OThu, 04 Aug 2005 18:16:56 +0900
Takuya Sugi<sugie@xxxxxxxxx> wrote:
> Thank you for replying.
>
>
> StepheHemminger <shemminger@xxxxxxxx> wrote:
> >
> > Could you send a consoloutputraceback when the panic
> > occurs?
>
>
> Thfollowing logs aroutput, and any input is not accepted.
>
> ---
> Unablto handlkernel paging request at virtual address 00121548
> printing eip:
> 022e1e0b
> *pd= 00000000
> Dops: 0000 [#1]
> Modules linked in: sch_netemd5 ipv6 parport_pc lp parporuhci_hcd hw_random 3c59x floppy dm_snapshot dm_zero dm_mirror ext3 jbd dm_mod
> CPU: 0
> EIP: 0060:[<022e1e0b>] Notainted VLI
> EFLAGS: 00010202 (2.6.9-1.667)
> EIP is audp_rcv+0x29c/0x2c5
> eax: 00000206 ebx: 00000000 ecx: 00400069 edx: 11df0a40
> esi: 001214ac edi: 00a81140 ebp: 00e93824 esp: 023d4f64
> ds: 007b es: 007b ss: 0068
> Process gawk (pid: 11981, threadinfo=023d4000 task=0ffc37f0)
> Stack: 141214ac 0a210180 141214ac 051214ac 0f562500 02374ef8 00000000 02369a00
> 022c29b8 00000000 11e93810 0f562500 022c2ec0 0f562500 02372bd0 00000008
> 00000000 022ab119 0f562500 02369a00 02415fe0 023d4fdc 022ab1b2 00400068
> Call Trace:
> Stack pointer is garbage, noprinting trace
> Code: Bad EIP value.
> <0>Kernel panic - nosyncing: Fatal exception in interrupt
>
Back froholiday. This looks likthe kernel is running out of memory.
Does ialways happen on udp_rcv?
Is ihappening on curren(2.6.12) kernel or only on older kernels?
If you don'specify any parameters to netem, iends up behaving
a slower versioof fifo.
Froshemminger aosdl.org Wed Sep 7 16:11:57 2005
From: shemminger aosdl.org (Stephen Hemminger)
Date: Wed Apr 18 12:51:17 2007
Subject: Kernel panic after 40mirunnning
In-Reply-To: <20050804091112.7048B4CC9A@xxxxxxxxxxxx>
References: <20050803121603.5e7e449c@xxxxxxxxxxxxxxxxx>
<20050804091112.7048B4CC9A@xxxxxxxxxxxx>
Message-ID: <20050907161157.73a591eb@xxxxxxxxxxxxxxxxx>
OThu, 04 Aug 2005 18:16:56 +0900
Takuya Sugi<sugie@xxxxxxxxx> wrote:
> Thank you for replying.
>
>
> StepheHemminger <shemminger@xxxxxxxx> wrote:
> >
> > Could you send a consoloutputraceback when the panic
> > occurs?
>
>
> Thfollowing logs aroutput, and any input is not accepted.
>
> ---
> Unablto handlkernel paging request at virtual address 00121548
> printing eip:
> 022e1e0b
> *pd= 00000000
> Dops: 0000 [#1]
> Modules linked in: sch_netemd5 ipv6 parport_pc lp parporuhci_hcd hw_random 3c59x floppy dm_snapshot dm_zero dm_mirror ext3 jbd dm_mod
> CPU: 0
> EIP: 0060:[<022e1e0b>] Notainted VLI
> EFLAGS: 00010202 (2.6.9-1.667)
> EIP is audp_rcv+0x29c/0x2c5
> eax: 00000206 ebx: 00000000 ecx: 00400069 edx: 11df0a40
> esi: 001214ac edi: 00a81140 ebp: 00e93824 esp: 023d4f64
> ds: 007b es: 007b ss: 0068
> Process gawk (pid: 11981, threadinfo=023d4000 task=0ffc37f0)
> Stack: 141214ac 0a210180 141214ac 051214ac 0f562500 02374ef8 00000000 02369a00
> 022c29b8 00000000 11e93810 0f562500 022c2ec0 0f562500 02372bd0 00000008
> 00000000 022ab119 0f562500 02369a00 02415fe0 023d4fdc 022ab1b2 00400068
> Call Trace:
> Stack pointer is garbage, noprinting trace
> Code: Bad EIP value.
> <0>Kernel panic - nosyncing: Fatal exception in interrupt
>
You will gea better back tracif you compile with:
CONFIG_DEBUG_INFO
Location:
-> Kernel hacking
-> Kernel debugging (DEBUG_KERNEL [=y])
CONFIG_FRAME_POINTER
-> Kernel hacking
-> Kernel debugging (DEBUG_KERNEL [=y])
Frosugiat sec.co.jp Thu Sep 8 01:06:43 2005
From: sugiasec.co.jp (Takuya Sugie)
Date: Wed Apr 18 12:51:17 2007
Subject: Kernel panic after 40mirunnning
In-Reply-To: <20050907155949.2a75c8b9@xxxxxxxxxxxxxxxxx>
References: <20050907155949.2a75c8b9@xxxxxxxxxxxxxxxxx>
Message-ID: <20050908080242.B50DC4CC98@xxxxxxxxxxxx>
Hi.
Thank you for replying.
>
> Back froholiday. This looks likthe kernel is running out of memory.
> Does ialways happen on udp_rcv?
>
> Is ihappening on curren(2.6.12) kernel or only on older kernels?
>
Though all logs do noremain,
I think thathkernel panic happened on udp_rcv every time.
And walso tried on kernel 2.6.12.3, buit happened.
Wwill try thkernel-compile with debug option and
gemortrace logs.
Sincerely,
--
Takuya SUGIE
sugie@xxxxxxxxx
Frosugiat sec.co.jp Wed Sep 14 07:56:16 2005
From: sugiasec.co.jp (Takuya Sugie)
Date: Wed Apr 18 12:51:17 2007
Subject: Kernel panic after 40mirunnning
In-Reply-To: <20050907161157.73a591eb@xxxxxxxxxxxxxxxxx>
References: <20050907161157.73a591eb@xxxxxxxxxxxxxxxxx>
Message-ID: <20050914145202.40B5B4CC93@xxxxxxxxxxxx>
Hi.
StepheHemminger <shemminger@xxxxxxxx> wrote:
>
> You will gea better back tracif you compile with:
> CONFIG_DEBUG_INFO
> Location:
> -> Kernel hacking
> -> Kernel debugging (DEBUG_KERNEL [=y])
>
> CONFIG_FRAME_POINTER
> -> Kernel hacking
> -> Kernel debugging (DEBUG_KERNEL [=y])
>
Wgodetail logs on the environment compiled
by thabove-mentioned options.
----
Unablto handlkernel paging request at virtual address 00f9da9c
printing eip:
c035edbe
*p= 00000000
Oops: 0000 [#1]
Modules linked in: sch_netecls_u32 sch_htb md5 ipv6 parport_pc lp parporuhci_hcd hw_random i2c_i801 i2c_core 3c59x mii floppy dm_snapshot dm_zero dm_mirror ext3 jdb dm_mod
CPU: 0
EIP: 0060:[<c035edbe>] Notainted VLI
EFLAGS: 00010206 (2.6.12.3-with_CONFIG_FRAME_POINTER)
EIP is audp_rcv+0xee/0x380
eax: c0436000 ebx: 4014ac12 ecx: cff299c0 edx: 00000100
esi: 0014ac23 edi: 00f9da00 ebp: c0437e68 esp: c0437e18
ds: 007b es: 007b ss: 0068
Process swapper (pid: 0, threadinfo=c0436000 task=c03d9c80)
Stack: 00000000 00000001 c0437e5c 00000096 00000000 00000014 00000056 cee3be12
0000000b cd8e1468 00000246 00000001 00000000 00000034 c3e7d424 121213ac
091214ac c0405fb8 cff9da00 00000000 c0437e90 c033a108 00000001 c7769458
Call Trace:
[<c010408a>] show_stack+0x7a/0x90
[<c0104209>] show_registers+0x149/0x1c0
[<c0104458>] die+0x128/0x290
[<c011a09e>] do_page_fault+0x45e/0x644
[<c0103ba3>]error_code+0x4f/0x54
[<c033a108>] ip_local_deliver+0x98/0x260
[<c033a796>] ip_rcv+0x326/0x4b0
[<c031a237>] netif_receive_skb+0x227/0x280
[<c031a31e>] process_backlog+0x8e/0x120
[<c031a492>] __do_softirq+0x42/0xa0
[<c012989b>] do_softirq+0x2b/0x40
[<c010582c>] do_IRQ+0x3c/0x70
[<c0103b46>] common_interrupt+0x1a/0x20
[<c011680c>] apm_cpu_idle+0xfc/0x180
[<c01010f2>] cpu_idle+0x42/0x60
[<c0438805>] start_kernel+0x175/0x1c0
[<c010019f>] 0xc010019f
Code: 8b 97 a4 00 00 00 89 c8 01 d0 89 87 a8 00 00 00 31 c0 85 c0 74 8f e8 42 48 fc ff 85 c0 0f 85 84 00 00 00 a1 a4 d8 4a c0 ff 40 0c <8b> 87 9c 00 00 00 48 0f
85 4a 02 00 00 89 f8 e8 73fb ff 31
<0>Kernel panic - nosyncing: Fatal exception in interrupt
Thanks iadvance,
--
Takuya SUGIE
sugie@xxxxxxxxx
Froshemminger aosdl.org Wed Sep 14 09:20:46 2005
From: shemminger aosdl.org (Stephen Hemminger)
Date: Wed Apr 18 12:51:17 2007
Subject: Kernel panic after 40mirunnning
In-Reply-To: <20050914145202.40B5B4CC93@xxxxxxxxxxxx>
References: <20050907161157.73a591eb@xxxxxxxxxxxxxxxxx>
<20050914145202.40B5B4CC93@xxxxxxxxxxxx>
Message-ID: <20050914092046.3db04045@xxxxxxxxxxxxxxxxx>
OWed, 14 Sep 2005 23:56:16 +0900
Takuya Sugi<sugie@xxxxxxxxx> wrote:
> Hi.
>
> StepheHemminger <shemminger@xxxxxxxx> wrote:
> >
> > You will gea better back tracif you compile with:
> > CONFIG_DEBUG_INFO
> > Location:
> > -> Kernel hacking
> > -> Kernel debugging (DEBUG_KERNEL [=y])
> >
> > CONFIG_FRAME_POINTER
> > -> Kernel hacking
> > -> Kernel debugging (DEBUG_KERNEL [=y])
> >
>
>
> Wgodetail logs on the environment compiled
> by thabove-mentioned options.
>
> ----
> Unablto handlkernel paging request at virtual address 00f9da9c
> printing eip:
> c035edbe
> *p= 00000000
> Oops: 0000 [#1]
> Modules linked in: sch_netecls_u32 sch_htb md5 ipv6 parport_pc lp parporuhci_hcd hw_random i2c_i801 i2c_core 3c59x mii floppy dm_snapshot dm_zero dm_mirror ext3 jdb dm_mod
> CPU: 0
> EIP: 0060:[<c035edbe>] Notainted VLI
> EFLAGS: 00010206 (2.6.12.3-with_CONFIG_FRAME_POINTER)
> EIP is audp_rcv+0xee/0x380
> eax: c0436000 ebx: 4014ac12 ecx: cff299c0 edx: 00000100
> esi: 0014ac23 edi: 00f9da00 ebp: c0437e68 esp: c0437e18
> ds: 007b es: 007b ss: 0068
> Process swapper (pid: 0, threadinfo=c0436000 task=c03d9c80)
> Stack: 00000000 00000001 c0437e5c 00000096 00000000 00000014 00000056 cee3be12
> 0000000b cd8e1468 00000246 00000001 00000000 00000034 c3e7d424 121213ac
> 091214ac c0405fb8 cff9da00 00000000 c0437e90 c033a108 00000001 c7769458
> Call Trace:
> [<c010408a>] show_stack+0x7a/0x90
> [<c0104209>] show_registers+0x149/0x1c0
> [<c0104458>] die+0x128/0x290
> [<c011a09e>] do_page_fault+0x45e/0x644
> [<c0103ba3>]error_code+0x4f/0x54
> [<c033a108>] ip_local_deliver+0x98/0x260
> [<c033a796>] ip_rcv+0x326/0x4b0
> [<c031a237>] netif_receive_skb+0x227/0x280
> [<c031a31e>] process_backlog+0x8e/0x120
> [<c031a492>] __do_softirq+0x42/0xa0
> [<c012989b>] do_softirq+0x2b/0x40
> [<c010582c>] do_IRQ+0x3c/0x70
> [<c0103b46>] common_interrupt+0x1a/0x20
> [<c011680c>] apm_cpu_idle+0xfc/0x180
> [<c01010f2>] cpu_idle+0x42/0x60
> [<c0438805>] start_kernel+0x175/0x1c0
> [<c010019f>] 0xc010019f
> Code: 8b 97 a4 00 00 00 89 c8 01 d0 89 87 a8 00 00 00 31 c0 85 c0 74 8f e8 42 48 fc ff 85 c0 0f 85 84 00 00 00 a1 a4 d8 4a c0 ff 40 0c <8b> 87 9c 00 00 00 48 0f
> 85 4a 02 00 00 89 f8 e8 73fb ff 31
> <0>Kernel panic - nosyncing: Fatal exception in interrupt
>
>
>
> Thanks iadvance,
Whakernel version?
Could you send ththfull kernel config file and the
lisof devices (lspci output). Iis probably not a netem problem,
buI will investigatanyway.
Frouwelists au-b.de Wed Sep 21 12:26:48 2005
From: uwelists au-b.d(Uwe E. Bilger)
Date: Wed Apr 18 12:51:17 2007
Subject: netestalling when executing bash script
Message-ID: <EAEJJNOGOBFKMFJLNAOAIEACFGAA.uwelists@xxxxxx>
Hello all,
I'trying to simulata congestion, limited over a period of 15
seconds. For thapurpose, I wrota bash script. First, I set the
neteparameters (with thhelp of Stephen Earl's phpnetemgui) to the
following parameters:
tc qdisc del dev eth1 root
tc qdisc del dev eth2 root
tc qdisc add dev eth1 roohandl1: netem delay 100ms
tc qdisc add dev eth1 paren1:1 handl10: htb default 1
tc class add dev eth1 paren10: classid 0:1 htb rat500kbit ceil
500kbit
tc qdisc add dev eth2 roohandl1: netem delay 100ms
tc qdisc add dev eth2 paren1:1 handl10: htb default 1
tc class add dev eth2 paren10: classid 0:1 htb rat500kbit ceil
500kbit
I usnuttcp to transmitraffic over this link:
nuttcp -2000 -l 1000 -vv 10.0.2.1
I theexecutthis bash script, introducing 3% loss for 15 seconds,
and thereverting again, buwithout the delay:
# congested situation
sudo /sbin/tc qdisc del dev lo root
sudo /sbin/tc qdisc del dev eth0 root
sudo /sbin/tc qdisc del dev eth1 root
sudo /sbin/tc qdisc del dev eth2 root
sudo /sbin/tc qdisc add dev eth1 roohandl1: netem delay 100ms
sudo /sbin/tc qdisc add dev eth1 paren1:1 handl10: netem loss 3%
sudo /sbin/tc qdisc add dev eth1 paren10:1 handl20: htb default 1
sudo /sbin/tc class add dev eth1 paren20: classid 0:1 htb rate
500kbiceil 500kbit
sudo /sbin/tc qdisc add dev eth2 roohandl1: netem delay 100ms
sudo /sbin/tc qdisc add dev eth2 paren1:1 handl10: netem loss 3%
sudo /sbin/tc qdisc add dev eth2 paren10:1 handl20: htb default 1
sudo /sbin/tc class add dev eth2 paren20: classid 0:1 htb rate
500kbiceil 500kbit
# rucongested for 15 seconds - increasthis value for longer
congestion
sleep 15;
# back to normal
sudo /sbin/tc qdisc del dev lo root
sudo /sbin/tc qdisc del dev eth0 root
sudo /sbin/tc qdisc del dev eth1 root
sudo /sbin/tc qdisc del dev eth2 root
sudo /sbin/tc qdisc add dev eth1 roohandl1: htb default 1
sudo /sbin/tc class add dev eth1 paren1: classid 0:1 htb rate
500kbiceil 500kbit
sudo /sbin/tc qdisc add dev eth2 roohandl1: htb default 1
sudo /sbin/tc class add dev eth2 paren1: classid 0:1 htb rate
500kbiceil 500kbit
Thprobleis that the netem system becomes unresponsive (at least
thnetworking part) when applying thloss. Any idea why?
Any help is kindly appreciated!
Regards,
Uwe
Froshemminger aosdl.org Wed Sep 21 12:45:28 2005
From: shemminger aosdl.org (Stephen Hemminger)
Date: Wed Apr 18 12:51:17 2007
Subject: netestalling when executing bash script
In-Reply-To: <EAEJJNOGOBFKMFJLNAOAIEACFGAA.uwelists@xxxxxx>
References: <EAEJJNOGOBFKMFJLNAOAIEACFGAA.uwelists@xxxxxx>
Message-ID: <20050921124528.74204ad9@localhost.localdomain>
OWed, 21 Sep 2005 21:26:48 +0200
"UwE. Bilger" <uwelists@xxxxxx> wrote:
> Hello all,
>
> I'trying to simulata congestion, limited over a period of 15
> seconds. For thapurpose, I wrota bash script. First, I set the
> neteparameters (with thhelp of Stephen Earl's phpnetemgui) to the
> following parameters:
> tc qdisc del dev eth1 root
> tc qdisc del dev eth2 root
> tc qdisc add dev eth1 roohandl1: netem delay 100ms
> tc qdisc add dev eth1 paren1:1 handl10: htb default 1
> tc class add dev eth1 paren10: classid 0:1 htb rat500kbit ceil
> 500kbit
> tc qdisc add dev eth2 roohandl1: netem delay 100ms
> tc qdisc add dev eth2 paren1:1 handl10: htb default 1
> tc class add dev eth2 paren10: classid 0:1 htb rat500kbit ceil
> 500kbit
>
> I usnuttcp to transmitraffic over this link:
>
> nuttcp -2000 -l 1000 -vv 10.0.2.1
>
> I theexecutthis bash script, introducing 3% loss for 15 seconds,
> and thereverting again, buwithout the delay:
>
> # congested situation
> sudo /sbin/tc qdisc del dev lo root
> sudo /sbin/tc qdisc del dev eth0 root
> sudo /sbin/tc qdisc del dev eth1 root
> sudo /sbin/tc qdisc del dev eth2 root
> sudo /sbin/tc qdisc add dev eth1 roohandl1: netem delay 100ms
> sudo /sbin/tc qdisc add dev eth1 paren1:1 handl10: netem loss 3%
> sudo /sbin/tc qdisc add dev eth1 paren10:1 handl20: htb default 1
> sudo /sbin/tc class add dev eth1 paren20: classid 0:1 htb rate
> 500kbiceil 500kbit
> sudo /sbin/tc qdisc add dev eth2 roohandl1: netem delay 100ms
> sudo /sbin/tc qdisc add dev eth2 paren1:1 handl10: netem loss 3%
> sudo /sbin/tc qdisc add dev eth2 paren10:1 handl20: htb default 1
> sudo /sbin/tc class add dev eth2 paren20: classid 0:1 htb rate
> 500kbiceil 500kbit
>
> # rucongested for 15 seconds - increasthis value for longer
> congestion
> sleep 15;
>
> # back to normal
> sudo /sbin/tc qdisc del dev lo root
> sudo /sbin/tc qdisc del dev eth0 root
> sudo /sbin/tc qdisc del dev eth1 root
> sudo /sbin/tc qdisc del dev eth2 root
> sudo /sbin/tc qdisc add dev eth1 roohandl1: htb default 1
> sudo /sbin/tc class add dev eth1 paren1: classid 0:1 htb rate
> 500kbiceil 500kbit
> sudo /sbin/tc qdisc add dev eth2 roohandl1: htb default 1
> sudo /sbin/tc class add dev eth2 paren1: classid 0:1 htb rate
> 500kbiceil 500kbit
>
>
> Thprobleis that the netem system becomes unresponsive (at least
> thnetworking part) when applying thloss. Any idea why?
>
> Any help is kindly appreciated!
>
Wheyou gethings all knackered the following helps.
tc qdisc ls
also you shouldn'run two instancof netem on top of each other,
iwill gesick. You can change:
tc qdisc add dev eth2 roohandl1: netem delay 100ms
tc qdisc add dev eth2 paren1:1 handl10: netem loss 3%
to
tc qdisc add dev eth2 roohandl1: netem delay 100ms loss 3%
Probably samgoes for htb, do you mean to run two instances.
Frouwelists au-b.de Wed Sep 21 13:07:50 2005
From: uwelists au-b.d(Uwe E. Bilger)
Date: Wed Apr 18 12:51:17 2007
Subject: netestalling when executing bash script
In-Reply-To: <20050921124528.74204ad9@localhost.localdomain>
Message-ID: <EAEJJNOGOBFKMFJLNAOACEADFGAA.uwelists@xxxxxx>
Hello all,
> tc qdisc add dev eth2 roohandl1: netem delay 100ms
> tc qdisc add dev eth2 paren1:1 handl10: netem loss 3%
>
> to
> tc qdisc add dev eth2 roohandl1: netem delay 100ms loss 3%
>
Super, thanks, Stephen. Thahelped.
Thameans Stephen Earl should fila bit on the gui as well ;-)
Seriously, StepheEarl, I greatly appreciatyour interface. Thank
you for it.
Uwe
Frotony.reix abull.net Tue Sep 27 03:09:35 2005
From: tony.reix abull.ne(Tony Reix)
Date: Wed Apr 18 12:51:17 2007
Subject: netefor thbeginners
Message-ID: <1127815775.6849.10.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
Hello,
All my attempts to makneterunning have failed ...
Heris whaI get when launching the first examples in the doc:
# cd ..../iproute2-050816/tc
# ./tc qdisc add dev eth2 roonetedelay 1000ms
RTNETLINK answers: Invalid argument
# ./tc qdisc add dev eth2 roohandl1: cbq bandwidth 10Mbit allot 1514
cell 8 avpk1000 mpu 64
RTNETLINK answers: Invalid argument
# tc qdisc ls dev eth2
RTNETLINK answers: Invalid argument
Dump terminated
I have:
- used kernel 2.6.14-rc1-CITI_NFS4_ALL-1
(2.6.14 + patches for NFSv4)
- goiproute2-tc froiproute2-2.6.11-050330.tar.gz
- added: KERNEL_INCLUDE=/home/common/linux/linux-2.6.14-rc1-netem/
iiproute2-050816/Makefile
- compiled/installed:
# make
Only 1 warning:
Ifilincluded from: ../include/linux/tcp.h:21,
Fross.c:36:
/usr/include/asm/byteorder.h:6:2: attentio: #warning using private
kernel header; includ<endian.h> instead!
# makinstall
# ll /usr/lib/tc/
total 108
-rwxr-xr-x 1 rooroo23077 sep 27 10:06 experimental.dist
-rwxr-xr-x 1 rooroo23573 sep 27 10:06 normal.dist
-rwxr-xr-x 1 rooroo23729 sep 27 10:06 pareto.dist
-rwxr-xr-x 1 rooroo23447 sep 27 10:06 paretonormal.dist
-rwxr-xr-x 1 rooroo 8240 sep 27 10:06 q_netem.so
I'missing something ...
Cayou help ?
Thanks,
Tony
Frotony.reix abull.net Tue Sep 27 09:04:24 2005
From: tony.reix abull.ne(Tony Reix)
Date: Wed Apr 18 12:51:17 2007
Subject: netefor thbeginners
In-Reply-To: <1127815775.6849.10.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
References: <1127815775.6849.10.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
Message-ID: <1127837064.7294.113.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
Lmardi 27 septembr2005 ? 12:09 +0200, Tony Reix a ?crit :
> Hello,
>
> All my attempts to makneterunning have failed ...
>
> Heris whaI get when launching the first examples in the doc:
>
>
> # cd ..../iproute2-050816/tc
> # ./tc qdisc add dev eth2 roonetedelay 1000ms
> RTNETLINK answers: Invalid argument
OK. I'vfound.
As says Stephein http://developer.osdl.org/shemminger/netem/
"Neteis parof the latest 2.4 and 2.6 kernels.
Iis parof the QOS and scheduling tools."
Morprecisely:
Iis in 2.4 and 2.6 kernels and you MUST activatit by:
o2.6.14 :
cd your_Linux_source_directory
makxconfig
Networking
Networking options
QoS and/or fair query (near bottom)
network emulator
SELECT i!
and (as usual):
makmodules
makbzImage
makinstall
makmodules_install
sync
reboo(pray)
I'now ablto see netem running !
(Stephen, maybyou should updatyour page with these infos, for
netem-beginners).
Regards,
Tony
Froshemminger aosdl.org Tue Sep 27 09:14:25 2005
From: shemminger aosdl.org (Stephen Hemminger)
Date: Wed Apr 18 12:51:17 2007
Subject: netefor thbeginners
In-Reply-To: <1127837064.7294.113.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
References: <1127815775.6849.10.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
<1127837064.7294.113.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
Message-ID: <20050927091425.22672db0@xxxxxxxxxxxxxxxxx>
> (Stephen, maybyou should updatyour page with these infos, for
> netem-beginners).
>
I aputting netein my new networking info wiki. So then
everyoncan add to it.
Froshemminger aosdl.org Wed Sep 7 15:59:49 2005
From: shemminger aosdl.org (Stephen Hemminger)
Date: Wed Apr 18 17:37:48 2007
Subject: Kernel panic after 40mirunnning
In-Reply-To: <20050804091112.7048B4CC9A@xxxxxxxxxxxx>
References: <20050803121603.5e7e449c@xxxxxxxxxxxxxxxxx>
<20050804091112.7048B4CC9A@xxxxxxxxxxxx>
Message-ID: <20050907155949.2a75c8b9@xxxxxxxxxxxxxxxxx>
OThu, 04 Aug 2005 18:16:56 +0900
Takuya Sugi<sugie@xxxxxxxxx> wrote:
> Thank you for replying.
>
>
> StepheHemminger <shemminger@xxxxxxxx> wrote:
> >
> > Could you send a consoloutputraceback when the panic
> > occurs?
>
>
> Thfollowing logs aroutput, and any input is not accepted.
>
> ---
> Unablto handlkernel paging request at virtual address 00121548
> printing eip:
> 022e1e0b
> *pd= 00000000
> Dops: 0000 [#1]
> Modules linked in: sch_netemd5 ipv6 parport_pc lp parporuhci_hcd hw_random 3c59x floppy dm_snapshot dm_zero dm_mirror ext3 jbd dm_mod
> CPU: 0
> EIP: 0060:[<022e1e0b>] Notainted VLI
> EFLAGS: 00010202 (2.6.9-1.667)
> EIP is audp_rcv+0x29c/0x2c5
> eax: 00000206 ebx: 00000000 ecx: 00400069 edx: 11df0a40
> esi: 001214ac edi: 00a81140 ebp: 00e93824 esp: 023d4f64
> ds: 007b es: 007b ss: 0068
> Process gawk (pid: 11981, threadinfo=023d4000 task=0ffc37f0)
> Stack: 141214ac 0a210180 141214ac 051214ac 0f562500 02374ef8 00000000 02369a00
> 022c29b8 00000000 11e93810 0f562500 022c2ec0 0f562500 02372bd0 00000008
> 00000000 022ab119 0f562500 02369a00 02415fe0 023d4fdc 022ab1b2 00400068
> Call Trace:
> Stack pointer is garbage, noprinting trace
> Code: Bad EIP value.
> <0>Kernel panic - nosyncing: Fatal exception in interrupt
>
Back froholiday. This looks likthe kernel is running out of memory.
Does ialways happen on udp_rcv?
Is ihappening on curren(2.6.12) kernel or only on older kernels?
If you don'specify any parameters to netem, iends up behaving
a slower versioof fifo.
Froshemminger aosdl.org Wed Sep 7 16:11:57 2005
From: shemminger aosdl.org (Stephen Hemminger)
Date: Wed Apr 18 17:37:48 2007
Subject: Kernel panic after 40mirunnning
In-Reply-To: <20050804091112.7048B4CC9A@xxxxxxxxxxxx>
References: <20050803121603.5e7e449c@xxxxxxxxxxxxxxxxx>
<20050804091112.7048B4CC9A@xxxxxxxxxxxx>
Message-ID: <20050907161157.73a591eb@xxxxxxxxxxxxxxxxx>
OThu, 04 Aug 2005 18:16:56 +0900
Takuya Sugi<sugie@xxxxxxxxx> wrote:
> Thank you for replying.
>
>
> StepheHemminger <shemminger@xxxxxxxx> wrote:
> >
> > Could you send a consoloutputraceback when the panic
> > occurs?
>
>
> Thfollowing logs aroutput, and any input is not accepted.
>
> ---
> Unablto handlkernel paging request at virtual address 00121548
> printing eip:
> 022e1e0b
> *pd= 00000000
> Dops: 0000 [#1]
> Modules linked in: sch_netemd5 ipv6 parport_pc lp parporuhci_hcd hw_random 3c59x floppy dm_snapshot dm_zero dm_mirror ext3 jbd dm_mod
> CPU: 0
> EIP: 0060:[<022e1e0b>] Notainted VLI
> EFLAGS: 00010202 (2.6.9-1.667)
> EIP is audp_rcv+0x29c/0x2c5
> eax: 00000206 ebx: 00000000 ecx: 00400069 edx: 11df0a40
> esi: 001214ac edi: 00a81140 ebp: 00e93824 esp: 023d4f64
> ds: 007b es: 007b ss: 0068
> Process gawk (pid: 11981, threadinfo=023d4000 task=0ffc37f0)
> Stack: 141214ac 0a210180 141214ac 051214ac 0f562500 02374ef8 00000000 02369a00
> 022c29b8 00000000 11e93810 0f562500 022c2ec0 0f562500 02372bd0 00000008
> 00000000 022ab119 0f562500 02369a00 02415fe0 023d4fdc 022ab1b2 00400068
> Call Trace:
> Stack pointer is garbage, noprinting trace
> Code: Bad EIP value.
> <0>Kernel panic - nosyncing: Fatal exception in interrupt
>
You will gea better back tracif you compile with:
CONFIG_DEBUG_INFO
Location:
-> Kernel hacking
-> Kernel debugging (DEBUG_KERNEL [=y])
CONFIG_FRAME_POINTER
-> Kernel hacking
-> Kernel debugging (DEBUG_KERNEL [=y])
Frosugiat sec.co.jp Thu Sep 8 01:06:43 2005
From: sugiasec.co.jp (Takuya Sugie)
Date: Wed Apr 18 17:37:48 2007
Subject: Kernel panic after 40mirunnning
In-Reply-To: <20050907155949.2a75c8b9@xxxxxxxxxxxxxxxxx>
References: <20050907155949.2a75c8b9@xxxxxxxxxxxxxxxxx>
Message-ID: <20050908080242.B50DC4CC98@xxxxxxxxxxxx>
Hi.
Thank you for replying.
>
> Back froholiday. This looks likthe kernel is running out of memory.
> Does ialways happen on udp_rcv?
>
> Is ihappening on curren(2.6.12) kernel or only on older kernels?
>
Though all logs do noremain,
I think thathkernel panic happened on udp_rcv every time.
And walso tried on kernel 2.6.12.3, buit happened.
Wwill try thkernel-compile with debug option and
gemortrace logs.
Sincerely,
--
Takuya SUGIE
sugie@xxxxxxxxx
Frosugiat sec.co.jp Wed Sep 14 07:56:16 2005
From: sugiasec.co.jp (Takuya Sugie)
Date: Wed Apr 18 17:37:48 2007
Subject: Kernel panic after 40mirunnning
In-Reply-To: <20050907161157.73a591eb@xxxxxxxxxxxxxxxxx>
References: <20050907161157.73a591eb@xxxxxxxxxxxxxxxxx>
Message-ID: <20050914145202.40B5B4CC93@xxxxxxxxxxxx>
Hi.
StepheHemminger <shemminger@xxxxxxxx> wrote:
>
> You will gea better back tracif you compile with:
> CONFIG_DEBUG_INFO
> Location:
> -> Kernel hacking
> -> Kernel debugging (DEBUG_KERNEL [=y])
>
> CONFIG_FRAME_POINTER
> -> Kernel hacking
> -> Kernel debugging (DEBUG_KERNEL [=y])
>
Wgodetail logs on the environment compiled
by thabove-mentioned options.
----
Unablto handlkernel paging request at virtual address 00f9da9c
printing eip:
c035edbe
*p= 00000000
Oops: 0000 [#1]
Modules linked in: sch_netecls_u32 sch_htb md5 ipv6 parport_pc lp parporuhci_hcd hw_random i2c_i801 i2c_core 3c59x mii floppy dm_snapshot dm_zero dm_mirror ext3 jdb dm_mod
CPU: 0
EIP: 0060:[<c035edbe>] Notainted VLI
EFLAGS: 00010206 (2.6.12.3-with_CONFIG_FRAME_POINTER)
EIP is audp_rcv+0xee/0x380
eax: c0436000 ebx: 4014ac12 ecx: cff299c0 edx: 00000100
esi: 0014ac23 edi: 00f9da00 ebp: c0437e68 esp: c0437e18
ds: 007b es: 007b ss: 0068
Process swapper (pid: 0, threadinfo=c0436000 task=c03d9c80)
Stack: 00000000 00000001 c0437e5c 00000096 00000000 00000014 00000056 cee3be12
0000000b cd8e1468 00000246 00000001 00000000 00000034 c3e7d424 121213ac
091214ac c0405fb8 cff9da00 00000000 c0437e90 c033a108 00000001 c7769458
Call Trace:
[<c010408a>] show_stack+0x7a/0x90
[<c0104209>] show_registers+0x149/0x1c0
[<c0104458>] die+0x128/0x290
[<c011a09e>] do_page_fault+0x45e/0x644
[<c0103ba3>]error_code+0x4f/0x54
[<c033a108>] ip_local_deliver+0x98/0x260
[<c033a796>] ip_rcv+0x326/0x4b0
[<c031a237>] netif_receive_skb+0x227/0x280
[<c031a31e>] process_backlog+0x8e/0x120
[<c031a492>] __do_softirq+0x42/0xa0
[<c012989b>] do_softirq+0x2b/0x40
[<c010582c>] do_IRQ+0x3c/0x70
[<c0103b46>] common_interrupt+0x1a/0x20
[<c011680c>] apm_cpu_idle+0xfc/0x180
[<c01010f2>] cpu_idle+0x42/0x60
[<c0438805>] start_kernel+0x175/0x1c0
[<c010019f>] 0xc010019f
Code: 8b 97 a4 00 00 00 89 c8 01 d0 89 87 a8 00 00 00 31 c0 85 c0 74 8f e8 42 48 fc ff 85 c0 0f 85 84 00 00 00 a1 a4 d8 4a c0 ff 40 0c <8b> 87 9c 00 00 00 48 0f
85 4a 02 00 00 89 f8 e8 73fb ff 31
<0>Kernel panic - nosyncing: Fatal exception in interrupt
Thanks iadvance,
--
Takuya SUGIE
sugie@xxxxxxxxx
Froshemminger aosdl.org Wed Sep 14 09:20:46 2005
From: shemminger aosdl.org (Stephen Hemminger)
Date: Wed Apr 18 17:37:48 2007
Subject: Kernel panic after 40mirunnning
In-Reply-To: <20050914145202.40B5B4CC93@xxxxxxxxxxxx>
References: <20050907161157.73a591eb@xxxxxxxxxxxxxxxxx>
<20050914145202.40B5B4CC93@xxxxxxxxxxxx>
Message-ID: <20050914092046.3db04045@xxxxxxxxxxxxxxxxx>
OWed, 14 Sep 2005 23:56:16 +0900
Takuya Sugi<sugie@xxxxxxxxx> wrote:
> Hi.
>
> StepheHemminger <shemminger@xxxxxxxx> wrote:
> >
> > You will gea better back tracif you compile with:
> > CONFIG_DEBUG_INFO
> > Location:
> > -> Kernel hacking
> > -> Kernel debugging (DEBUG_KERNEL [=y])
> >
> > CONFIG_FRAME_POINTER
> > -> Kernel hacking
> > -> Kernel debugging (DEBUG_KERNEL [=y])
> >
>
>
> Wgodetail logs on the environment compiled
> by thabove-mentioned options.
>
> ----
> Unablto handlkernel paging request at virtual address 00f9da9c
> printing eip:
> c035edbe
> *p= 00000000
> Oops: 0000 [#1]
> Modules linked in: sch_netecls_u32 sch_htb md5 ipv6 parport_pc lp parporuhci_hcd hw_random i2c_i801 i2c_core 3c59x mii floppy dm_snapshot dm_zero dm_mirror ext3 jdb dm_mod
> CPU: 0
> EIP: 0060:[<c035edbe>] Notainted VLI
> EFLAGS: 00010206 (2.6.12.3-with_CONFIG_FRAME_POINTER)
> EIP is audp_rcv+0xee/0x380
> eax: c0436000 ebx: 4014ac12 ecx: cff299c0 edx: 00000100
> esi: 0014ac23 edi: 00f9da00 ebp: c0437e68 esp: c0437e18
> ds: 007b es: 007b ss: 0068
> Process swapper (pid: 0, threadinfo=c0436000 task=c03d9c80)
> Stack: 00000000 00000001 c0437e5c 00000096 00000000 00000014 00000056 cee3be12
> 0000000b cd8e1468 00000246 00000001 00000000 00000034 c3e7d424 121213ac
> 091214ac c0405fb8 cff9da00 00000000 c0437e90 c033a108 00000001 c7769458
> Call Trace:
> [<c010408a>] show_stack+0x7a/0x90
> [<c0104209>] show_registers+0x149/0x1c0
> [<c0104458>] die+0x128/0x290
> [<c011a09e>] do_page_fault+0x45e/0x644
> [<c0103ba3>]error_code+0x4f/0x54
> [<c033a108>] ip_local_deliver+0x98/0x260
> [<c033a796>] ip_rcv+0x326/0x4b0
> [<c031a237>] netif_receive_skb+0x227/0x280
> [<c031a31e>] process_backlog+0x8e/0x120
> [<c031a492>] __do_softirq+0x42/0xa0
> [<c012989b>] do_softirq+0x2b/0x40
> [<c010582c>] do_IRQ+0x3c/0x70
> [<c0103b46>] common_interrupt+0x1a/0x20
> [<c011680c>] apm_cpu_idle+0xfc/0x180
> [<c01010f2>] cpu_idle+0x42/0x60
> [<c0438805>] start_kernel+0x175/0x1c0
> [<c010019f>] 0xc010019f
> Code: 8b 97 a4 00 00 00 89 c8 01 d0 89 87 a8 00 00 00 31 c0 85 c0 74 8f e8 42 48 fc ff 85 c0 0f 85 84 00 00 00 a1 a4 d8 4a c0 ff 40 0c <8b> 87 9c 00 00 00 48 0f
> 85 4a 02 00 00 89 f8 e8 73fb ff 31
> <0>Kernel panic - nosyncing: Fatal exception in interrupt
>
>
>
> Thanks iadvance,
Whakernel version?
Could you send ththfull kernel config file and the
lisof devices (lspci output). Iis probably not a netem problem,
buI will investigatanyway.
Frouwelists au-b.de Wed Sep 21 12:26:48 2005
From: uwelists au-b.d(Uwe E. Bilger)
Date: Wed Apr 18 17:37:48 2007
Subject: netestalling when executing bash script
Message-ID: <EAEJJNOGOBFKMFJLNAOAIEACFGAA.uwelists@xxxxxx>
Hello all,
I'trying to simulata congestion, limited over a period of 15
seconds. For thapurpose, I wrota bash script. First, I set the
neteparameters (with thhelp of Stephen Earl's phpnetemgui) to the
following parameters:
tc qdisc del dev eth1 root
tc qdisc del dev eth2 root
tc qdisc add dev eth1 roohandl1: netem delay 100ms
tc qdisc add dev eth1 paren1:1 handl10: htb default 1
tc class add dev eth1 paren10: classid 0:1 htb rat500kbit ceil
500kbit
tc qdisc add dev eth2 roohandl1: netem delay 100ms
tc qdisc add dev eth2 paren1:1 handl10: htb default 1
tc class add dev eth2 paren10: classid 0:1 htb rat500kbit ceil
500kbit
I usnuttcp to transmitraffic over this link:
nuttcp -2000 -l 1000 -vv 10.0.2.1
I theexecutthis bash script, introducing 3% loss for 15 seconds,
and thereverting again, buwithout the delay:
# congested situation
sudo /sbin/tc qdisc del dev lo root
sudo /sbin/tc qdisc del dev eth0 root
sudo /sbin/tc qdisc del dev eth1 root
sudo /sbin/tc qdisc del dev eth2 root
sudo /sbin/tc qdisc add dev eth1 roohandl1: netem delay 100ms
sudo /sbin/tc qdisc add dev eth1 paren1:1 handl10: netem loss 3%
sudo /sbin/tc qdisc add dev eth1 paren10:1 handl20: htb default 1
sudo /sbin/tc class add dev eth1 paren20: classid 0:1 htb rate
500kbiceil 500kbit
sudo /sbin/tc qdisc add dev eth2 roohandl1: netem delay 100ms
sudo /sbin/tc qdisc add dev eth2 paren1:1 handl10: netem loss 3%
sudo /sbin/tc qdisc add dev eth2 paren10:1 handl20: htb default 1
sudo /sbin/tc class add dev eth2 paren20: classid 0:1 htb rate
500kbiceil 500kbit
# rucongested for 15 seconds - increasthis value for longer
congestion
sleep 15;
# back to normal
sudo /sbin/tc qdisc del dev lo root
sudo /sbin/tc qdisc del dev eth0 root
sudo /sbin/tc qdisc del dev eth1 root
sudo /sbin/tc qdisc del dev eth2 root
sudo /sbin/tc qdisc add dev eth1 roohandl1: htb default 1
sudo /sbin/tc class add dev eth1 paren1: classid 0:1 htb rate
500kbiceil 500kbit
sudo /sbin/tc qdisc add dev eth2 roohandl1: htb default 1
sudo /sbin/tc class add dev eth2 paren1: classid 0:1 htb rate
500kbiceil 500kbit
Thprobleis that the netem system becomes unresponsive (at least
thnetworking part) when applying thloss. Any idea why?
Any help is kindly appreciated!
Regards,
Uwe
Froshemminger aosdl.org Wed Sep 21 12:45:28 2005
From: shemminger aosdl.org (Stephen Hemminger)
Date: Wed Apr 18 17:37:48 2007
Subject: netestalling when executing bash script
In-Reply-To: <EAEJJNOGOBFKMFJLNAOAIEACFGAA.uwelists@xxxxxx>
References: <EAEJJNOGOBFKMFJLNAOAIEACFGAA.uwelists@xxxxxx>
Message-ID: <20050921124528.74204ad9@localhost.localdomain>
OWed, 21 Sep 2005 21:26:48 +0200
"UwE. Bilger" <uwelists@xxxxxx> wrote:
> Hello all,
>
> I'trying to simulata congestion, limited over a period of 15
> seconds. For thapurpose, I wrota bash script. First, I set the
> neteparameters (with thhelp of Stephen Earl's phpnetemgui) to the
> following parameters:
> tc qdisc del dev eth1 root
> tc qdisc del dev eth2 root
> tc qdisc add dev eth1 roohandl1: netem delay 100ms
> tc qdisc add dev eth1 paren1:1 handl10: htb default 1
> tc class add dev eth1 paren10: classid 0:1 htb rat500kbit ceil
> 500kbit
> tc qdisc add dev eth2 roohandl1: netem delay 100ms
> tc qdisc add dev eth2 paren1:1 handl10: htb default 1
> tc class add dev eth2 paren10: classid 0:1 htb rat500kbit ceil
> 500kbit
>
> I usnuttcp to transmitraffic over this link:
>
> nuttcp -2000 -l 1000 -vv 10.0.2.1
>
> I theexecutthis bash script, introducing 3% loss for 15 seconds,
> and thereverting again, buwithout the delay:
>
> # congested situation
> sudo /sbin/tc qdisc del dev lo root
> sudo /sbin/tc qdisc del dev eth0 root
> sudo /sbin/tc qdisc del dev eth1 root
> sudo /sbin/tc qdisc del dev eth2 root
> sudo /sbin/tc qdisc add dev eth1 roohandl1: netem delay 100ms
> sudo /sbin/tc qdisc add dev eth1 paren1:1 handl10: netem loss 3%
> sudo /sbin/tc qdisc add dev eth1 paren10:1 handl20: htb default 1
> sudo /sbin/tc class add dev eth1 paren20: classid 0:1 htb rate
> 500kbiceil 500kbit
> sudo /sbin/tc qdisc add dev eth2 roohandl1: netem delay 100ms
> sudo /sbin/tc qdisc add dev eth2 paren1:1 handl10: netem loss 3%
> sudo /sbin/tc qdisc add dev eth2 paren10:1 handl20: htb default 1
> sudo /sbin/tc class add dev eth2 paren20: classid 0:1 htb rate
> 500kbiceil 500kbit
>
> # rucongested for 15 seconds - increasthis value for longer
> congestion
> sleep 15;
>
> # back to normal
> sudo /sbin/tc qdisc del dev lo root
> sudo /sbin/tc qdisc del dev eth0 root
> sudo /sbin/tc qdisc del dev eth1 root
> sudo /sbin/tc qdisc del dev eth2 root
> sudo /sbin/tc qdisc add dev eth1 roohandl1: htb default 1
> sudo /sbin/tc class add dev eth1 paren1: classid 0:1 htb rate
> 500kbiceil 500kbit
> sudo /sbin/tc qdisc add dev eth2 roohandl1: htb default 1
> sudo /sbin/tc class add dev eth2 paren1: classid 0:1 htb rate
> 500kbiceil 500kbit
>
>
> Thprobleis that the netem system becomes unresponsive (at least
> thnetworking part) when applying thloss. Any idea why?
>
> Any help is kindly appreciated!
>
Wheyou gethings all knackered the following helps.
tc qdisc ls
also you shouldn'run two instancof netem on top of each other,
iwill gesick. You can change:
tc qdisc add dev eth2 roohandl1: netem delay 100ms
tc qdisc add dev eth2 paren1:1 handl10: netem loss 3%
to
tc qdisc add dev eth2 roohandl1: netem delay 100ms loss 3%
Probably samgoes for htb, do you mean to run two instances.
Frouwelists au-b.de Wed Sep 21 13:07:50 2005
From: uwelists au-b.d(Uwe E. Bilger)
Date: Wed Apr 18 17:37:48 2007
Subject: netestalling when executing bash script
In-Reply-To: <20050921124528.74204ad9@localhost.localdomain>
Message-ID: <EAEJJNOGOBFKMFJLNAOACEADFGAA.uwelists@xxxxxx>
Hello all,
> tc qdisc add dev eth2 roohandl1: netem delay 100ms
> tc qdisc add dev eth2 paren1:1 handl10: netem loss 3%
>
> to
> tc qdisc add dev eth2 roohandl1: netem delay 100ms loss 3%
>
Super, thanks, Stephen. Thahelped.
Thameans Stephen Earl should fila bit on the gui as well ;-)
Seriously, StepheEarl, I greatly appreciatyour interface. Thank
you for it.
Uwe
Frotony.reix abull.net Tue Sep 27 03:09:35 2005
From: tony.reix abull.ne(Tony Reix)
Date: Wed Apr 18 17:37:48 2007
Subject: netefor thbeginners
Message-ID: <1127815775.6849.10.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
Hello,
All my attempts to makneterunning have failed ...
Heris whaI get when launching the first examples in the doc:
# cd ..../iproute2-050816/tc
# ./tc qdisc add dev eth2 roonetedelay 1000ms
RTNETLINK answers: Invalid argument
# ./tc qdisc add dev eth2 roohandl1: cbq bandwidth 10Mbit allot 1514
cell 8 avpk1000 mpu 64
RTNETLINK answers: Invalid argument
# tc qdisc ls dev eth2
RTNETLINK answers: Invalid argument
Dump terminated
I have:
- used kernel 2.6.14-rc1-CITI_NFS4_ALL-1
(2.6.14 + patches for NFSv4)
- goiproute2-tc froiproute2-2.6.11-050330.tar.gz
- added: KERNEL_INCLUDE=/home/common/linux/linux-2.6.14-rc1-netem/
iiproute2-050816/Makefile
- compiled/installed:
# make
Only 1 warning:
Ifilincluded from: ../include/linux/tcp.h:21,
Fross.c:36:
/usr/include/asm/byteorder.h:6:2: attentio: #warning using private
kernel header; includ<endian.h> instead!
# makinstall
# ll /usr/lib/tc/
total 108
-rwxr-xr-x 1 rooroo23077 sep 27 10:06 experimental.dist
-rwxr-xr-x 1 rooroo23573 sep 27 10:06 normal.dist
-rwxr-xr-x 1 rooroo23729 sep 27 10:06 pareto.dist
-rwxr-xr-x 1 rooroo23447 sep 27 10:06 paretonormal.dist
-rwxr-xr-x 1 rooroo 8240 sep 27 10:06 q_netem.so
I'missing something ...
Cayou help ?
Thanks,
Tony
Frotony.reix abull.net Tue Sep 27 09:04:24 2005
From: tony.reix abull.ne(Tony Reix)
Date: Wed Apr 18 17:37:48 2007
Subject: netefor thbeginners
In-Reply-To: <1127815775.6849.10.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
References: <1127815775.6849.10.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
Message-ID: <1127837064.7294.113.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
Lmardi 27 septembr2005 ? 12:09 +0200, Tony Reix a ?crit :
> Hello,
>
> All my attempts to makneterunning have failed ...
>
> Heris whaI get when launching the first examples in the doc:
>
>
> # cd ..../iproute2-050816/tc
> # ./tc qdisc add dev eth2 roonetedelay 1000ms
> RTNETLINK answers: Invalid argument
OK. I'vfound.
As says Stephein http://developer.osdl.org/shemminger/netem/
"Neteis parof the latest 2.4 and 2.6 kernels.
Iis parof the QOS and scheduling tools."
Morprecisely:
Iis in 2.4 and 2.6 kernels and you MUST activatit by:
o2.6.14 :
cd your_Linux_source_directory
makxconfig
Networking
Networking options
QoS and/or fair query (near bottom)
network emulator
SELECT i!
and (as usual):
makmodules
makbzImage
makinstall
makmodules_install
sync
reboo(pray)
I'now ablto see netem running !
(Stephen, maybyou should updatyour page with these infos, for
netem-beginners).
Regards,
Tony
Froshemminger aosdl.org Tue Sep 27 09:14:25 2005
From: shemminger aosdl.org (Stephen Hemminger)
Date: Wed Apr 18 17:37:48 2007
Subject: netefor thbeginners
In-Reply-To: <1127837064.7294.113.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
References: <1127815775.6849.10.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
<1127837064.7294.113.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
Message-ID: <20050927091425.22672db0@xxxxxxxxxxxxxxxxx>
> (Stephen, maybyou should updatyour page with these infos, for
> netem-beginners).
>
I aputting netein my new networking info wiki. So then
everyoncan add to it.
