From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Erik K." To: documentation@lists.ipfire.org Subject: Re: IPTraf-ng problem with version 1.1.4 Date: Thu, 29 Aug 2013 10:31:16 +0200 Message-ID: <08D07391-C82C-4E50-93D2-807D084E17AF@ipfire.org> In-Reply-To: <0D1AF94E-D44E-4DDC-A6C0-D2471ABBF43D@ipfire.org> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="===============1992142960792491601==" List-Id: --===============1992142960792491601== Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable The patch seems to solve it.=20 Fwd: [PATCH] BUGFIX: fix "Floating point exception" in tcplog_flowrate_msg() > make sure interval is at least 1 (bug was made in > "tcplog_flowrate_msg(): cleanup and fix") >=20 > Reported-by: Erik K. > Signed-off-by: Vitezslav Samel > --- > src/tcptable.c | 2 ++ > 1 files changed, 2 insertions(+), 0 deletions(-) >=20 > diff --git a/src/tcptable.c b/src/tcptable.c > index a4133d9..e217b19 100644 > --- a/src/tcptable.c > +++ b/src/tcptable.c > @@ -437,6 +437,8 @@ static char *tcplog_flowrate_msg(struct tcptableent *en= try, char *buf, > size_t bufsize) > { > time_t interval =3D time(NULL) - entry->conn_starttime; > + if (interval < 1) > + interval =3D 1; >=20 > char rbuf[64]; > rate_print(entry->bcount / interval, rbuf, sizeof(rbuf)); > --=20 > 1.7.8.4 i will go for some further tests and bring it also to IPFires bugzilla for ev= entual news and/or other fixes. Greetings=20 Erik Am 28.08.2013 um 20:03 schrieb Erik K.: > Hi Michael, >=20 > Am 28.08.2013 um 13:26 schrieb Michael Tremer: >=20 >> Hi Erik, >>=20 >> On Wed, 2013-08-28 at 13:13 +0200, Erik K. wrote: >>> Hi all, >>> i want to report a problem with the unreleased, updated version of IPTraf= -ng-1.1.4 . The problem occurs if i turn on the IPTraf logging, after a while= i get an "Floating point exception" under the "Packets catured:" line and IP= Traf-ng quits. Until now i couldn=C2=B4t localize the problem, but it seems l= ike a bug in the new version.=20 >>> The IPTraf-ng version 1.1.3 works without this problem. >>=20 >> This is most certainly a division by zero. If you do that, the floating >> point unit of the CPU throws a floating point exception. >> It is best to report that to the iptraf-ng developers and open a bug >> report about this on our own bugtracker as well, so that we can update >> as soon as a fix is available. >=20 > i have sent an mail to IPTraf mailinglist and i=C2=B4am currently compile t= he patched iptraf-ng source with some fixes from the community. Hope this sol= ves the problem, will let you know this after some testing results. >=20 >>=20 >>> Beside info: >>> I have added an IPTraf-ng configuration for the logrotate.conf. In here -= -> http://git.ipfire.org/?p=3Dpeople/ummeegge/ipfire-2.x.git;a=3Dcommit;h=3D1= 6bcdaebca3586fda824e03822f82eecb796dc65 the diff can be found. >>=20 >> Should this be merged? Why is iptraf-ng writing log files any way? >=20 > I think version 1.1.3 is also pretty much up to date (on my Fedora 16 syste= m version 1.1.1 works) and a good possibility to use IPTraf-ng nevertheless i= n IPFire, i don=C2=B4t know how much time it takes that the above mentioned p= atch will be provided in the official source code and also if it solves the p= roblem. >=20 > The logging of IPTraf-ng is very nice i think, especially with the ability = of the filter usage the output can be very clear and applicable for different= needs also easy to use for further processing with shell scripting. The conf= iguration window allows to activate or deactivate the logging, so i thought i= t might be a good idea to add it also in logrotate.conf. >=20 >>=20 >> -Michael >>=20 >> _______________________________________________ >> Development mailing list >> Development(a)lists.ipfire.org >> http://lists.ipfire.org/mailman/listinfo/development >=20 --===============1992142960792491601==--