Re: [LARTC] can all internet traffic be directed thru 1 computer on aRouter?
Ian, Let me try to understand. You have a local network were you have many computers that have access to the internet. They all go through one modem/router. So now you want to put a gateway server betwen your LAN and the outside world so you can manage the traffic? It course can be done. If there is anything else you can serve us to make an analisys. Att, Nataniel Klug Gerente Cyber Nett Brazil - Original Message - From: Ian stuart Turnbull [EMAIL PROTECTED] To: lartc@mailman.ds9a.nl Sent: Monday, February 13, 2006 5:32 PM Subject: [LARTC] can all internet traffic be directed thru 1 computer on aRouter? Hello al, Is it possible [indeed is this the right place] to add iptables to force all internet traffic to go thru a particular computer on a LAN? I have a 4 port Router/modem that contains a Busybox v0.61 Linux system. I am able to add entries to the iptables tho' I don't really know what it does yet. I want to be able to use Ethereal on this one computer to check what web pages my children are visiting - being fairly strict I don't want them visiting some of the more perverse sites. A friend told me this is possible. Can anyone help please. _ The new MSN Search Toolbar now includes Desktop search! http://toolbar.msn.co.uk/ ___ LARTC mailing list LARTC@mailman.ds9a.nl http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc ___ LARTC mailing list LARTC@mailman.ds9a.nl http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc
Re: [LARTC] filter performance/optimization questions
Jakub Wartak wrote: Dnia środa, 8 lutego 2006 18:29, Imre Gergely napisał: hi i'm using htb + u32 filters, and i was wondering if there is something one can optimize at this stage. i have a lot of filters (~ 50.000 / interface, and there are two interfaces), and around 4500 classes / interface. the traffic going through this machine is something around 210-230mbit/s at 50kpps. as you can imagine, the load is pretty high. in fact (as it's a dual xeon at 2.4ghz), one CPU is always at 100% when the traffic increases. i did some tests with esfq (that brought down the classes to around 150), but the filters remained, and the load was still 100%. and i get some packet loss because of that. not much, around 1-2%, but it's enough :) is there something i could do to bring the load down? short of replacing the whole system? i didn't find anything performance-related on the net, or in any documentation. thanks. Show your dmesg, cat /proc/interrupts ( or use itop to determine which card/interface is hogging ), lsmod and .config from kernel compilation Also show us ip -s link [EMAIL PROTECTED] root]# cat /proc/interrupts CPU0 CPU1 0: 55921457 383025821IO-APIC-edge timer 1:342259IO-APIC-edge i8042 2: 0 0 XT-PIC cascade 8: 0 0IO-APIC-edge rtc 14: 1 13IO-APIC-edge ide0 24: 23261179891473249 IO-APIC-level ioc0, eth1 25: 305396 1034030719 IO-APIC-level ioc1, eth2 28: 625322546645 IO-APIC-level eth0 NMI: 111277 253384 LOC: 438830354 438830358 ERR: 0 MIS: 0 (eth1 is the download interface. eth2 is the upload, on which currently is no htb) dmesg attached. [EMAIL PROTECTED] root]# lsmod Module Size Used by bcm5700 132208 0 e100 34304 0 mii 5440 1 e100 .config, ip -s link output attached. What ethcards do you have, is NAPI enabled on them ? 02:09.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5704 Gigabit Ethernet (rev 03) 02:09.1 Ethernet controller: Broadcom Corporation NetXtreme BCM5704 Gigabit Ethernet (rev 03) You could also disable connection tracking if that's not done already. iptables is used only on INPUT, for firewall. And finally, are you using any libpcap based application ? only occasionaly, for a couple of seconds. note: the initial system as of starting the thread was replaced with this one. Bootdata ok (command line is root=/dev/md0 nousb) Linux version 2.6.9-2.ast-smp ([EMAIL PROTECTED]) (gcc version 3.3.3 20040412 (Red Hat Linux 3.3.3-7)) #1 SMP Sat Dec 18 13:31:32 EET 2004 BIOS-provided physical RAM map: BIOS-e820: - 0009fc00 (usable) BIOS-e820: 0009fc00 - 000a (reserved) BIOS-e820: 000e - 0010 (reserved) BIOS-e820: 0010 - 4000 (usable) BIOS-e820: ff7c - 0001 (reserved) No mptable found. On node 0 totalpages: 262144 DMA zone: 4096 pages, LIFO batch:1 Normal zone: 258048 pages, LIFO batch:16 HighMem zone: 0 pages, LIFO batch:1 ACPI: Unable to locate RSDP Intel MultiProcessor Specification v1.4 Virtual Wire compatibility mode. OEM ID: TYAN 6Product ID: S28806APIC at: 0xFEE0 Processor #0 15:5 APIC version 16 Processor #1 15:5 APIC version 16 I/O APIC #2 Version 17 at 0xFEC0. I/O APIC #3 Version 17 at 0xFEBFE000. I/O APIC #4 Version 17 at 0xFEBFF000. Processors: 2 Built 1 zonelists Kernel command line: root=/dev/md0 nousb console=tty0 Initializing CPU#0 PID hash table entries: 4096 (order: 12, 131072 bytes) time.c: Using 1.193182 MHz PIT timer. time.c: Detected 1793.890 MHz processor. Console: colour VGA+ 80x25 Dentry cache hash table entries: 262144 (order: 9, 2097152 bytes) Inode-cache hash table entries: 131072 (order: 8, 1048576 bytes) Memory: 1026520k/1048576k available (1809k kernel code, 21300k reserved, 664k data, 176k init) Calibrating delay loop... 3522.56 BogoMIPS (lpj=1761280) Mount-cache hash table entries: 256 (order: 0, 4096 bytes) CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line) CPU: L2 Cache: 1024K (64 bytes/line) Using local APIC NMI watchdog using perfctr0 CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line) CPU: L2 Cache: 1024K (64 bytes/line) CPU0: AMD Opteron(tm) Processor 244 stepping 08 per-CPU timeslice cutoff: 1024.01 usecs. task migration cache decay timeout: 2 msecs. Booting processor 1/1 rip 6000 rsp 10037f25f58 Initializing CPU#1 Calibrating delay loop... 3579.90 BogoMIPS (lpj=1789952) CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line) CPU: L2 Cache: 1024K (64 bytes/line) AMD Opteron(tm) Processor 244 stepping 08 Total of 2 processors activated (7102.46 BogoMIPS). Using IO-APIC 2 Using IO-APIC 3 Using IO-APIC 4 Using local APIC timer interrupts. Detected 12.457 MHz APIC timer.
[LARTC] Re: filter performance/optimization questions (Imre Gergely)
Can you also post : mpstat -P ALL 1 20 iostat -x 1 10 and opreport --symbols ?? ___ LARTC mailing list LARTC@mailman.ds9a.nl http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc
Re: [LARTC] Guarantee ICMP respond time ?
Hi Robin , I didn'd want to fake ICMP echo_reply i forgot to mention that in this test i'm pinging my gateway to be shure that ping response is not bigger for some other reasonds i find that ping response is getting bigger some times with about 10ms but some times it;s doubles or even more but in most time is like constant. Here is some data if you find it intresting with shaper enabled 64 octets from 213.91.166.1: icmp_seq=22 ttl=254 time=30.9 ms 64 octets from 213.91.166.1: icmp_seq=23 ttl=254 time=40.9 ms 64 octets from 213.91.166.1: icmp_seq=24 ttl=254 time=14.3 ms 64 octets from 213.91.166.1: icmp_seq=25 ttl=254 time=14.4 ms 64 octets from 213.91.166.1: icmp_seq=26 ttl=254 time=34.2 ms 64 octets from 213.91.166.1: icmp_seq=27 ttl=254 time=14.2 ms 64 octets from 213.91.166.1: icmp_seq=28 ttl=254 time=14.2 ms 64 octets from 213.91.166.1: icmp_seq=29 ttl=254 time=14.2 ms 64 octets from 213.91.166.1: icmp_seq=30 ttl=254 time=31.1 ms 64 octets from 213.91.166.1: icmp_seq=31 ttl=254 time=14.3 ms 64 octets from 213.91.166.1: icmp_seq=32 ttl=254 time=14.2 ms 64 octets from 213.91.166.1: icmp_seq=33 ttl=254 time=130.9 ms without shaper enabled 64 octets from 213.91.166.1: icmp_seq=10 ttl=254 time=517.2 ms 64 octets from 213.91.166.1: icmp_seq=11 ttl=254 time=545.4 ms 64 octets from 213.91.166.1: icmp_seq=12 ttl=254 time=573.8 ms 64 octets from 213.91.166.1: icmp_seq=13 ttl=254 time=628.6 ms 64 octets from 213.91.166.1: icmp_seq=14 ttl=254 time=635.3 ms 64 octets from 213.91.166.1: icmp_seq=15 ttl=254 time=666.0 ms 64 octets from 213.91.166.1: icmp_seq=16 ttl=254 time=694.3 ms 64 octets from 213.91.166.1: icmp_seq=17 ttl=254 time=718.1 ms 64 octets from 213.91.166.1: icmp_seq=18 ttl=254 time=746.2 ms 64 octets from 213.91.166.1: icmp_seq=19 ttl=254 time=749.8 ms 64 octets from 213.91.166.1: icmp_seq=20 ttl=254 time=778.1 ms Hammond, Robin-David%KB3IEN wrote: well if you want the line to look less conjested to a casual observer you can fake the ICMP echo_reply. (best know which hosts are infact on-line first). Faking the reply does not preclude actualy sending the echo request, but allowing a duplicate (real) reply might look weird... On Tue, 14 Feb 2006, Stanislav Nedelchev wrote: Date: Tue, 14 Feb 2006 22:35:40 +0200 From: Stanislav Nedelchev [EMAIL PROTECTED] To: lartc@mailman.ds9a.nl Subject: [LARTC] Guarantee ICMP respond time ? Hello to all people there . Can i guarantee ICMP respond time no metter how loaded is internet line . i have typical NATed enviroiment like External IP |linux router| LAN - 192.168.0.0/24 i have example setup with IMQ but is it possible to be done also if i attache htb to eth0 and eth1 for example . if i start shaper ping i better that without shaper but it's not guarantted i mean response time is not like constant. Maybe i'm missing something. Is it possible with HTB ot with something else like CBQ ? here is my example setup echo Loading Traffic Shaper IMQ0 Upload tc qdisc del dev imq0 root tc qdisc add dev imq0 root handle 2: htb default 333 r2q 1 tc class add dev imq0 parent 2: classid 2:2 htb rate 192kbit #ICMP tc class add dev imq0 parent 2:2 classid 2:30 htb rate 32kbit prio 0 tc filter add dev imq0 parent 2:0 protocol ip handle 5 fw classid 2:30 tc qdisc add dev imq0 parent 2:30 handle 30: sfq perturb 1 tc class add dev imq0 parent 2:2 classid 2:24 htb rate 96kbit ceil 160kbit prio 1 tc filter add dev imq0 parent 2:0 protocol ip handle 1 fw classid 2:24 tc qdisc add dev imq0 parent 2:24 handle 24: sfq perturb 10 tc class add dev imq0 parent 2:2 classid 2:26 htb rate 32kbit ceil 128kbit prio 3 tc filter add dev imq0 parent 2:0 protocol ip handle 2 fw classid 2:26 #tc qdisc add dev imq0 parent 2:26 handle 26: sfq perturb 10 tc class add dev imq0 parent 2:2 classid 2:28 htb rate 16kbit ceil 64kbit prio 5 tc filter add dev imq0 parent 2:0 protocol ip handle 3 fw classid 2:28 #tc qdisc add dev imq0 parent 2:28 handle 28: sfq perturb 10 tc class add dev imq0 parent 2:2 classid 2:333 htb rate 16kbit ceil 128kbit prio 7 tc qdisc add dev imq0 parent 2:333 handle 333: sfq perturb 10 echo Done #- #- echo Loading Traffic Shaper imq1 Upload tc qdisc del dev imq1 root tc qdisc add dev imq1 root handle 2: htb default 333 r2q 1 tc class add dev imq1 parent 2: classid 2:2 htb rate 192kbit #ICMP tc class add dev imq1 parent 2:2 classid 2:30 htb rate 32kbit prio 0 tc filter add dev imq1 parent 2:0 protocol ip handle 5 fw classid 2:30 tc qdisc add dev imq1 parent 2:30 handle 30: sfq perturb 1 tc class add dev imq1 parent 2:2 classid 2:24 htb rate 96kbit ceil 160kbit prio 1 tc filter add dev imq1 parent 2:0 protocol ip handle 1 fw classid 2:24 tc qdisc add dev imq1 parent 2:24 handle 24: sfq perturb 10 tc class add