.0/0'
option int_ports '1024-65535'
option action 'allow'
config perm_rule
option comment 'Deny Low Ports'
option ext_ports '0-1024'
option int_addr '0.0.0.0/0'
option int_ports '0-1024'
option action 'deny'
config perm_rule
option comment 'Main Network'
option ext_ports '1024-65335'
option int_addr '192.168.1.0/24'
option int_ports '1024-65335'
option action 'allow'
config perm_rule
option comment 'Ps5'
option ext_ports '1024-65335'
option int_addr '192.168.1.186'
option int_ports '1024-65335'
option action 'allow'
this is working....
This is not working just so we are clear...
1 Like
How did all this UPnP discussion become part of the Qosify thread??
4 Likes
It might say its working but how can it be with those rules my friend... I suggest you try my setup just edit ip's to your network...
1 Like
sry ill stop talking now just saw something that was wrong just wanted to point it out
1 Like
If a class doesn't have bulk_trigger then the default bulk_trigger isn't applied? This config isn't working for me:
config defaults
option dscp_default_tcp tin1
option dscp_default_udp tin1
option dscp_bulk tin0
option bulk_trigger_timeout 5
option bulk_trigger_pps 100
config class tin0
option ingress LE
option egress LE
config class tin1
option ingress AF11
option egress AF11
Bulk traffic tagged as tin1 isn't going to tin0 unless i use same settings inside the class. Is that an expected behavior?
If a class is referenced, the bulk/prio settings from the class are used. You could fix your config by adding a tin1_default class and moving the bulk_trigger settings there
3 Likes
I would recommend to manually use overhead 44 atm mpu 64 but that is just me... If you really want t drill into this, may I recommend you go and peruse the following project.instructions.
(Side-note: these named options in cake attempt to make a gnarly problem better, instead of having to de novo deduce the applicable overhead/encapsulation one only needs to find the respective information in the documentation from one's ISP; it is just that not all ISPs actually give this information in the terms that cake uses).
2 Likes
Ok so i choose option type manual ? It's just
thanks it' s work fine
egress status:
qdisc cake 1: root refcnt 2 bandwidth 16Mbit diffserv4 dual-srchost nat nowash no-ack-filter split-gso rtt 100ms atm overhead 44 mpu 64
Do you think that DNS hostnames IP should be retrieve at start? And then updated dinamically with the dns. I have services like DoH that doesn't use dnsmasq and their domains doesn't get triggered at start unless i manually resolve them in a computer. As a fix i added the IP of that domains with their dscp tag.
Are you sure this is ADSL 2? As i thought ADSL 2 is 24Mbit down 1.3Mbit upload max?
Might be VDSL 2 on Fiber FTTC (Fiber to the cabinet)
1 Like
no, me i'm on vdsl2 but it was a test for my brother who his has 15 and 800kpbps
for the moment my config of qosify is bad i prefer sqm because in game i take a bullshit with qosify but no with sqm 
Oh sorry, i thought you maybe mixed up with ADSL 2 and VDSL 2 hehe 
1 Like
Its gonna be tricky with VDSL 2 FTTC lines as your speed can change on the DSLAM, So one day you could be on 56Mb download and 15Mb upload and then when the DSLAM moves from you 56 to 61Mb etc.
Possible you might be experiencing the same?
You might be better of with xDSL FTTH and avoid docsis.cable modem ISPs
no my real bandwith is always the same
my fai is orange in france