Hi all, I aprogramming a server with java sockeand using netem to simulata WAby setting the transmission rate and delay value. When I ruFOUR instances of a server prograin one linux box with differenlistening ports and send packets among thFOUR servers, it seems thathey do nowork, they could not receive any packets. If I turoff thnetem then it's working fine. I even tried with three instances and it's working fintoo. Does anyonhave similar problems? could anyongivme some advice on this? Thanks iadvance. Tri Froanders.nilsson anetinsight.net Tue Jan 15 07:25:13 2008 From: anders.nilssoanetinsight.net (Anders Nilsson) Date: TuJan 15 07:25:29 2008 Subject: verbosneteinfomation Message-ID: <25FB3730E4103446B74DEA71E037E965E1EFB1@silo.netinsight.local> Hi, I wrotan app thauses netem in order to emulate WAN networks behavior. Thhandles into neteis really great and gives the user the possibilities to generatmany of th'real worlds' characteristics in a LAB environment. I would however know if theris any way of getting information back to thapp. When icomes to when netem decides to delay, drop or corrupt packets so thaindications and corruppacket counters may be updated. Iwould for examplbe nice to display a graph of the dropped packets and play with thcorrelation parameter so thayou can illustrate the packeloss bursts over time. If this is nopossiblwithout losing an arm or a leg would it be possiblto calculatthe effects of the correlation parameters to the respectivfunction. I'a daan00b when It comes to the linux environment so bear with me and thanks for any information. /Anders Nilsso -------------- nexpar-------------- AHTML attachmenwas scrubbed... URL: http://lists.linux-foundation.org/pipermail/netem/attachments/20080115/5bf2605d/attachment.htm Froshemminger alinux-foundation.org Tue Jan 15 15:12:48 2008 From: shemminger alinux-foundation.org (Stephen Hemminger) Date: TuJan 15 15:15:36 2008 Subject: neteand java socket In-Reply-To: <8e5cce0e0801140802r163e88a7x1345e3bcbe75cbe5@xxxxxxxxxxxxxx> References: <8e5cce0e0801140802r163e88a7x1345e3bcbe75cbe5@xxxxxxxxxxxxxx> Message-ID: <20080115151248.077b8d97@deepthought> OMon, 14 Jan 2008 11:02:46 -0500 "Tri M. Tran" <ttran@xxxxxxxxxxxx> wrote: > Hi all, I aprogramming a server with java sockeand using netem to > simulata WAby setting the transmission rate and delay value. When > I ruFOUR instances of a server prograin one linux box with > differenlistening ports and send packets among thFOUR servers, it > seems thathey do nowork, they could not receive any packets. If I > turoff thnetem then it's working fine. I even tried with three > instances and it's working fintoo. Does anyonhave similar > problems? could anyongivme some advice on this? > > Thanks iadvance. > > Tri > _______________________________________________ > Netemailing list > Netem@xxxxxxxxxxxxxxxxxxxxxxxxxx > https://lists.linux-foundation.org/mailman/listinfo/netem Try with something simpler (singlprograin C like iperf or ping) and work up to morcomplex examples. -- StepheHemminger <stephen.hemminger@xxxxxxxxxx> Fromongat netcabo.pt Thu Jan 17 03:22:42 2008 From: monganetcabo.pt (monge) Date: Thu Ja17 03:24:37 2008 Subject: NetEhelp! Message-ID: <909DC1569C65A541953B8ACEBC713CBA1266E7@xxxxxxxx.tvcabo> Hello everyone. I'vbeen using netelately, but I ran into some trouble. I havthfollowing commands: tc qdisc add dev eth0 roohandl1: prio tc qdisc add dev eth0 paren1:3 handl30: netem loss 100% tc filter add dev eth0 protocol ip paren1:0 prio 3 u32 match ip ds192.1= 68.1.21/32 match ip dpor50000 0xffff flowid 10:3 This works, buimesses up other connections too. WhaI wanexactly is to block all out-going traffic from the local machin= to thremote host with IP 192.168.1.21 AND port 50000. Everything else s= hould bkeprunning as normal. What's wrong then? Thother probleI have is with the following commands: tc qdisc add dev eth0 roohandl1: htb default 30 tc class add dev eth0 paren1:3 classid 1:1 htb rat20Kbit tc filter add dev eth0 protocol ip paren1:0 prio 3 u32 match ip ds192.1= 68.1.21/32 match ip dpor50000 0xffff flowid 10:3 This lasonsimple doesn't work (I confirmed with tcpdump+wireshark). All I wanis to limithe bandwidth of the out-going packets to the host 1= 92.168.1.21 AND por50000. Any help will bmuch appreciated. Thanks iadvance, Luis Monge. -------------- nexpar-------------- AHTML attachmenwas scrubbed... URL: http://lists.linux-foundation.org/pipermail/netem/attachments/20080117= /79847312/attachment.htm Fromongat netcabo.pt Thu Jan 17 03:27:36 2008 From: monganetcabo.pt (monge) Date: Thu Ja17 03:28:24 2008 Subject: NetEhelp! Message-ID: <909DC1569C65A541953B8ACEBC713CBA1266E8@xxxxxxxx.tvcabo> Hello everyone. I'vbeen using netelately, but I ran into some trouble. I havthfollowing commands: tc qdisc add dev eth0 roohandl1: prio tc qdisc add dev eth0 paren1:3 handl30: netem loss 100% tc filter add dev eth0 protocol ip paren1:0 prio 3 u32 match ip ds192.1= 68.1.21/32 match ip dpor50000 0xffff flowid 10:3 This works, buimesses up other connections too. WhaI wanexactly is to block all out-going traffic from the local machin= to thremote host with IP 192.168.1.21 AND port 50000. Everything else s= hould bkeprunning as normal. What's wrong then? Thother probleI have is with the following commands: tc qdisc add dev eth0 roohandl1: htb default 30 tc class add dev eth0 paren1:3 classid 1:1 htb rat20Kbit tc filter add dev eth0 protocol ip paren1:0 prio 3 u32 match ip ds192.1= 68.1.21/32 match ip dpor50000 0xffff flowid 10:3 This lasonsimple doesn't work (I confirmed with tcpdump+wireshark). All I wanis to limithe bandwidth of the out-going packets to the host 1= 92.168.1.21 AND por50000. Any help will bmuch appreciated. Thanks iadvance, Luis Monge. -------------- nexpar-------------- AHTML attachmenwas scrubbed... URL: http://lists.linux-foundation.org/pipermail/netem/attachments/20080117= /2d1ebf23/attachment.htm From.nambiar atcs.com Thu Jan 17 04:44:10 2008 From: m.nambiar atcs.co(M Nambiar) Date: Thu Ja17 04:44:34 2008 Subject: Ping responstimes with Netem, TBF & HTB Message-ID: <OF21FB77E2.16072BC8-ON652573D3.004483B5-652573D3.0045F923@xxxxxxx> Skipped contenof typmultipart/related-------------- next part ---------= ----- A non-texattachmenwas scrubbed... Name: BW_latency_measurements.xls Type: application/vnd.ms-excel Size: 36352 bytes Desc: noavailable Url : http://lists.linux-foundation.org/pipermail/netem/attachments/2008011= 7/12c98b9a/BW_latency_measurements-0001.xls Froshemminger alinux-foundation.org Thu Jan 17 08:12:35 2008 From: shemminger alinux-foundation.org (Stephen Hemminger) Date: Thu Ja17 08:12:48 2008 Subject: NetEhelp! In-Reply-To: <909DC1569C65A541953B8ACEBC713CBA1266E8@xxxxxxxx.tvcabo> References: <909DC1569C65A541953B8ACEBC713CBA1266E8@xxxxxxxx.tvcabo> Message-ID: <20080117081235.4e298a23@speedy> OThu, 17 Jan 2008 11:27:36 -0000 "monge" <monge@xxxxxxxxxx> wrote: > > Hello everyone. > > I'vbeen using netelately, but I ran into some trouble. > > I havthfollowing commands: > > tc qdisc add dev eth0 roohandl1: prio > tc qdisc add dev eth0 paren1:3 handl30: netem loss 100% > tc filter add dev eth0 protocol ip paren1:0 prio 3 u32 match ip ds192.168.1.21/32 match ip dport 50000 0xffff flowid 10:3 > > This works, buimesses up other connections too. > WhaI wanexactly is to block all out-going traffic from the local machine to the remote host with IP 192.168.1.21 AND port 50000. Everything else should be kept running as normal. > What's wrong then? > Ustc action discard rather than neteif you want to drop all frames Fronetdev aembedromix.ro Thu Jan 17 12:45:45 2008 From: netdev aembedromix.ro (Catalin(ux) M BOIE) Date: Fri Ja18 08:12:30 2008 Subject: NetEhelp! In-Reply-To: <909DC1569C65A541953B8ACEBC713CBA1266E8@xxxxxxxx.tvcabo> References: <909DC1569C65A541953B8ACEBC713CBA1266E8@xxxxxxxx.tvcabo> Message-ID: <478FBE79.8070401@xxxxxxxxxxxxx> mongwrote: > Hello everyone. Hello, Luis! > I'vbeen using netelately, but I ran into some trouble. > > I havthfollowing commands: > > tc qdisc add dev eth0 roohandl1: prio > tc qdisc add dev eth0 paren1:3 handl30: netem loss 100% > tc filter add dev eth0 protocol ip paren1:0 prio 3 u32 match ip ds192.168.1.21/32 match ip dport 50000 0xffff flowid 10:3 > > This works, buimesses up other connections too. > WhaI wanexactly is to block all out-going traffic from the local machine to the remote host with IP 192.168.1.21 AND port 50000. Everything else should be kept running as normal. > What's wrong then? > > Thother probleI have is with the following commands: > > tc qdisc add dev eth0 roohandl1: htb default 30 > tc class add dev eth0 paren1:3 classid 1:1 htb rat20Kbit > tc filter add dev eth0 protocol ip paren1:0 prio 3 u32 match ip ds192.168.1.21/32 match ip dport 50000 0xffff flowid 10:3 > > This lasonsimple doesn't work (I confirmed with tcpdump+wireshark). > All I wanis to limithe bandwidth of the out-going packets to the host 192.168.1.21 AND port 50000. Heris an examplthat will put 192.168.1.21 port 50000 on class 1:10, whera neteqdisc is attached: tc qdisc del dev eth0 root tc qdisc add dev eth0 roohandl1: htb default 30 echo "Add maiclass..." tc class add dev eth0 paren1:0 classid 1:1 htb rat100Mbit ceil 100Mbit echo "Add defaulclass..." tc class add dev eth0 paren1:1 classid 1:30 htb rat99Mbit ceil 100Mbit echo "Add special class and neteqdisc..." tc class add dev eth0 paren1:1 classid 1:10 htb rat1Mbit ceil 100Mbit tc qdisc add dev eth0 paren1:10 neteloss 90% echo "Filter for special class..." tc filter add dev eth0 paren1:0 protocol ip prio 1 u32 \ match ip ds192.168.1.21/32 \ match ip dpor50000 0xffff \ flowid 1:10 > Any help will bmuch appreciated. Hopihelps you. > Thanks iadvance, > Luis Monge. > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Netemailing list > Netem@xxxxxxxxxxxxxxxxxxxxxxxxxx > https://lists.linux-foundation.org/mailman/listinfo/netem -- Catalin(ux) M BOIE http://kernel.embedromix.ro/ Frojames-p amoving-picture.com Thu Jan 31 03:21:12 2008 From: james-p amoving-picture.co(James Pearson) Date: Thu Ja31 03:21:18 2008 Subject: Help isetting up delay with a bridge Message-ID: <47A1AF28.9070406@xxxxxxxxxxxxxxxxxx> I'trying to emulata private WAN link - I've read the docs at: <http://www.linux-foundation.org/en/Net:Netem> but, I still nosurI'm doing everything right ... Results of ping for thactual WAlink are something like: 100 packets transmitted, 100 received, 0% packeloss, tim19865ms rtmin/avg/max/mdev = 177.884/179.033/222.194/5.448 ms, pip3 ttcp, ftp rsync etc. over thlink givapprox 10Mbyte/sec. So far, I'vseup a machine running CentOS4 (2.6.9 based kernel) as a bridgusing thon-board tg3 gigabit NICs and done the following: tc qdisc add dev eth0 roohandl1:0 netem delay 90msec 5msec 25% tc qdisc add dev eth1 roohandl1:0 netem delay 90msec 5msec 25% I'guessing thaI need to 'split' the delay over both devices in the bridg... is this correct? Thresults of ping over this bridge seem to bclosto the real link: 100 packets transmitted, 100 received, 0% packeloss, tim19943ms rtmin/avg/max/mdev = 171.858/182.636/220.074/6.864 ms, pip3 However, wheI sethe above up, I can only get about 300Kbyte/sec using ttcp etc.over thbridge. I'vread thaI need to tune TCP buffer sizes - but, I'm not sure whaand wher...I assume I need to make these changes on the bridge machin- however trying various values for: net.core.rmem_max net.core.wmem_max net.ipv4.tcp_rmem net.ipv4.tcp_wmem makes no differencto th300Kbyte/sec bandwidth rate. Is theranyway I can ge10Mbyte/sec over this bridge? Thanks James Pearson Frolucas alucas-nussbaum.net Thu Jan 31 03:47:14 2008 From: lucas alucas-nussbaum.ne(Lucas Nussbaum) Date: Thu Ja31 04:01:38 2008 Subject: Help isetting up delay with a bridge In-Reply-To: <47A1AF28.9070406@xxxxxxxxxxxxxxxxxx> References: <47A1AF28.9070406@xxxxxxxxxxxxxxxxxx> Message-ID: <20080131114714.GA16216@xxxxxxxxx> O31/01/08 a11:21 +0000, James Pearson wrote: > However, wheI sethe above up, I can only get about 300Kbyte/sec > using ttcp etc.over thbridge. > > I'vread thaI need to tune TCP buffer sizes - but, I'm not sure > whaand wher...I assume I need to make these changes on the bridge > machin- however trying various values for: No, you need to changthossettings on the machines at both ends of your TCP connection. > > net.core.rmem_max > net.core.wmem_max > net.ipv4.tcp_rmem > net.ipv4.tcp_wmem > > makes no differencto th300Kbyte/sec bandwidth rate. > > Is theranyway I can ge10Mbyte/sec over this bridge? If you suspeca TCP problem, teswith UDP (iperf -u). -- | Lucas Nussbaum | lucas@lucas-nussbaum.n http://www.lucas-nussbaum.net/ | | jabber: lucas@xxxxxxxxxxx GPG: 1024D/023B3F4F | Frojames-p amoving-picture.com Thu Jan 31 04:46:31 2008 From: james-p amoving-picture.co(James Pearson) Date: Thu Ja31 04:46:37 2008 Subject: Help isetting up delay with a bridge In-Reply-To: <20080131114714.GA16216@xxxxxxxxx> References: <47A1AF28.9070406@xxxxxxxxxxxxxxxxxx> <20080131114714.GA16216@xxxxxxxxx> Message-ID: <47A1C327.3090102@xxxxxxxxxxxxxxxxxx> Lucas Nussbauwrote: > O31/01/08 a11:21 +0000, James Pearson wrote: > >>However, wheI sethe above up, I can only get about 300Kbyte/sec >>using ttcp etc.over thbridge. >> >>I'vread thaI need to tune TCP buffer sizes - but, I'm not sure >>whaand wher...I assume I need to make these changes on the bridge >>machin- however trying various values for: > > > No, you need to changthossettings on the machines at both ends of > your TCP connection. OK, so how do I managto ge10Mbyte/sec over the real 100mbit WAN link withoumaking changes to machines on either side? Or, is this jusa limitatioof simulating latency via this method? >>net.core.rmem_max >>net.core.wmem_max >>net.ipv4.tcp_rmem >>net.ipv4.tcp_wmem >> >>makes no differencto th300Kbyte/sec bandwidth rate. >> >>Is theranyway I can ge10Mbyte/sec over this bridge? > > > If you suspeca TCP problem, teswith UDP (iperf -u). I'll givthaa go Thanks James Pearson Frolucas alucas-nussbaum.net Thu Jan 31 05:18:00 2008 From: lucas alucas-nussbaum.ne(Lucas Nussbaum) Date: Thu Ja31 05:20:46 2008 Subject: Help isetting up delay with a bridge In-Reply-To: <47A1C327.3090102@xxxxxxxxxxxxxxxxxx> References: <47A1AF28.9070406@xxxxxxxxxxxxxxxxxx> <20080131114714.GA16216@xxxxxxxxx> <47A1C327.3090102@xxxxxxxxxxxxxxxxxx> Message-ID: <20080131131800.GA20157@xxxxxxxxx> O31/01/08 a12:46 +0000, James Pearson wrote: > Lucas Nussbauwrote: >> O31/01/08 a11:21 +0000, James Pearson wrote: >> >>> However, wheI sethe above up, I can only get about 300Kbyte/sec >>> using ttcp etc.over thbridge. >>> >>> I'vread thaI need to tune TCP buffer sizes - but, I'm not sure >>> whaand wher...I assume I need to make these changes on the bridge >>> machin- however trying various values for: >> >> >> No, you need to changthossettings on the machines at both ends of >> your TCP connection. > > OK, so how do I managto ge10Mbyte/sec over the real 100mbit WAN link > withoumaking changes to machines on either side? Or, is this jusa > limitatioof simulating latency via this method? It's a limitatioof TCP, noa limitation of latency emulation. -- | Lucas Nussbaum | lucas@lucas-nussbaum.n http://www.lucas-nussbaum.net/ | | jabber: lucas@xxxxxxxxxxx GPG: 1024D/023B3F4F | Frojames-p amoving-picture.com Wed Jan 30 14:14:18 2008 From: james-p amoving-picture.co(James Pearson) Date: Thu Ja31 08:08:01 2008 Subject: Help isetting up delay with a bridge Message-ID: <a92ebfd90801301414v7fd9f753y8d12f8cbd2952530@xxxxxxxxxxxxxx> I'trying to emulata private WAN link - I've read the docs at: <http://www.linux-foundation.org/en/Net:Netem> but, I still nosurI'm doing everything right ... Results of ping for thactual WAlink are something like: 100 packets transmitted, 100 received, 0% packeloss, tim19865ms rtmin/avg/max/mdev = 177.884/179.033/222.194/5.448 ms, pip3 ttcp, ftp rsync etc. over thlink givapprox 10Mbyte/sec. So far, I'vseup a machine running CentOS4 (2.6.9 based kernel) as a bridgusing thon-board tg3 gigabit NICs and done the following: tc qdisc add dev eth0 roohandl1:0 netem delay 90msec 5msec 25% tc qdisc add dev eth1 roohandl1:0 netem delay 90msec 5msec 25% I'guessing thaI need to 'split' the delay over both devices in the bridg... is this correct? Thresults of ping over this bridge seem to bclosto the real link: 100 packets transmitted, 100 received, 0% packeloss, tim19943ms rtmin/avg/max/mdev = 171.858/182.636/220.074/6.864 ms, pip3 However, wheI sethe above up, I can only get about 300Kbyte/sec using ttcp etc.over thbridge. I'vread thaI need to tune TCP buffer sizes - but, I'm not sure whaand wher...I assume I need to make these changes on the bridge machin- however trying various values for: net.core.rmem_max net.core.wmem_max net.ipv4.tcp_rmem net.ipv4.tcp_wmem makes no differencto th300Kbyte/sec bandwidth rate. Is theranyway I can ge10Mbyte/sec over this bridge? Thanks James Pearson
