Hello Everyone, I tried to add corruptioby using netem. Buif I am not wrong this corruption is introduced in the communication link and not in the memory. Is therany way thaI can introduce memory corruption? using netem or some other tool. Thanks Froosenbach aus.ibm.com Wed Aug 5 15:02:00 2009 From: osenbach aus.ibm.co(Bryan Osenbach) Date: Wed, 5 Aug 2009 16:02:00 -0600 Subject: AUTO: BryaOsenbach is ouof the office with no access to email and Sametime. (returning 08/12/2009) Message-ID: <OF83A311B8.AA5C0B6D-ON87257609.0079089C-87257609.0079089C@xxxxxxxxxx> I aouof the office until 08/12/2009. I will respond to your messagwhen I return. Note: This is aautomated responsto your message Netem Digest, Vol 52, Issu1 senon 8/5/09 15:52:23. This is thonly notification you will receivwhile this person is away. -------------- nexpar-------------- AHTML attachmenwas scrubbed... URL: http://lists.linux-foundation.org/pipermail/netem/attachments/20090805/d7a92447/attachment.ht Froalmeidaraf agmail.com Fri Aug 7 07:15:41 2009 From: almeidaraf agmail.co(Rafael Almeida) Date: Fri, 7 Aug 2009 11:15:41 -0300 Subject: Correptioin memory In-Reply-To: <20090805215221.13053.qmail@xxxxxxxxx> References: <20090805215221.13053.qmail@xxxxxxxxx> Message-ID: <6de6b1650908070715h2507f028j84e30ec4dab0f28f@xxxxxxxxxxxxxx> OWed, Aug 5, 2009 a6:52 PM, Sushil Chaudhari<schaudhari at mzeal.com> wrote: > Hello Everyone, > > I tried to add corruptioby using netem. Buif I am not wrong this corruption is introduced in the communication link and not in the memory. > > Is therany way thaI can introduce memory corruption? using netem or some other tool. > I'afraid I don'understand what you're asking. What do you mean by memory corruption? Neteis a queudiscipline and it works with network packets. Thcorruption is added to a packeas it passes through netem's queue. Frothis perspective, how would memory corruptiolook like? Frovijaya.vardhan aap.sony.com Thu Aug 13 05:26:35 2009 From: vijaya.vardhaaap.sony.com (Vardhan, Vijaya) Date: Thu, 13 Aug 2009 17:56:35 +0530 Subject: How to deletqdisc Message-ID: <7B7EF7F090B9804A830ACC82F2CDE95D0101D8F6@xxxxxxxxxxxxxxxxxxxxxxx> Hi, Othbelow command, -------------------- # tc qdisc show -------------------- I gean outpuas ------------------------ Qdisc pfifo_fas0: dev eth0 roobands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 ----------------------- Now I wanto deletthis impairment and would like to make my network normally, How do I do that? WheI tried ---------------------------- # tc qdisc del dev eth0 root -------------------------- I goan outpuas --------------------- RTNETLINK answers: No such filor directory -------------------- Pleashelp me, as iis blocking my work. Regards, Hanok ------------------------------------------------------------------- This email is confidential and intended only for thusof the individual or entity named above and may contain information that is privileged. If you are not the intended recipient, you are notified that any dissemination, distribution or copying of this email is strictly prohibited. If you have received this email in error, please notify us immediately by return email or telephone and destroy the original message. - This mail is sent via Sony Asia Pacific Mail Gateway. ------------------------------------------------------------------- -------------- nexpar-------------- AHTML attachmenwas scrubbed... URL: http://lists.linux-foundation.org/pipermail/netem/attachments/20090813/4d1a5251/attachment.ht Froshemminger alinux-foundation.org Fri Aug 14 08:04:10 2009 From: shemminger alinux-foundation.org (Stephen Hemminger) Date: Fri, 14 Aug 2009 08:04:10 -0700 Subject: How to deletqdisc In-Reply-To: <7B7EF7F090B9804A830ACC82F2CDE95D0101D8F6@xxxxxxxxxxxxxxxxxxxxxxx> References: <7B7EF7F090B9804A830ACC82F2CDE95D0101D8F6@xxxxxxxxxxxxxxxxxxxxxxx> Message-ID: <20090814080410.59e7af62@nehalam> OThu, 13 Aug 2009 17:56:35 +0530 "Vardhan, Vijaya" <vijaya.vardhaaap.sony.com> wrote: > Hi, > > > > Othbelow command, > > -------------------- > > # tc qdisc show > > -------------------- > > > > > > > I gean outpuas > > > > ------------------------ > > Qdisc pfifo_fas0: dev eth0 roobands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 > 1 1 1 1 1 > > ----------------------- > > > > Now I wanto deletthis impairment and would like to make my network > normally, You wernorunning netem. "pfifo_fast" is the default builtin qdisc thais used when nothing elsis configured. You can't delete this one, you jusreplacit by setting some other qdisc. Fromehk5abi ayahoo.com Sat Aug 22 03:49:13 2009 From: mehk5abi ayahoo.co(Mehk Punjabi) Date: Sat, 22 Aug 2009 03:49:13 -0700 (PDT) Subject: Netewith ICMP/QoS/UDP traffic? Message-ID: <211073.14576.qm@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> Hello, Casomeonplease help me understand the pros/cons and feasibility of using Netem with ICMP/QoS or UDP traffic? Is there a set of recommended Netem parameters for such traffic? Or, is Netem primarily intended and recommended for TCP/IP apps? Would really appreciatsomguidance. Thank you. -------------- nexpar-------------- AHTML attachmenwas scrubbed... URL: http://lists.linux-foundation.org/pipermail/netem/attachments/20090822/02c3cc51/attachment.ht Frostefano.salsano auniroma2.it Thu Aug 27 09:07:01 2009 From: stefano.salsano auniroma2.i(Stefano Salsano) Date: Thu, 27 Aug 2009 18:07:01 +0200 Subject: fixing correlated loss model inetem In-Reply-To: <67C096F8AA5F0245A7300A7689847639947974@rosie.stockholm.netinsight.local> References: <4A4CAAE6.6080804@xxxxxxxxxxx> <67C096F8AA5F0245A7300A7689847639947974@rosie.stockholm.netinsight.local> Message-ID: <4A96AF25.7060200@xxxxxxxxxxx> Dear all, after my announcemenabounetem2 I have received a comment in a privatmessag(reported below). The message requested for some new features to badded to thloss generator. We have generalized the requests and whavnow added: - improved logging of loss events - generatioof a deterministic loss pattern Moreover, whavimproved the possibility to generate loss patterns according ththGilbert and Gilbert-Elliot loss models widely used in thliterature. (In thprevious version, we had emulated the Gilbert and Gilbert-Elliomodels using our own 4-statmodel, now we have a direcsupporusing their original markov models) By thway whave created a MAN page for netem2, which was missing in thcurrendistribution, where we have included the existing documentatioof neteand we have added the documentation for all our stuff... thMApage is available at: http://netgroup.uniroma2.it/twiki/bin/view.cgi/Main/Netem2ManPage and iis obviously linked within our Netem2 projecpage: http://netgroup.uniroma2.it/twiki/bin/view.cgi/Main/NetEm2 As for logging, iis now possiblto add a new parameter called "logging", where logging 0 => no logging (default) logging 1 => enabllogging for example tc qdisc add/changdev eth0 roonetem2 logging 1 loss_bern 50 will activatlogging which is outpuon dmesg. Thlogging is donas follows, for each loss event the log will include rows idmesg like: netem2 loss even23 RFPLE 2 netem2 loss even24 RFPLE 1 netem2 loss even24 RFPLE 15 wherRFPLE stands for "Received FroPrevious Loss Event" and it counts thgood packets received between two loss events this means thaif you hava burst of consecutive losses RFPLE will be 0, for examplfor a bursof thre consecutive losses: netem2 loss even25 RFPLE 14 netem2 loss even26 RFPLE 0 netem2 loss even27 RFPLE 0 As for thgeneration of a deterministic loss pattern, wcan provide in inpua filwith a sequence of "1" and "0" where "1" are the loss events and "0" arthregular transmissions of packets for examplif you storthe following string to a file: 1111100000000000000000000000000000011111 you will hav5 loss events, then 30 transmissions then 5 loss events loss_patterfilenam[repetitions] wherrepetition is thnumber of requested "replicas" of the loss patterfile repetitions is optional and by defaulis 0 which means infinite repetitioof thloss pattern Waiting for comments... Besregards, Stefano > Stefano, > > Really greawork froyou guys, > > I havanother question thayou might also want to address. There is > another problewhen using Netethat I've not found a solution to. In > Neteyou can specify a certain percentagof packet losses using a > certaidistribution model/correlation. When running a targesystem > abovneteover a period you have affected the target according to your > settings buwin a lab environment need to know what happened at what > timin thnetem engine in order to trace back happenings in the target > system. A easy implementatiowould bto increase a counter every time > theris a decision to drop a packeaccessible externally so that for > exampla graph could periodically poll thcounter and generate a read > oucorresponding to thpacket losses/time. It would also be nice to be > ablto generata instantaneous (one off) selectable burst length drop > of packets > > For example: > > tc qdisc changdev eth0 roonetem burstloss x > > wherx is thnumber of consecutively dropped packets when the command > executes. > > This is a really importanfeaturwhen developing and testing FEC > implementations robustness to packedrops. > > Oncagain thank you for improving NETEM. > -- ******************************************************************* Stefano Salsano Dipartimento Ingegneria Elettronica Universita' di Roma "Tor Vergata" Via del Politecnico, 1 - 00133 Roma - ITALY http://netgroup.uniroma2.it/Stefano_Salsano/ E-mail : stefano.salsano auniroma2.it Cell. : +39 320 4307310 Offic : (Tel.) +39 06 72597770 (Fax.) +39 06 72597435 ******************************************************************* Frostefano.salsano auniroma2.it Thu Aug 27 09:25:06 2009 From: stefano.salsano auniroma2.i(Stefano Salsano) Date: Thu, 27 Aug 2009 18:25:06 +0200 Subject: fixing correlated loss model inetem In-Reply-To: <4A96AF25.7060200@xxxxxxxxxxx> References: <4A4CAAE6.6080804@xxxxxxxxxxx> <67C096F8AA5F0245A7300A7689847639947974@rosie.stockholm.netinsight.local> <4A96AF25.7060200@xxxxxxxxxxx> Message-ID: <4A96B362.2090509@xxxxxxxxxxx> Sorry, I forgoto say thathe new patches (for kernel 2.6.29 and 2.6.30) aravailablwithin our Netem2 project page: http://netgroup.uniroma2.it/twiki/bin/view.cgi/Main/NetEm2 Cheers, Stefano Stefano Salsano wrote: > Dear all, > > after my announcemenabounetem2 I have received a comment in a > privatmessag(reported below). The message requested for some new > features to badded to thloss generator. We have generalized the > requests and whavnow added: > > - improved logging of loss events > - generatioof a deterministic loss pattern > > Moreover, whavimproved the possibility to generate loss patterns > according ththGilbert and Gilbert-Elliot loss models widely used in > thliterature. (In thprevious version, we had emulated the Gilbert > and Gilbert-Elliomodels using our own 4-statmodel, now we have a > direcsupporusing their original markov models) > > By thway whave created a MAN page for netem2, which was missing in > thcurrendistribution, where we have included the existing > documentatioof neteand we have added the documentation for all our > stuff... thMApage is available at: > http://netgroup.uniroma2.it/twiki/bin/view.cgi/Main/Netem2ManPage > > and iis obviously linked within our Netem2 projecpage: > http://netgroup.uniroma2.it/twiki/bin/view.cgi/Main/NetEm2 > > As for logging, iis now possiblto add a new parameter called > "logging", where > logging 0 => no logging (default) > logging 1 => enabllogging > > for example > tc qdisc add/changdev eth0 roonetem2 logging 1 loss_bern 50 > > will activatlogging which is outpuon dmesg. > > Thlogging is donas follows, for each loss event the log will include > rows idmesg like: > > netem2 loss even23 RFPLE 2 > netem2 loss even24 RFPLE 1 > netem2 loss even24 RFPLE 15 > > wherRFPLE stands for "Received FroPrevious Loss Event" and it counts > thgood packets received between two loss events > this means thaif you hava burst of consecutive losses RFPLE will be > 0, for examplfor a bursof thre consecutive losses: > > netem2 loss even25 RFPLE 14 > netem2 loss even26 RFPLE 0 > netem2 loss even27 RFPLE 0 > > As for thgeneration of a deterministic loss pattern, wcan provide in > inpua filwith a sequence of "1" and "0" where "1" are the loss > events and "0" arthregular transmissions of packets > > for examplif you storthe following string to a file: > > 1111100000000000000000000000000000011111 > > you will hav5 loss events, then 30 transmissions then 5 loss events > > loss_patterfilenam[repetitions] > > wherrepetition is thnumber of requested "replicas" of the loss > patterfile > repetitions is optional and by defaulis 0 which means infinite > repetitioof thloss pattern > > Waiting for comments... > > Besregards, > Stefano > >> Stefano, >> >> Really greawork froyou guys, >> >> I havanother question thayou might also want to address. There is >> another problewhen using Netethat I've not found a solution to. In >> Neteyou can specify a certain percentagof packet losses using a >> certaidistribution model/correlation. When running a targesystem >> abovneteover a period you have affected the target according to your >> settings buwin a lab environment need to know what happened at what >> timin thnetem engine in order to trace back happenings in the target >> system. A easy implementatiowould bto increase a counter every time >> theris a decision to drop a packeaccessible externally so that for >> exampla graph could periodically poll thcounter and generate a read >> oucorresponding to thpacket losses/time. It would also be nice to be >> ablto generata instantaneous (one off) selectable burst length drop >> of packets >> >> For example: >> >> tc qdisc changdev eth0 roonetem burstloss x >> >> wherx is thnumber of consecutively dropped packets when the command >> executes. >> >> This is a really importanfeaturwhen developing and testing FEC >> implementations robustness to packedrops. >> >> Oncagain thank you for improving NETEM. >> > -- ******************************************************************* Stefano Salsano Dipartimento Ingegneria Elettronica Universita' di Roma "Tor Vergata" Via del Politecnico, 1 - 00133 Roma - ITALY http://netgroup.uniroma2.it/Stefano_Salsano/ E-mail : stefano.salsano auniroma2.it Cell. : +39 320 4307310 Offic : (Tel.) +39 06 72597770 (Fax.) +39 06 72597435 ******************************************************************* Frodwjames aavaya.com Mon Aug 31 16:02:14 2009 From: dwjames aavaya.co(James, David W (David)) Date: Mon, 31 Aug 2009 17:02:14 -0600 Subject: Unablto preven re-order when using variable delay. Message-ID: <71C3797C8397694EA4F9FD754B0FAFFA01DA59FA@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> For my applicatioI need to providvariable delay, without reordering. Neteworked finuntil the enhancement allowing reorder of packets based on variable delay. https://lists.linux-foundation.org/pipermail/netem/2005-November/000865.html After thenhancemenwas put in to allow re-ordering (tfifo qdisc) I modified my system to changed the internal qdisc for netem from tfifo to pfifo. (http://www.linuxfoundation.org/en/Net:Netem#How_to_reorder_packets_based_on_jitter.3F) However, wheI upgraded to thFedora 11 distribution ( kernel 2.6.29.5-191.fc11.i686.PAE, iproute: 2.6.29-2.fc11.i586) I was no longer able to add the pfifo qdisc. My tesscript: /sbin/tc qdisc del dev eth2 roo2>&1 /sbin/tc qdisc add dev eth2 roohandl1: prio 2>&1 /sbin/tc qdisc add dev eth2 paren1:3 handl30: netem delay 1ms 2>&1 /sbin/tc qdisc add dev eth2 handl40: paren30:1 pfifo limit 16000 2>&1 /sbin/tc qdisc show dev eth2 2>& reports: RTNETLINK answers: Invalid argument qdisc prio 1: roobands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 qdisc nete30: paren1:3 limit 1000 delay 999u After working with thfolks aredhat [Bug 515522], I discover that netem has been changed in a way that it appears pfifo will not be supported. "sch_netem: Removclassful functionality" http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=02201464119334690fe209849843881b8e9cfa9f Has anyonelsexperienced this issue? Perhaps theris a differenway to set netem up to provide variable delay, without reorder; any suggestions are welcome. David David W. James | SolutioVerification - Automation Tea| Avaya | 1300 W. 120th Ave. | B2-B42 | Westminster, CO 80234 | Voice/Fax (303) 538 3986 | dwjames at avaya.com | -------------- nexpar-------------- AHTML attachmenwas scrubbed... URL: http://lists.linux-foundation.org/pipermail/netem/attachments/20090831/9d5ac70c/attachment.ht
