On 12/22/2013, 3:20 AM, "Pekka Pietikäinen" <p...@ee.oulu.fi> wrote:
>On 22/12/13 02:22, Fred Klassen wrote: >> ough of that. Time to try to show you what I have been working on. I >> am happy to make a few announcements: >> >> Firstly we are working on a new wiki http://tcpreplay.appneta.com >> <http://tcpreplay.appneta.com/>. This is a work-in-progress site and >> we are not simply doing copy/paste from the old site. We expect the >> content to be fully updated in the next 3-4 weeks. >> >> Secondly, the GitHub source code repository has now moved to >> https://github.com/appneta/tcpreplay. If your not sure how to >> download, please visit our new wiki. >> >> Finally, I am excited to announce our first software release, version >> 4.0.0 Beta 1. It brings together several features from the existing >> 4.0.0 Alpha 1 branch, but also contains several of our expanded >> capabilities and increased performance, including: >> >> > >Hi and thanks for stepping up and revitalizing this awesome project! > >Got a checkout done and was able to build on Fedora 20 (no such luck >with the previous 4.0 alphas), Great. I have not tested with Fedora. I have tested with Ubuntu 10.04. > >yum install autogen-libopts-devel-5.18-1.fc20.x86_64 >./configure --enable-dynamic-link --disable-local-libopts Unfortunately I didn't discover --disable-local-libopts until I read Aaron's release notes. I built the final tarball for 4.0.0beta1 but didn't do regression testing on OS X. It appears I broke it, so there will be a 4.0.0beta2 coming out soon. > >+ a build fix (remains of didsig in the code which had been refactored) >did the trick, pull request sent Saw the pull request. Tested and accepted. Will be in 4.0.0beta2. > >Some notes: > >Some old annoyances are still around ( >https://github.com/synfinatic/tcpreplay/pull/9 ), now trying to figure >out how they apply to the new code base and fix them :-). They're quite >easy to run into, run tcpreplay >with --loop 0 with really small (1-10 packets is typical) pcaps. I'll try to look into it. If you find a fix, make sure you put the pull request on the new root repo https://github/appneta/tcpreplay. It is much easier to Handle the pull request. > >Used to run like: >tcpreplay -K --stats 1 -i virbr0 -M 50 --loop 0 out.pcap > >where ^C didn't actually exit. Fixed in pull request. Great! I fixed a few things around ^C including printing full report rather than just the summary. Also report will print out even if only a few packets were sent. > >The rest still investigating how they apply. >I'm running tcpreplay in a small wrapper written in nodejs (simplified >option set and supports start from 100Mbps and slow down to 0 Mbps, then >exits + dump stats out in JSON), hence the fflush after printing stats. I believe the lowest setting that works is --mbps=0.001 and --mbps=0 is equivalent to -t option. > >Noticed -K is missing, replaced with --preload-pcap... High cpu use >despite low packet rate, will investigate. -K was before my time. Also, CPU was high when I took over. I agree with the current methodology of spinning and sleeping. Accuracy is way higher and as long you sleep in the main loop, the system remains responsive. Most applications would not notice the overhead, but it has bad optics. > > > > > > > >-------------------------------------------------------------------------- >---- >Rapidly troubleshoot problems before they affect your business. Most IT >organizations don't have a clear picture of how application performance >affects their revenue. With AppDynamics, you get 100% visibility into >your >Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics >Pro! >http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktr >k >_______________________________________________ >Tcpreplay-users mailing list >Tcpreplay-users@lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/tcpreplay-users >Support Information: http://tcpreplay.synfin.net/trac/wiki/Support ------------------------------------------------------------------------------ Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk _______________________________________________ Tcpreplay-users mailing list Tcpreplay-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/tcpreplay-users Support Information: http://tcpreplay.synfin.net/trac/wiki/Support
