[collectd] Collectd on RedHat with virtuall network devices !
Michael Theurl
michael.theurl at smog.at
Sun Mar 12 01:56:15 CET 2006
Hy Florian,
Thanxs for this fast feedback :)
my /proc/net/dev look like this:
< - snip - >
Inter-| Receive |
Transmit
face |bytes packets errs drop fifo frame compressed multicast|bytes
packets errs drop fifo colls carrier compressed
lo: 71546 644 0 0 0 0 0 0
71546 644 0 0 0 0 0 0
eth0: 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
eth1:3507098520 3316184 0 0 0 0 0 45269
693998005 2489918 0 0 0 0 0 0
sit0: 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
vmnet2: 0 87639 0 0 0 0 0 0
0 151506 0 0 0 0 0 0
vmnet1: 0 25957 0 0 0 0 0 0
0 55353 0 0 0 0 0 0
tap0:2190178848 1600016 0 0 0 0 0 0
50846286 727736 0 0 0 0 0 0
< - snap - >
and the rrdtool fetch for the vmnet1 looks this:
< - snip - >
1142124000: 0.0000000000e+00 0.0000000000e+00
1142124060: 0.0000000000e+00 0.0000000000e+00
1142124120: 0.0000000000e+00 0.0000000000e+00
1142124180: 0.0000000000e+00 0.0000000000e+00
1142124240: 0.0000000000e+00 0.0000000000e+00
1142124300: 0.0000000000e+00 0.0000000000e+00
1142124360: 0.0000000000e+00 0.0000000000e+00
1142124420: 0.0000000000e+00 0.0000000000e+00
1142124480: 0.0000000000e+00 0.0000000000e+00
1142124540: nan nan
< - snap - >
You right the rrd database works. I see the "dev" entry for the virtual
devices look strange !
but the tap0 my looks good in the "dev" entry but in the rrd database
there are only this 0.000 entrys !!
Any idea ?
cu
Michael
On Sun, 2006-03-12 at 01:38 +0100, Florian Forster wrote:
> Hello Michael,
>
> On Sat, Mar 11, 2006 at 08:36:14PM +0100, Michael Theurl wrote:
> > The main problem is the only rrd file for the eth1 interface have
> > collected the bits !
> > the other rrd are created but the rrd database is zero. I use the
> > "traffic" file from the contrib folder.
> >
> > As a sample i checked the database state with "rrdtool info
> > /var/lib/collectd/traffic-tap0.rrd"
>
> that displays information about the database `layout' only. To check the
> values stored in this layout use something like this:
> -- 8< --
> octo at alice:~/build $ rrdtool fetch /var/lib/collectd/leeloo.lan.home.verplant.org/traffic-eth0.rrd AVERAGE -s -300
> timestamp incoming outgoing
>
> 1142123100: 1.9956200000e+03 2.0372000000e+03
> 1142123160: 1.7583166667e+03 1.0541666667e+03
> 1142123220: 2.0173090909e+03 1.1511803030e+03
> 1142123280: 5.8442757576e+02 6.6627136364e+02
> 1142123340: 2.1038875758e+03 7.8401954545e+02
> 1142123400: 8.6518242424e+02 6.5879878788e+02
> 1142123460: 5.9996000000e+02 7.5366666667e+02
> -- >8 --
>
> If many `NaN's (Not a Number) are displayed, somethings is going wrong.
> If there are just many very small values, it's likely working..
>
> > is it Possible that the collectd have some problems with virtual
> > network devices ?
>
> collectd simply reads and interprets the content of </proc/net/dev>
> (when running on Linux, that is). Take a look at that file and check if
> the interfaces you're interested in are listed here. If the interface is
> a virtual interface or a physical one doesn't matter, as long as it's
> listed here (and doesn't contain a colon..).
>
> > here my collectd.conf:
>
> The config looks okay..
>
> > Is there a sollution for this problem ?
>
> Check the two points above and send what you found out, and there'll be
> a good chance we'll get this sorted out ;)
>
> Regards,
> -octo
> _______________________________________________
> Collectd mailing list
> Collectd at verplant.org
> http://mailman.verplant.org/listinfo/collectd
--
[] this Email is made of 100% Recyclable elektrons
[] visit : www.smog.at ; www.theurl.at ; www.lugv.at
[] private: http://members.smog.at/michael
[] mailto : michael.theurl at smog.at
[] key : www.smog.at/key
More information about the Collectd
mailing list