Date: Thu, 8 Dec 2005 01:20:07 -0800 (PST) From: Jeremy Chadwick <freebsd@jdc.parodius.com> To: FreeBSD-gnats-submit@FreeBSD.org Subject: bin/90098: traceroute "pretty printing" of multipath routes is broken Message-ID: <20051208092007.9386A5FD0@mx1.parodius.com> Resent-Message-ID: <200512080930.jB89U32H094547@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 90098 >Category: bin >Synopsis: traceroute "pretty printing" of multipath routes is broken >Confidential: no >Severity: non-critical >Priority: medium >Responsible: freebsd-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Thu Dec 08 09:30:02 GMT 2005 >Closed-Date: >Last-Modified: >Originator: Jeremy Chadwick >Release: FreeBSD 6.0-STABLE i386 >Organization: Parodius Networking >Environment: FreeBSD icarus.home.lan 6.0-STABLE FreeBSD 6.0-STABLE #0: Fri Dec 2 02:22:17 PST 2005 root@icarus.home.lan:/usr/obj/usr/src/sys/ICARUS i386 >Description: This problem applies (presently) to HEAD and RELENG_6* only. CVS commit 1.27 (and 1.26.8.1 for RELENG_6_*) breaks traceroute output in numerous circumstances. The code change in question: http://www.freebsd.org/cgi/cvsweb.cgi/src/contrib/traceroute/traceroute.c.diff?r1=1.26&r2=1.27&f=h The breakage: icarus# /usr/sbin/traceroute gw traceroute to gw.home.lan (192.168.1.1), 64 hops max, 40 byte packets 1 * gw (192.168.1.1) 0.762 ms 0.784 ms What you see above is NOT a copy/paste error, nor a formatting error. It's the result of the commit. After the aforementioned "pretty printing" is removed, proper formatting is restored: icarus# ./traceroute gw traceroute to gw.home.lan (192.168.1.1), 64 hops max, 40 byte packets 1 * gw (192.168.1.1) 0.853 ms 0.786 ms Apologies for the criticism, but stuff like this really pisses me off. No one tested this thoroughly; it never should have made it past HEAD. For weeks I've been bashing my head thinking this was a terminal problem, some sort-of IFS or globbing problem with /bin/sh, or a gcc optimisation bug -- only to find the above commit... Please back this out of all trees promptly, especially RELENG_6*. >How-To-Repeat: Run a traceroute to a host that doesn't respond to packets during the first probe attempt. >Fix: Back the patch out and have the author (ras@e-gerbil.net) take responsibility for fixing this properly. >Release-Note: >Audit-Trail: >Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20051208092007.9386A5FD0>