Date: Mon, 19 Oct 1998 16:56:49 +0200 (MEST) From: Werner Griessl <croot@btp1da.phy.uni-bayreuth.de> To: "John W. DeBoskey" <jwd@unx.sas.com> Cc: freebsd-current@FreeBSD.ORG, dg@root.com Subject: Re: Conflicting ping performance Message-ID: <XFMail.981019165649.croot@btp1da.phy.uni-bayreuth.de> In-Reply-To: <199810191438.KAA23348@bb01f39.unx.sas.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 19-Oct-98 John W. DeBoskey wrote: > Hi, > > Are you telling me it 'should' work this way now? Or that this > behaviour is a bug and was introduced in 1.40? It would appear > that the -i option is adversely affecting the -f option, and, yes, > I am running this test as root so -c should allow infinity. > > If it's a bug, I'll look into a submitting a diff... > > Thanks, > John > No, I believe it means: use "time ping -c 500 -i 0 localhost > /dev/null" as root instead. Werner > >> > We're seeing some very interesting ping timings. Wall-clock vs. >> >recorded ping statistics. >> > >> > In the statistics below, the -current system is recording ping >> >statistics 1/2 the values of the march snap. However, the march >> >snap completes in 0.05 seconds, while the -current system requires >> >10 seconds. A system running the 980825-SNAP gives the same strange >> >results as the -current system. >> > >> > The network controllers should not have any effect since we are >> >testing against localhost. (however, the two machines below are >> >identical hardware configs using fxp0). >> > >> > This problem replicates when testing the actual network devices. >> > >> > Can anyone else replicate this problem? >> > >> > Comments, Critiques, or diffs to fix this are appreciated. >> >> Caused by this: >> >> RCS file: /home/ncvs/src/sbin/ping/ping.c,v >> ... >> ---------------------------- >> revision 1.41 >> date: 1998/08/26 18:51:37; author: des; state: Exp; lines: +1 -13 >> Remove -c restrictions from previous commit. >> ---------------------------- >> revision 1.40 >> date: 1998/08/26 01:58:39; author: dillon; state: Exp; lines: +38 -11 >> (well tested at BEST): -i option can now take FP values (e.g. -i 0.1), >> extremely useful for networking testing. Other options secured from >> user-level D.O.S. attacks. -f, -s now root-only. -i wait times < 1.0 >> root-only. -c count limited to 100 and defaults to 16 when ping run >> by non-root user. >> >> >> -DG >> >> David Greenman >> Co-founder/Principal Architect, The FreeBSD Project >> > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-current" in the body of the message ---------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.981019165649.croot>