similar to: Measuring SFQ performance

Anyone notices that using sfq qdisc adds latency compared to pfifo? With "sfq perturb 10" ping latency is around 100-120ms, while with "pfifo limit 3" drops to ~10-30ms (qdiscs belong to htb class). I''m missing something or is this expected? { $TC_CLASS parent 1:1 classid 1:20 htb prio 0 rate 900kbit ceil 1000kbit burst 7k $TC_QDISC parent 1:20 handle 200: {pfifo

Hello all, I have a few question regarding PRIO and what it is. Is it used to prioritize what classes get what? I read the how to but it isn''t exactly clear. Also about SFQ. Does this disciplines keep traffic within a queue fair? for example if I classify all http connections to 1:10 and I have packets in the queue and another person is also sending packets to the queue.

Hi, I have a 3Mbit (up,down) connection going through a Linux box (Debian 600mhz, 500mb ram) using NAT to approx 125 users. Presently I am shaping by marking packets by their port numbers. I''m prioritizing 22, 23, 25, 80, 81, 110, 443, 500, 3389, 1214, 6881:6889, etc, into their appropriate classes depending on weather there getting more or less bandwidth. This has worked pretty

Please see below Jon Flechsenhaar Boeing WNW Team Network Services (714)-762-1231 202-E7 -----Original Message----- From: Flechsenhaar, Jon J Sent: Wednesday, September 27, 2006 9:35 AM To: lartc@mailman.ds9a.nl Subject: [LARTC] 2.6.14 - HTB/SFQ QoS broken? Jody: I don''t know if you answer basic HTB questions but I''l try anyway. I''m implementing AF and EF with a

Hello, I have one Q. What is better when using as leaf in one leaf with several hosts (like subnet) to splice traffic equaly to user number? Like 2 users = 256/2, 3 = 256/3 and so on. WRR was pretty good for this job, but it''s only for 2.4.21 kernels and mine''s 2.4.22. So what is better: SFQ or TBF? -- Artūras Šlajus _______________________________________________

The limit parameter for SFQ qdiscs isn''t listed in the howto. Based on a mail from this list from 4 may 2002 ("SFQ queue lengths - generalization"), I''d guess it limits the number of packets/bytes in in the queue, but wether it does so by number of bytes, packets, or time spent in the queue I don''t know. Perhaps someone in the know could add a line to the

I set up a little test to see what the behaviour of (e)sfq was - because I couldn''t work it out from the source :-) . I wanted to see where from a slot the packets got dropped when the queue was full. (e)sfq drops from the longest slot to make space for an incoming packet, so it''s not tail drop as such, but the results show me it does drop from the tail of the slot - which

http://ds9a.nl/lartc/manpages I finished some more manpages, and improved others. New are tc-tbf.8, tc-sfq.8, tc-prio.8 and tc-red.8. Please check & use them. http://ds9a.nl/lartc/manpages -- http://www.PowerDNS.com Versatile DNS Software & Services Trilab The Technology People Netherlabs BV / Rent-a-Nerd.nl -

Hi folks, I`ve a problem to use HTB and SFQ. The first script, below, to show a simple configuration, does work fine...!!! But, in the second example, does not work, becouse i put more code to clasify the traffic by protocol, http and ftp in this case. Somebody can tell me the errors? Thx, in advance.- NOTICE: IMQ device is to asociate with ETH1 my external iface. SCRIPT que funciona:

Hi to all, i''m following this guide (http://www.opalsoft.net/qos/DS-28.htm), is very detailed, but i''m a bit confused about queuing disciplinse of leaf classes. In this guide the author uses PFIFO (see the scheme that i attached at message) in this way: # tc class add dev eth0 parent 1:21 handle 210: pfifo lmit 10 # tc class add dev eth0 parent 1:22 handle 220: pfifo lmit 10 #

I''d like to implement something like SFQ, but with a cap of a rate per flow. Essentially, I''d like to be able to limit citrix connections to 30KBps without having an overall traffic flow cap for all citrix connections. Slightly preferable to this would be something that would actually work per flow instead of the stochastic algorithm. Is this possible? Thanks, Bill --

Hi all! I''m trying to shape traffic in a dorm''s network (4 mbit symmetrical internet link, about 200 computers, heavy p2p usage). The router is a p4xeon running linux 2.6.9 with the qnet patches (http://kem.p.lodz.pl/~peter/qnet/). When I activate ip_forward I get >20% packet loss and a lot of duplicates. Any ideas? I attach my shaping script. Thank you very much in advance,

Hello, I wrote a simple traffic shaping script (below) that should have allowed me to shape my internet traffic a bit (ppp0 - adsl 128kbit/64kbit; local interface eth0). The script works only partially - the speed is being limited but too much. Without running this script my download rate is about 10kBytes (with second computer also downloading at about 6kBytes). After running it my download speed

Dear all, I''m reading codes in sch_sqf.c in kernel 2.4 of Linux. I find I messed up with the concept of slot, bucket and hash tables. Could anyone explain the data stucture of the hash tables a little to me. Thanks very much. Best, Franklin BTW: I can not understand why the number of entries of dep[SFQ_DEPTH*] should be twice:( _______________________________________________ LARTC

> From: Martin Devera <devik@cdi.cz> > Subject: [LARTC] (E)SFQ suggestion > Hi, > just simple note. Maybe it is already in progress :) > > There are attempts to replace hashing routine in SFQ to > consider IPs or ports. > What about to use HRR - roundrobin around bunch of IP > adresses and then smaller WRR for ports per IP ? > It would solve both

I''m not a programmer per se (including C), but... I''d like to be able to give the define in sch_sfq.c (of, say, 2.4.20), SFS_DEPTH other values than 128 as an argument on the tc commandline. It could be powers of two up to 2^7 (128) as it seems that 128 is the current maximum. I''m a little anxious to ask the question "How do I do that?" ;-) Instead

> Subject: [LARTC] Small ISP problems (CBQ) > First of all, this is what we want (in network priority order): > 1: SSH - to be realtime always. I don''t think you want this to always be high prio - that includes scp. > 2: HTTP to be fast, always. Clearly can''t be done if you have more http requests than your bandwidth can handle. > 3-> ftp, direct-connect,

Hi, Is it possible to attach another qdisq under a HFSC classfull ? I try it and the configuraton works : $TC qdisc add dev $dev root handle 1: hfsc default 30 $TC class add dev $dev parent 1: classid 1:1 hfsc sc rate ${RATE}mbit ul rate ${RATE}mbit $TC class add dev $dev parent 1:1 classid 1:10 hfsc rt umax 1500b dmax 30ms rate 200kbit $TC class add dev $dev

I read the sched/qdiscs code from kernel source ... and I have some questions : When the .enqueue, .dequeue, .drop, .requeue functions are called ? What is the event that triggers them and how often this event apears ( per second ? ) ? When a qdisc is dequeued ( when it''s limit is reached ) ? Thanks _______________________________________________ LARTC mailing list /