What is wrong here (continued...)?
Hi all,
Here is a RRD graph ive just made of my packet shaping setup:
http://www.overclockers.com.au/~mwp/temp/tc-1hour.png
The app creating the P-25 traffic is bittorrent on the Linux box also doing the shaping.
P-22 and P-23 is created by FTP transfers.
Starts with bittorrent running.
At 00:03, an FTP transfer is started on "cool.comp" (windows box). At 00:09, bittorrent is stopped.
At 00:13, an FTP transfer is started on "ian.comp" (another windows box). At 00:18, bittorrent is restarted.
At 00:21, FTP on "ian.comp" is stopped.
At 00:25, FTP on "cool.comp" is stopped.
Now whats happening in the graph is the exact opposite of what i want to happen. P-25 is clearly getting more priority than P-22 and P23.
I just cant see how this can be happening with the config below:
#root device
$TC qdisc add dev $INTERFACE root handle 1: htb default 25
#root class
$TC class add dev $INTERFACE parent 1: classid 1:1 htb rate 256kbit
#20 - ICMP, DNS, SYN, SSH
$TC class add dev $INTERFACE parent 1:1 classid 1:20 htb rate 128kbit ceil 256kbit prio 1 $TC qdisc add dev $INTERFACE parent 1:20 handle 20: sfq perturb 10
$TC filter add dev $INTERFACE parent 1:0 prio 0 protocol ip handle 20 fw flowid 1:20
#21 - HTTP/FTP out
$TC class add dev $INTERFACE parent 1:1 classid 1:21 htb rate 128kbit ceil 256kbit prio 2 $TC qdisc add dev $INTERFACE parent 1:21 handle 21: sfq perturb 10
$TC filter add dev $INTERFACE parent 1:0 prio 0 protocol ip handle 21 fw flowid 1:21
#22 - All cool.comp traffic
$TC class add dev $INTERFACE parent 1:1 classid 1:22 htb rate 128kbit ceil 256kbit prio 3 $TC qdisc add dev $INTERFACE parent 1:22 handle 22: sfq perturb 10
$TC filter add dev $INTERFACE parent 1:0 prio 0 protocol ip handle 22 fw flowid 1:22
#23 - All ian.comp traffic
$TC class add dev $INTERFACE parent 1:1 classid 1:23 htb rate 64kbit ceil 256kbit prio 4 $TC qdisc add dev $INTERFACE parent 1:23 handle 23: sfq perturb 10
$TC filter add dev $INTERFACE parent 1:0 prio 0 protocol ip handle 23 fw flowid 1:23
#24 - HTTP/FTP in
$TC class add dev $INTERFACE parent 1:1 classid 1:24 htb rate 64kbit ceil 256kbit prio 5 $TC qdisc add dev $INTERFACE parent 1:24 handle 24: sfq perturb 10
$TC filter add dev $INTERFACE parent 1:0 prio 0 protocol ip handle 24 fw flowid 1:24
#25 - Everything else
$TC class add dev $INTERFACE parent 1:1 classid 1:25 htb rate 15kbit ceil 180kbit prio 8 $TC qdisc add dev $INTERFACE parent 1:25 handle 25: sfq perturb 10
$TC filter add dev $INTERFACE parent 1:0 prio 0 protocol ip handle 25 fw flowid 1:25
As you can see class 25 has priority 8, so all other traffic should get preference, which is not happening.
It almost seems like all the priorities are reversed...
Any ideas?
Thanks,
Mark Williams.
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Quoted 7 lines
Quoted 4 lines
As you can see class 25 has priority 8, so all other traffic should get preference, which is not happening.
It almost seems like all the priorities are reversed...Err... I haven't checked the docs, but I think that's because the priorities are the other way around...
Nope, the HTB man page says:
prio priority
In the round-robin process, classes with the lowest priority field are tried for packets first. Manda- tory.
I did actually try reversed priorities after i made the graph, etc though. It just made things worse... so they are in the correct order.
Mark.
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Quoted 9 lines · Mark Williams (MWP)
Quoted 6 lines
Quoted 3 lines
As you can see class 25 has priority 8, so all other traffic should get preference, which is not happening.
It almost seems like all the priorities are reversed...Err... I haven't checked the docs, but I think that's because the priorities are the other way around...
Nope, the HTB man page says:
Quoted 7 lines
prio priority
In the round-robin process, classes with the lowest priority field are tried for packets first. Manda- tory.I did actually try reversed priorities after i made the graph, etc though. It just made things worse... so they are in the correct order.
If you have classes with different prio's, you have to know what's going on. The class with the lowest prio has the highest priority. This means that that class will get the lowes delay ONLY if this class is not sending more then the configured rate. This also means that this class will get the remaining bandwidth of the parent AFTER the other child classes are served.
Stef
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Quoted 15 lines · Mark Williams (MWP)
Quoted 14 lines
Quoted 6 lines
Quoted 3 lines
As you can see class 25 has priority 8, so all other traffic should get preference, which is not happening.
It almost seems like all the priorities are reversed...Err... I haven't checked the docs, but I think that's because the priorities are the other way around...
Nope, the HTB man page says:
prio priority
In the round-robin process, classes with the lowest priority field are tried for packets first. Manda- tory.
Quoted 8 lines
Quoted 2 lines
I did actually try reversed priorities after i made the graph, etc though. It just made things worse... so they are in the correct order.
If you have classes with different prio's, you have to know what's going on. The class with the lowest prio has the highest priority. This means that that class will get the lowes delay ONLY if this class is not sending more then the configured rate. This also means that this class will get the remaining bandwidth of the parent AFTER the other child classes are served.
Ok, so priorities are ignored as soon as used bandwidth of a class goes above the "rate" setting?
Thanks.
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Quoted 25 lines · Mark Williams (MWP)
Quoted 18 lines · Mark Williams (MWP)
Quoted 9 lines
Quoted 6 lines
Quoted 3 lines
As you can see class 25 has priority 8, so all other traffic should get preference, which is not happening.
It almost seems like all the priorities are reversed...Err... I haven't checked the docs, but I think that's because the priorities are the other way around...
Nope, the HTB man page says:
Quoted 7 lines
prio priority
In the round-robin process, classes with the lowest priority field are tried for packets first. Manda- tory.I did actually try reversed priorities after i made the graph, etc though. It just made things worse... so they are in the correct order.
Quoted 6 lines
If you have classes with different prio's, you have to know what's going on. The class with the lowest prio has the highest priority. This means that that class will get the lowes delay ONLY if this class is not sending more then the configured rate. This also means that this class will get the remaining bandwidth of the parent AFTER the other child classes are served.
Quoted 2 lines
Ok, so priorities are ignored as soon as used bandwidth of a class goes above the "rate" setting?
If a class goes aboive the rate setting, the class will get a bad (big) delay. If it stays under the rate, it will get a low delay.
So prio does 2 things, and in most cases you don't need prio to do what you want to do. Prio is perfect to speed up low rate traffic like ack, telnet, ssh (no scp), ...
Stef
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Quoted 26 lines · Mark Williams (MWP)
Quoted 25 lines
Quoted 24 lines · Mark Williams (MWP)
Quoted 9 lines
Quoted 6 lines
Quoted 3 lines
As you can see class 25 has priority 8, so all other traffic should get preference, which is not happening.
It almost seems like all the priorities are reversed...Err... I haven't checked the docs, but I think that's because the priorities are the other way around...
Nope, the HTB man page says:
Quoted 7 lines
prio priority
In the round-robin process, classes with the lowest priority field are tried for packets first. Manda- tory.I did actually try reversed priorities after i made the graph, etc though. It just made things worse... so they are in the correct order.
If you have classes with different prio's, you have to know what's going on. The class with the lowest prio has the highest priority. This means that that class will get the lowes delay ONLY if this class is not sending more then the configured rate. This also means that this class will get the remaining bandwidth of the parent AFTER the other child classes are served.
Quoted 8 lines
Quoted 2 lines
Ok, so priorities are ignored as soon as used bandwidth of a class goes above the "rate" setting?
If a class goes aboive the rate setting, the class will get a bad (big) delay. If it stays under the rate, it will get a low delay.
So prio does 2 things, and in most cases you don't need prio to do what you want to do. Prio is perfect to speed up low rate traffic like ack, telnet, ssh (no scp), ...
I see...
Ok, ive just tried:
class add dev $INTERFACE parent 1: classid 1:1 htb rate 256kbit
class add dev $INTERFACE parent 1:1 classid 1:20 htb rate 256kbit ceil 256kbit prio 1 class add dev $INTERFACE parent 1:1 classid 1:25 htb rate 32kbit ceil 128kbit prio 8
Even with this, class 25 still seems to get preference of traffic... it just sticks at 128kbit outbound no matter what class 20 is doing.
With what youve said above, i cant see how this is the case?
Thanks again, and Merry Christmas to all,
Mark.
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/