Re: [LARTC] which NIC is which
On Tue, 9 Jul 2002, Michael T. Babcock wrote: Arthur van Leeuwen wrote: Unfortunately, in systems with identical cards that are configured using plug-and-play methods such as those used by PCI random is the best shot you have... 'Deterministic' is more accurate. It seems to be random, on first boot. But it will almost never change after that unless you make hardware changes, in my experience. Not in mine. Every time the PCI bus settings (can't recall the name right now, have been playing with a fixed hardware platform for quite a while) get reset there will probably be a different assignation. Yeah, it'll probably be deterministic *somehow*, but you'll be hard pressed to come up with any rules. Doei, Arthur. -- /\/ | [EMAIL PROTECTED] | Work like you don't need the money /__\ / | A friend is someone with whom | Love like you have never been hurt /\/__ | you can dare to be yourself | Dance like there's nobody watching ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] automatic classes
On Wed, 10 Jul 2002, [iso-8859-1] John Bäckstrand wrote: [snip] 2) I would _want_ to traffic shape based on mac, not IP, but this doesnt seem possible. It isnt vital for me though, ip will work. Actually, it is possible, using netfilter and fwmarks. Netfilter can actually match based on MAC address. Problem is that you can't have it without specifying shit yourself. You could conceivably write a script to take a list of MAC addresses to shape and generate the corresponding tc calls... and then seed that from your ARP table. That would be the closest to automatic you can get. Doei, Arthur. -- /\/ | [EMAIL PROTECTED] | Work like you don't need the money /__\ / | A friend is someone with whom | Love like you have never been hurt /\/__ | you can dare to be yourself | Dance like there's nobody watching ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
[LARTC] ingress qdisc on kernel 2.2.21 with ds8 patch
Since some other constraints require me to still run 2.2 kernel. (drivers for some hardware not working under 2.4) I tried to control the ingress traffic with the ingress qdisc with no success whatsoever. I got 2.2.21 kernel and installed the ds8 patch to have the ingress qdisc too. Recompiled, installed kernel and modules. Recompiled tc with diffserv=y. I tried a setup very similar to the example Edge2: ipchains -A input -p tcp -d 0/0 20 --mark 20 ipchains -A input -p tcp -d 0/0 21 --mark 20 tc qdisc add dev eth0 handle : ingress tc filter add dev eth0 parent : protocol ip prio 50 handle 20 fw police rate 256kbit burst 25k mtu 1.5k drop flowid :1 I tried uploading with a ftp client and there was no traffic limit imposed on what my box received. tc -s qdisc ls shows all counters as 0 (zero). I looked around for this issue and managed to find exactly my problem described in a forum at http://www.geocrawler.com/archives/3/1/2001/5/0/5806401/ but the message was unaswered there. If anyone has any suggestions, they are most welcome. -- ing. Andrei Boros mailto:[EMAIL PROTECTED] / +40-1-303-1870 Centrul pt. Tehnologia Informatiei Societatea Romana de Radiodifuziune ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] Priority Queueing on Linux
Michael T. Babcock wrote: bert hubert wrote: On Tue, Jul 09, 2002 at 12:02:40PM +0800, Patrick Chan wrote: There is priority queueing in Cisco router. Please dig up a link so we can see what 'priority queueing' actually *is*. But I bet that tc has it. I can almost guarantee Patrick is asking about diffserv support. nope, here it is. http://www.cisco.com/univercd/cc/td/doc/product/software/ios121/121cgcr/qos_c/qcprt2/qcdconmg.htm#23965 -- Anton Yurchenko[EMAIL PROTECTED] Digital Generation ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] HZ to be 1000 - 2.5.25 may have this
Adam B. Fineberg wrote: Can this be done in 2.4.18 by changing HZ in include/asm-i386/param.h to 1000? Yes. Would anything else need to be changed, like CLOCKS_PER_SEC (also in param.h)? Hmm, CLOCKS_PER_SEC doesn't look right, particularly if you look at include/asm-ia64/ia32.h:IA32_CLOCKS_PER_SEC. It's only used for some obscure parameter-passing mechanism in ELF, so the damage should be quite limited. (I.e. I've never noticed anything going wrong when changing HZ, and I didn't realize the CLOCKS_PER_SEC dependency until now.) - Werner -- _ / Werner Almesberger, Buenos Aires, Argentina [EMAIL PROTECTED] / /_http://icapeople.epfl.ch/almesber/_/ ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] Priority Queueing on Linux
it is the same as PRIO queue in linux tc. devik On Tue, 9 Jul 2002, bert hubert wrote: On Tue, Jul 09, 2002 at 12:02:40PM +0800, Patrick Chan wrote: There is priority queueing in Cisco router. Is there any equivalent implementation for TC on Linux? If yes, how can I configure and can you give me example? Please dig up a link so we can see what 'priority queueing' actually *is*. But I bet that tc has it. Regards, bert -- http://www.PowerDNS.com Versatile DNS Software Services http://www.tk the dot in .tk http://lartc.org Linux Advanced Routing Traffic Control HOWTO ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/ ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] Questions about HTB classes with an SFQ
sfq MUST be leaf because it is classless. devik On Mon, 8 Jul 2002, Joshua Snyder wrote: I have read most of the documents regarding both Htb and Sfq and I still am not sure if it is best to put the Sfq on the leaf Htb or to put it closer to the root class. All of the docs for Htb say to put it on the leaf classes but the Sfq stuff says to put it closer to the root(but not on the root). If my case every leaf class is only going to be one machine/user. I want to have fairness between machines(I don't want one person to get all of the bandwidth). I don't think putting Sfq on the leaf classes will get me this, but I haven't been able to find a place that says this for sure... josh ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/ ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] HTB does not work
On Wed, Jul 10, 2002 at 11:07:20AM +0300, Alexander Trotsai wrote: [root@watcher root]# tc qdisc add dev eth0 root handle 1: htb default 20 Make doubly sure that you are running a patched tc. Verify which tc you are running - perhaps an old one is lying around. ps. Could I use htb queuening on dotQ subinterfaces? I think you can, but don't trust me on that. Regards, bert -- http://www.PowerDNS.com Versatile DNS Software Services http://www.tk the dot in .tk http://lartc.org Linux Advanced Routing Traffic Control HOWTO ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] routing broadcast messages
On Wed, Jul 10, 2002 at 04:35:16AM +0400, Poltorak Serguei wrote: Hello. I would like to route broadcast messages. For now, if I ping a.b.c.255 from m.n.o.w the packet is passing through each router, except the last, a.b.c.1 (m.n.o.p, other external address) and only he replys to that packet, but not from a.b.c.1, he does it from m.n.o.p address (logic, it's the address of the output interface). Broadcast messages don't leave their subnet. If you want that, you don't need a router but a bridge! Regards, bert -- http://www.PowerDNS.com Versatile DNS Software Services http://www.tk the dot in .tk http://lartc.org Linux Advanced Routing Traffic Control HOWTO ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] which NIC is which
On Tue, Jul 09, 2002 at 02:14:05PM -0700, John Telford wrote: I'm building routers. It's difficult to tell in advance which NIC will be assigned eth0 and which will assigned eth1 when using two NICs. Ping testing usually clears up this simple problem. The identification problem gets worse when adding a third NIC, after sorting out the first two NICs. Frequently the eth0 or eth1 assignments for the first two NICs change. There are tricks to configure based on the MAC address of your interface instead of on its place in the probe. Perhaps googling on that will help. Regards, bert -- http://www.PowerDNS.com Versatile DNS Software Services http://www.tk the dot in .tk http://lartc.org Linux Advanced Routing Traffic Control HOWTO ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: Re: [LARTC] HTB does not work
On Wed, Jul 10, 2002 at 10:29:43AM +0200, bert hubert wrote: bhOn Wed, Jul 10, 2002 at 11:07:20AM +0300, Alexander Trotsai wrote: bh [root@watcher root]# tc qdisc add dev eth0 root handle 1: bh htb default 20 bhMake doubly sure that you are running a patched tc. Verify which tc you are bhrunning - perhaps an old one is lying around. Thanks Seems to be that binary tc from site won't work with my kernel So I get patch, rehdat iproute src.rpm, apply patch and can setup htb policy bh ps. Could I use htb queuening on dotQ subinterfaces? bhI think you can, but don't trust me on that. bhRegards, -- Best regard, Alexander Trotsai aka MAGE-RIPE aka MAGE-UANIC My PGP at ftp://blackhole.adamant.net/pgp/trotsai.key[.asc] Big trouble - old inkjet cartridges emanate barium-based fumes ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] Priority Queueing on Linux
Title: Re: [LARTC] Priority Queueing on Linux Hi, Below is the explanation of priority queueing http://www.cisco.com/univercd/cc/td/doc/cisintwk/ito_doc/qos.htm#xtocid22 I am developing a linux router. There is both data and voice traffic passed thru it. When I use priority queueing on Cisco router, the voice quality is good, evening downloading a large file. But I use the following tc config, the voice quality is not smooth when downloading a large file: tc qdisc add dev ppp0 root handle 10: cbq bandwidth 112Kbit avpkt 1000 tc class add dev ppp0 parent 10:0 classid 10:100 cbq bandwidth 112Kbit rate 57Kb it allot 1514 weight 5Kbit prio 2 maxburst 20 avpkt 1000 isolated tc class add dev ppp0 parent 10:0 classid 10:200 cbq bandwidth 112Kbit rate 55Kb it allot 1514 weight 5Kbit prio 8 maxburst 20 avpkt 1000 bounded tc qdisc add dev ppp0 parent 10:100 sfq quantum 1514b perturb 15 tc qdisc add dev ppp0 parent 10:200 sfq quantum 1514b perturb 15 tc filter add dev ppp0 parent 10:0 protocol ip prio 2 u32 match ip tos 0x8 0xff flowid 10:100 tc filter add dev ppp0 parent 10:0 protocol ip prio 8 u32 match ip tos 0x00 0xff flowid 10:200 voice packet is tagged with 0x8 in TOS field. Voice traffic only uses 34K bandwidth. So the bandwidth allocated to voice is enough. Is there any better tc config? -- Date: Tue, 9 Jul 2002 11:46:58 +0200 From: bert hubert [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Re: [LARTC] Priority Queueing on Linux On Tue, Jul 09, 2002 at 12:02:40PM +0800, Patrick Chan wrote: There is priority queueing in Cisco router. Is there any equivalent implementation for TC on Linux? If yes, how can I configure and can you give me example? Please dig up a link so we can see what 'priority queueing' actually *is*. But I bet that tc has it. Regards, bert -- http://www.PowerDNS.com Versatile DNS Software Services http://www.tk the dot in .tk http://lartc.org Linux Advanced Routing Traffic Control HOWTO Date: Tue, 09 Jul 2002 09:19:11 -0400 From: Michael T. Babcock [EMAIL PROTECTED] Organization: FibreSpeed Ltd. To: lartc [EMAIL PROTECTED] Subject: Re: [LARTC] Priority Queueing on Linux bert hubert wrote: On Tue, Jul 09, 2002 at 12:02:40PM +0800, Patrick Chan wrote: There is priority queueing in Cisco router. Please dig up a link so we can see what 'priority queueing' actually *is*. But I bet that tc has it. I can almost guarantee Patrick is asking about diffserv support. -- Michael T. Babcock CTO, FibreSpeed Ltd.
[LARTC] QoS on data and voice traffic
Title: QoS on data and voice traffic Hi, Sorry for asking so much questions in this mailing list. Because I have to develop a Linux router, there are both data and voice packets passed thru it. The voice packets are already tagged with 0x8 in TOS field. The bandwidth of the serial link is 115200bps. A voice call takes up 34Kbytes bandwidth. Now, I use the following tc config: tc qdisc add dev ppp0 root handle 10: cbq bandwidth 112Kbit avpkt 1000 tc class add dev ppp0 parent 10:0 classid 10:100 cbq bandwidth 112Kbit rate 57Kb it allot 1514 weight 5Kbit prio 2 maxburst 20 avpkt 1000 isolated tc class add dev ppp0 parent 10:0 classid 10:200 cbq bandwidth 112Kbit rate 55Kb it allot 1514 weight 5Kbit prio 8 maxburst 20 avpkt 1000 bounded tc qdisc add dev ppp0 parent 10:100 sfq quantum 1514b perturb 15 tc qdisc add dev ppp0 parent 10:200 sfq quantum 1514b perturb 15 tc filter add dev ppp0 parent 10:0 protocol ip prio 2 u32 match ip tos 0x8 0xff flowid 10:100 tc filter add dev ppp0 parent 10:0 protocol ip prio 8 u32 match ip tos 0x00 0xff flowid 10:200 10:100 is isolated and the bandwidth allocated to this class is 57Kb, the voice packets are successfully put to class 10:100 because I know it by tc -s qdisc theoretically, the voice quality should be very smooth. But the voice is not smooth, when I start to download a file. If I turn off tc, the quality is even terrible, when download a file. Sorry, I just learn tc for 1 month. Hope someone can help me.
Re: [LARTC] routing broadcast messages
Hello but packets are going To their subnetwork. then m.n.o.w sends packet to a.b.c.255 gateways other than a.b.c.1 doesn't know that a.b.c.255 is a broadcast. it's only a.b.c.1 (m.n.o.p) who discards the packet may be I should redraw my pic. a.b.c.0/24,brd+ -[ a.b.c.1, m.n.o.p ]-m.n.o.w ---pings are going in that direction So, packets are going TO their subnet. Any idea??? thanks, PoltoS/ On Wed, 10 Jul 2002, bert hubert wrote: ;On Wed, Jul 10, 2002 at 04:35:16AM +0400, Poltorak Serguei wrote: ; Hello. ; ; I would like to route broadcast messages. ; For now, if I ping a.b.c.255 from m.n.o.w the packet is passing through ; each router, except the last, a.b.c.1 (m.n.o.p, other external address) ; and only he replys to that packet, but not from a.b.c.1, he does it from ; m.n.o.p address (logic, it's the address of the output interface). ; ;Broadcast messages don't leave their subnet. If you want that, you don't ;need a router but a bridge! ; ;Regards, ; ;bert ; ;-- ;http://www.PowerDNS.com Versatile DNS Software Services ;http://www.tk the dot in .tk ;http://lartc.org Linux Advanced Routing Traffic Control HOWTO ; ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] QoS on data and voice traffic
Hi, You are wrong! The packets must spend less time in your box for 0x08 class. So, if you are using cbq and sfq I don't belive that you will have good results. First you need a scheduler between 0x08 class and 0x00 like PRIO. You can use sfq for 0x00 class or tbf to limit the bandwidth. For 0x08 try to use pfifo. Also you can do the same configuration using htb. Please, for 0x08, use pfifo and for 0x00 sfq! A +, R. --- Patrick Chan [EMAIL PROTECTED] wrote: Hi, Sorry for asking so much questions in this mailing list. Because I have to develop a Linux router, there are both data and voice packets passed thru it. The voice packets are already tagged with 0x8 in TOS field. The bandwidth of the serial link is 115200bps. A voice call takes up 34Kbytes bandwidth. Now, I use the following tc config: tc qdisc add dev ppp0 root handle 10: cbq bandwidth 112Kbit avpkt 1000 tc class add dev ppp0 parent 10:0 classid 10:100 cbq bandwidth 112Kbit rate 57Kb it allot 1514 weight 5Kbit prio 2 maxburst 20 avpkt 1000 isolated tc class add dev ppp0 parent 10:0 classid 10:200 cbq bandwidth 112Kbit rate 55Kb it allot 1514 weight 5Kbit prio 8 maxburst 20 avpkt 1000 bounded tc qdisc add dev ppp0 parent 10:100 sfq quantum 1514b perturb 15 tc qdisc add dev ppp0 parent 10:200 sfq quantum 1514b perturb 15 tc filter add dev ppp0 parent 10:0 protocol ip prio 2 u32 match ip tos 0x8 0xff flowid 10:100 tc filter add dev ppp0 parent 10:0 protocol ip prio 8 u32 match ip tos 0x00 0xff flowid 10:200 10:100 is isolated and the bandwidth allocated to this class is 57Kb, the voice packets are successfully put to class 10:100 because I know it by tc -s qdisc theoretically, the voice quality should be very smooth. But the voice is not smooth, when I start to download a file. If I turn off tc, the quality is even terrible, when download a file. Sorry, I just learn tc for 1 month. Hope someone can help me. __ Do You Yahoo!? Sign up for SBC Yahoo! Dial - First Month Free http://sbc.yahoo.com ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] Priority Queueing on Linux
Anton Yurchenko wrote: nope, here it is. http://www.cisco.com/univercd/cc/td/doc/product/software/ios121/121cgcr/qos_c/qcprt2/qcdconmg.htm#23965 From the page (for those who don't follow links, or for the archives of this list): PQ [Priority Queuing] allows you to define how traffic is prioritized in the network. You configure four traffic priorities. You can define a series of filters based on packet characteristics to cause the router to place traffic into these four queues; the queue with the highest priority is serviced first until it is empty, then the lower queues are serviced in sequence. -- Michael T. Babcock CTO, FibreSpeed Ltd. ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
[LARTC] Re: automatic classes
Is this possible with linux TC ? My problem is I dont know each and every IP-address that will be used. Just wanted to let everyone know that I think I found my solution: http://wipl-wrr.sourceforge.net It automatically creates a classes based on either IP or MAC. --- John Bäckstrand ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] routing broadcast messages
Hello All, What your are trying to do is called directed broadcast, and the linux networking gods believe it is evil (i.e. a security hole) and should not be implemented by routers. See http://www.uwsg.iu.edu/hypermail/linux/net/9707.3/0030.html for example. Eran. Poltorak Serguei wrote: Hello but packets are going To their subnetwork. then m.n.o.w sends packet to a.b.c.255 gateways other than a.b.c.1 doesn't know that a.b.c.255 is a broadcast. it's only a.b.c.1 (m.n.o.p) who discards the packet may be I should redraw my pic. a.b.c.0/24,brd+ -[ a.b.c.1, m.n.o.p ]-m.n.o.w ---pings are going in that direction So, packets are going TO their subnet. Any idea??? thanks, PoltoS/ On Wed, 10 Jul 2002, bert hubert wrote: ;On Wed, Jul 10, 2002 at 04:35:16AM +0400, Poltorak Serguei wrote: ; Hello. ; ; I would like to route broadcast messages. ; For now, if I ping a.b.c.255 from m.n.o.w the packet is passing through ; each router, except the last, a.b.c.1 (m.n.o.p, other external address) ; and only he replys to that packet, but not from a.b.c.1, he does it from ; m.n.o.p address (logic, it's the address of the output interface). ; ;Broadcast messages don't leave their subnet. If you want that, you don't ;need a router but a bridge! ; ;Regards, ; ;bert ; ;-- ;http://www.PowerDNS.com Versatile DNS Software Services ;http://www.tk the dot in .tk ;http://lartc.org Linux Advanced Routing Traffic Control HOWTO ; ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/ -- Eran Mann Direct : 972-4-9936297 Senior Software Engineer Fax : 972-4-9890430 Optical AccessEmail : [EMAIL PROTECTED] ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] which NIC is which
Thus spake bert hubert: On Tue, Jul 09, 2002 at 02:14:05PM -0700, John Telford wrote: I'm building routers. It's difficult to tell in advance which NIC will be assigned eth0 and which will assigned eth1 when using two NICs. Ping testing usually clears up this simple problem. The identification problem gets worse when adding a third NIC, after sorting out the first two NICs. Frequently the eth0 or eth1 assignments for the first two NICs change. There are tricks to configure based on the MAC address of your interface instead of on its place in the probe. Perhaps googling on that will help. A script to rename interfaces with ip link ... set name ... after modprobe according to their hw addresses should suffice. I never tried this because in my experience order of detection was always consistant between reboots assuming there are no hardware chasnges. ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] which NIC is which
On Wed, Jul 10, 2002 at 04:38:55PM +0400, Vladimir B. Savkin wrote: A script to rename interfaces with ip link ... set name ... after modprobe according to their hw addresses should suffice. I never tried this because in my experience order of detection was always consistant between reboots assuming there are no hardware chasnges. If you turn on APIC you are in for a surprise I gather :-) Regards, bert -- http://www.PowerDNS.com Versatile DNS Software Services http://www.tk the dot in .tk http://lartc.org Linux Advanced Routing Traffic Control HOWTO ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] automatic classes
1) I want to deploy a box in bridge mode first of all. 2) I would _want_ to traffic shape based on mac, not IP, but this doesnt seem possible. It isnt vital for me though, ip will work. 3) I want each ip (well, preferrably MAC, but...) to have 3 mbit of bandwidth. On one level this makes no sense. Well, I was a bit unclear, or rather unsure of what I really wanted. I actually want clients that use more bandwidth to have less priority, which is what wrr can do. What if you have more clients than bandwidth? Clearly you want to lower the 3MB to bandwidth/#clients. What if you have extra bandwidth? You probably want to share the excess rather than waste the bandwidth. In fact you probably want to share the bandwidth dynamically among the active clients. All of this is just what's done by SFQ Interesting. But would SFQ for example, result in low latency on a heavily used connection for clients that doesnt use much bandwidth? --- John Bäckstrand ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
RE: [LARTC] HZ to be 1000 - 2.5.25 may have this
HZ is set to 100 in that patch.. : -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Dmitriy Sent: Wednesday, July 10, 2002 1:30 AM To: [EMAIL PROTECTED] Subject: Re: [LARTC] HZ to be 1000 - 2.5.25 may have this On Tue, Jul 09, 2002 at 06:17:10PM -0700, Adam B. Fineberg wrote: bert hubert wrote: From lwn.net: The current development kernel remains 2.5.24. Linus has not released any kernels - or surfaced on the linux-kernel mailing list - since before OLS and the Kernel Summit. Some patches are beginning to show up in his BitKeeper tree, however; they include some SCSI updates, an NTFS update, and, interestingly, a change of the internal x86 clock frequency to 1000 Hz. 1000Hz would mean great things for us shaping people! Can this be done in 2.4.18 by changing HZ in include/asm-i386/param.h to 1000? Would anything else need to be changed, like CLOCKS_PER_SEC (also in param.h)? Try to download this url: http://www.ibiblio.org/gentoo/distfiles/linux-gentoo-2.4.19-crypto-r7.pa tch.bz2 This is gentoo Linux patch to 2.4.18, which adds Low Latency Scheduling, Preemptible Kernel, rmap, Ingo Molnar o(1) scheduler and by default it has HZ equal to 1000. Also there is XFS, Grsec and lots of other usefull staff. Patch is about 6.4Mbyte. Luck. Best regards, Adam Fineberg [EMAIL PROTECTED] ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/ -- With Respect Dmitriy Kuznetsov ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/ ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
[LARTC] Subnet/routing question
Forgive me if this is something so simple that i should already know it, but i need to understand if i can accomplish this with Linux routing. We have a /26 subnet from our ISP, and we have been using a Linux box as a firewall to put all our workstations behind NAT, with port forwarding for any box that's providing a service to the whole world. We now need to implement a system which will not support any kind of NAT - it requires having an external IP. Is it possible to use Linux routing to break up the /26 subnet into two /27 subnets, and to do NAT on addresses in one of the /27 subnets and to route addresses on the other /27 subnet straight through to that internal network? This is what i've tried, which i haven't gotten to work: ISP [x.y.z.193/26] | | [x.y.z.194/27 eth0] [x.y.z.225/27 eth0:0] Linux Firewall [192.168.0.0/24 eth1] (for NAT connections through the x.y.z.192/27 subnet) [x.y.z.226/27 eth1:0] (for straight through routing of IP addresses in the x.y.z.224/27 subnet) I've added rules to the routing table to create the connection, but i cannot get a packet with an address in the x.y.z.224/27 range to cross over between eth0 and eth1 in either direction. (Connections using NAT work fine) And the firewall is not stopping them, because the packets still don't go through even when i turn the firewall off. I'm definitely a newbie to the routing area, so maybe my routing table is wrong. What would i need in it? BTW, i'm running Mandrake Linux 8.2 right out of the box. Do i have to recompile the kernel to get some of these options? Thanks! I can't tell you how much i'd appreciate some light on this problem! Larry ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] HZ to be 1000 - 2.5.25 may have this
CIT/Paul wrote: HZ is set to 100 in that patch.. : Try to download this url: http://www.ibiblio.org/gentoo/distfiles/linux-gentoo-2.4.19-crypto-r7.pa tch.bz2 Acutally in this patch the clock rate becomes configurable. HZ is set to CONFIG_JIFFIES which defaults to 1000. Best regards, Adam Fineberg [EMAIL PROTECTED]
[LARTC] Anything out there that is similar to Cisco's WFQ?
From: CIT/Paul [EMAIL PROTECTED] Any help would be greatly appreciated :) This is much better than SFQ : Sounds like SFQ to me. Can you tell us what the differences are? ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] Subnet/routing question
Try: eth0 (external) - x.y.z.193/27 eth1 (internal) - x.y.z.225/27 (non-nat) eth2 (internal) - 192.168.0.0/24 (nat) eth0 - turn on proxy_arp eth1 - turn on proxy_arp eth2 - leave proxy_arp off. This should work just fine. Connections for the eth1-connected addresses will 'forward' through the box (set up your firewall rules appropriately) from eth0 (and vice-versa). To explain what I mean: ipchains -A forward -s x.y.z.255/27 --jump ACCEPT ipchains -A forward -s 192.168.0.0/24 --jump MASQ ... have fun. -- Michael T. Babcock CTO, FibreSpeed Ltd. ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] Anything out there that is similar to Cisco's WFQ?
Don Cohen wrote: From: CIT/Paul [EMAIL PROTECTED] Any help would be greatly appreciated :) This is much better than SFQ : Sounds like SFQ to me. Can you tell us what the differences are? PRIO'd SFQ. If you had classful PRIO with SFQ on each band, you'd probably have a similar effect to what's been described; just a guess though. It seems the desire is to 'ignore' low-priority bands if high-priority bands have traffic, and to balance between those sessions. -- Michael T. Babcock CTO, FibreSpeed Ltd. ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
[LARTC] RE: Anything out there that is similar to Cisco's WFQ?
Paul writes: No SFQ is not like WFQ... WRR is the closest thing to cisco's fair-queue.. WRR keeps track of the connections using the ip_conntrack .. that's sort of what cisco's fair-queue does and it checks the bandwidth streams and gives lower priority to the higher streams and larger packets.. it's meant to reduce latency for traffic shaping and it does :) I haven't tried WRR but it looks like the closest thing to it although it doesn't take everything in to account as cisco's flow based WFQ does.. This is not very convincing. Do you actually know how WFQ works? If so, please tell us. The doc you sent did not describe how it works but what the effects are, and those are entirely consistent with what SFQ does. High bandwidth flows are limited, low bandwidth flows get lower latency. Can you describe some effect that's different? ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] RE: Anything out there that is similar to Cisco's WFQ?
This is not very convincing. Do you actually know how WFQ works? If so, please tell us. The doc you sent did not describe how it works but what the effects are, and those are entirely consistent with what SFQ does. High bandwidth flows are limited, low bandwidth flows get lower latency. Can you describe some effect that's different? I read a bit on WFQ earlier, Im not grasping it totally and I dont know every implementation detail, but I think its basically WRR but taking actual bandwidth usage into account, and not just packet-counts. Well, try this: http://www.sics.se/~ianm/WFQ/wfq_descrip/node21.html Im sure you all can get more out of it than me, a total newbie to queueing theory and QoS. --- John Bäckstrand ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] RE: Anything out there that is similar to Cisco's WFQ?
=?iso-8859-1?Q?John_B=E4ckstrand?= writes: This is not very convincing. Do you actually know how WFQ works? If so, please tell us. The doc you sent did not describe how it works but what the effects are, and those are entirely consistent with what SFQ does. High bandwidth flows are limited, low bandwidth flows get lower latency. Can you describe some effect that's different? I read a bit on WFQ earlier, Im not grasping it totally and I dont know every implementation detail, but I think its basically WRR but taking actual bandwidth usage into account, and not just packet-counts. Well, try this: http://www.sics.se/~ianm/WFQ/wfq_descrip/node21.html This sounds just like SFQ except for the weights. I have a variant of SFQ that does support weights if that's important. It's easy to add. (The hard part is the code that allows you to configure the weights.) I was under the impression that the weights of WFQ isnt actually supposed to be set manually, but rather automatically. This page has a nice picture of WFQ (I think) http://www.cisco.com/warp/public/732/Tech/wfq/ It says: Weight determined by: *Required QoS (IP Procedure, RSVP) *Flow throughput inversely proportional *Frame relay FECN, BECN, DE (for FR Traffic) Only think I actually understood was Flow throughput inversely proportional which is a property I am looking for when trying to find a traffic control implementation. --- John Bäckstrand ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Re: [LARTC] RE: Anything out there that is similar to Cisco's WFQ?
Btw, about the original question (havent got the original email left), there is a WFQ implementation for ALTQ and FreeBSD, but it seems to not work too well: http://www.criticalsoftware.com/research/pdf/Paper-PS.p df http://corn.eos.nasa.gov/qos/qos_results_summary_july98 .html --- John Bäckstrand ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
[LARTC] RE: Priority Queueing on Linux
Hi Patrick I have implement voice and data on a linux machine with the assistance of tc. I have found limited success with tc to this end. I am using an ADSL link of 1.5Mb/256Kb. I previously had a 64Kb ISDN link which I could not really get top quality out of with data and voice. (So I just throw lots of bandwidth at the problem!!) I experimented with cbq and sfq but did not go with this solution as they try to be fair about the data it is handling. Voice does not want fairness as voice is a beast that demands attention and it wants it now!! Instead I have implemented PRIO with tbf on the upstream and have tinkered with tbf on an ingress filter on the downstream. I have no direct experience with imq or htb but I have a feeling they may not assist with the immediacy voice requires. If someone can help us with that point it would be most appreciated. The biggest difference I found was changing the mtu to 300. I have assigned voice to TOS 0x10 and used iptables to mark the traffic appropriately. heres is my tc script: tc qdisc del dev eth0 root /dev/null 21 tc qdisc del dev eth0 ingress /dev/null 21 tc qdisc add dev eth0 root handle 1: prio bands 4 priomap 1 3 2 3 3 3 0 3 3 3 3 3 3 3 3 3 tc qdisc add dev eth0 parent 1:2 handle 20: tbf rate 208kbit buffer 4kb limit 12kb mpu 50 tc qdisc add dev eth0 parent 1:3 handle 30: tbf rate 208kbit buffer 4kb limit 12kb mpu 50 tc qdisc add dev eth0 parent 1:4 handle 40: tbf rate 208kbit buffer 4kb limit 12kb mpu 50 tc qdisc add dev eth0 handle : ingress tc filter add dev eth0 parent : protocol ip prio 2 u32 match ip protocol 17 0xff match ip dport 1101 0x police rate 1536kbit burst 10k drop flowid :1 tc filter add dev eth0 parent : protocol ip prio 6 u32 match ip src 0.0.0.0/0 police rate 1312kbit burst 10k drop flowid :1 I have also played with the priomap, this modification is not necessary. All voice (tos 0x10) goes through the highest priority which I am not rate limiting. The rest of the bands I am rate limiting to 208Kb. On the ingress side I have identified my voice packets and let them have full bandwidth, the rest of the traffic I am rate limiting to 1312Kb. I don't know whether this helps the voice. If someone can comment on this it would be helpful. With this implementation I can have good quality voice conversations whilst downloading (or loading) My assumptions I have made about tc and voice and data are by no means authoritive, so any input on this subject from the group are welcome. Regards Rod Blennerhassett Hi, Below is the explanation of priority queueing http://www.cisco.com/univercd/cc/td/doc/cisintwk/ito_doc/qos.htm#xtocid22 I am developing a linux router. There is both data and voice traffic passed thru it. When I use priority queueing on Cisco router, the voice quality is good, evening downloading a large file. But I use the following tc config, the voice quality is not smooth when downloading a large file: tc qdisc add dev ppp0 root handle 10: cbq bandwidth 112Kbit avpkt 1000 tc class add dev ppp0 parent 10:0 classid 10:100 cbq bandwidth 112Kbit rate 57Kb it allot 1514 weight 5Kbit prio 2 maxburst 20 avpkt 1000 isolated tc class add dev ppp0 parent 10:0 classid 10:200 cbq bandwidth 112Kbit rate 55Kb it allot 1514 weight 5Kbit prio 8 maxburst 20 avpkt 1000 bounded tc qdisc add dev ppp0 parent 10:100 sfq quantum 1514b perturb 15 tc qdisc add dev ppp0 parent 10:200 sfq quantum 1514b perturb 15 tc filter add dev ppp0 parent 10:0 protocol ip prio 2 u32 match ip tos 0x8 0xff flowid 10:100 tc filter add dev ppp0 parent 10:0 protocol ip prio 8 u32 match ip tos 0x00 0xff flowid 10:200 voice packet is tagged with 0x8 in TOS field. Voice traffic only uses 34K bandwidth. So the bandwidth allocated to voice is enough. Is there any better tc config? ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
RE: [LARTC] Priority Queueing on Linux
Title: Message You can use the prio options in qdisc and channel traffic thro' different qdiscs. Another option is to set TOS marks and route to qdiscs using the mark filters in u32 classifier. Why don't you look up http://www.docum.net . Staf has a good site going. I've benefitted from it. Mohan -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Patrick ChanSent: Tuesday, July 09, 2002 9:33 AMTo: [EMAIL PROTECTED]Subject: [LARTC] Priority Queueing on Linux There is priority queueing in Cisco router. Is there any equivalent implementation for TC on Linux? If yes, how can I configure and can you give me example? Thx.
Re: [LARTC] ingress qdisc on kernel 2.2.21 with ds8 patch
May be you use ftp in passive mode, where port 20 not used 10.07.2002 11:57:48, Andrei Boros [EMAIL PROTECTED] wrote: Since some other constraints require me to still run 2.2 kernel. (drivers for some hardware not working under 2.4) I tried to control the ingress traffic with the ingress qdisc with no success whatsoever. I got 2.2.21 kernel and installed the ds8 patch to have the ingress qdisc too. Recompiled, installed kernel and modules. Recompiled tc with diffserv=y. I tried a setup very similar to the example Edge2: ipchains -A input -p tcp -d 0/0 20 --mark 20 ipchains -A input -p tcp -d 0/0 21 --mark 20 tc qdisc add dev eth0 handle : ingress tc filter add dev eth0 parent : protocol ip prio 50 handle 20 fw police rate 256kbit burst 25k mtu 1.5k drop flowid :1 I tried uploading with a ftp client and there was no traffic limit imposed on what my box received. tc -s qdisc ls shows all counters as 0 (zero). I looked around for this issue and managed to find exactly my problem described in a forum at http://www.geocrawler.com/archives/3/1/2001/5/0/5806401/ but the message was unaswered there. If anyone has any suggestions, they are most welcome. -- ing. Andrei Boros mailto:[EMAIL PROTECTED] / +40-1-303-1870 Centrul pt. Tehnologia Informatiei Societatea Romana de Radiodifuziune ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/ --- mailto:[EMAIL PROTECTED] BR Alexey Talikov FORTEK --- ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
[LARTC] Re: Ethernet interface shuts down
Bert thanks for your reply It seems routing daemons are not creating problem. with: ifconfig eth1 up I find that I am able to make interface up again, but it disappears after a while if I do not do anything on this machine for a while and comes back again with ifconfig eth1 up I have another similar kind of problem. I will describe it: (R1)eth0---eth1(R2) If for certain time I do not work on the machine then I am not able to ping the remote interface .eg eth0 from R2 and eth1 from R1.While local interfaces are pingable and ifconfig/ip addr shows that that interfaces are up on both sides. Initailly when I rebooted both machines these two boxes were mutually pingable.ip route also recognizes the availability of the networks. In this case also I am not running Any routing daemon. Any help on this will be very appreciable. Thanks in advance, Nitin __ Do You Yahoo!? Sign up for SBC Yahoo! Dial - First Month Free http://sbc.yahoo.com ___ LARTC mailing list / [EMAIL PROTECTED] http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/