Ian Diddams | 21 Jul 18:04 2014

migrating MRTG

I’m moving MRTG from a solaris box to a linux (centos 6) server, running mrtg 2.17.4 as a fresh installation.


It’s all working fine BUT …  I would rather like the old history from the previous server to be migrated too, so that we get the full last-year-graph as well.


I’ve tested an idea for the cpu logs we have by concatenation the old file beneath the new file…  but while I can see from the epoch dates in it the logged data is “there” it doesn’t result in a yearly/monthly file etc.


What need I be doing to achieve this?







This message is intended for the addressee only. It may contain information of a confidential or legally privileged nature. If you have received this message in error please notify the sender and destroy the message immediately. All attachments have been scanned for viruses. However IS Solutions plc cannot accept liability for any loss or damage you may incur as a result of virus infection.
mrtg mailing list
mrtg <at> lists.oetiker.ch
Tyler, Rodney | 8 Jul 23:36 2014

WAN Speed

I have setup mrtg to monitor traffic for a specific Ethernet interface on a Cisco Catalyst 6509 that connects two campuses over a fiber WAN link.  The third party fiber provider says they are giving us 100Mbps across the fiber, but we need to verify what we are actually getting.  I  have come to realize that simply monitoring the WAN interface gives me a nice graph and actual kB/s (or MB/s) data passed, as shown below; however, the percentages are based on my Maxbytes value that is based on what the ISP says we are getting (100Mpbs or 12.5MB/s).  In other words, I can change the percentages by adjusting MaxBytes.  Is there any way to use mrtg to verify the available bandwidth over the WAN, or do I need a different tool?
Max Average Current In Out
776.4 kB/s (6.2%) 143.6 kB/s (1.1%) 89.3 kB/s (0.7%)
11.2 MB/s (89.5%) 542.1 kB/s (4.3%) 819.2 kB/s (6.6%)
Thanks for any assistance,
Rodney Tyler
mrtg mailing list
mrtg <at> lists.oetiker.ch
Petr Lázňovský | 28 Jun 13:51 2014

MRTG got undef from extrenal script

Periodically monitoring connection to many network devices for few years, using MRTG and
mrtg-ping-probe script. It does work fine for long, long time, but in last few days it shows huge
packetloss to all machines, but this is not happens in reality.


Monitoring machine load is relativelly normal


I see all 'undef' instead of values in debug output:

c:\>perl C:\monitoring\mrtg\bin\mrtg c:\monitoring\pingp\packetloss.cfg --debug=snpo
Daemonizing MRTG ...
Do Not close this window. Or MRTG will die
--snpo: run external perl.exe c:\monitoring\pingp\mrtg-ping-probe.pl -t 42 -l 1000 -p loss/loss cisticz.pankrac.czf
--snpo: External result:undef out:undef uptime:unknown name:unknown
--snpo: run external perl.exe c:\monitoring\pingp\mrtg-ping-probe.pl -t 42 -l 1000 -p loss/loss lazna.pankrac.czf
--snpo: External result:undef out:undef uptime:unknown name:unknown

But If I start script not from mrtg but manually from cmdline, than expected numbers are on stdout:

c:\monitoring\pingp>perl mrtg-ping-probe.pl -t 42 -l 1000 -p loss/loss lazna.pankrac.czf

c:\monitoring\pingp>perl mrtg-ping-probe.pl -t 42 -l 1000 -p loss/loss cisticz.pankrac.czf

This is perl, v5.10.1 built for MSWin32-x86-multi-thread

Perl version is not new, but until now I have no reason to upgrade it, everything worked fine. Anybody have
idea, what can be wrong?

NetLink | 3 Jun 20:56 2014

Problem with cfgmaker

New installation from scatch.
Windows Server 2012 R2
Mrtg 2.17.4
Perl 5.16.3

perl mrtg returns without error.
perl cfgmaker public <at> gives this output:

PS C:\Program Files (x86)\mrtg-2.17.4\bin> perl cfgmaker public <at>
--base: Get Device Info on public <at>
Undefined subroutine &SNMP_util::encode_oid called at C:/Program Files
(x86)/mrtg-2.17.4/bin\..\lib\mrtg2/SNMP_util.pm line 1346.

Any idea what is the problem?
Greg Cory | 19 May 18:41 2014

Looking for some advice on the gaps we get in MRTG graph at peak traffic

We have mrtg set up and running, it’s been working good till a few months ago as our traffic increased.


Now during the peak of our traffic the mrtg graphs get some gaps in them.  Usually ever night around 19:00 to 22:00.


If you can see this picture below this is what we are getting.


Any advice would be greatly appreciated.










Attachment (image001.jpg): application/octet-stream, 27 KiB
mrtg mailing list
mrtg <at> lists.oetiker.ch
Beluc | 25 Mar 12:23 2014

monitoring iptables


i have a server with some extra failover public interfaces where i'm sending some kind of traffic on a round robin way using iptables SNAT :

auto eth1:0
iface eth1:0 inet static

auto eth1:1
iface eth1:1 inet static

auto eth1:2
iface eth1:2 inet static

# iptables -L -n -t nat -v
Chain POSTROUTING (policy ACCEPT 10 packets, 685 bytes)
 pkts bytes target     prot opt in     out     source               destination         
  546 32760 SNAT       all  --  *      eth1       statistic mode nth every 3 to:
  546 32760 SNAT       all  --  *      eth1       statistic mode nth every 2 to:
  546 32760 SNAT       all  --  *      eth1       statistic mode nth every 1 to:
 614K   41M MASQUERADE  all  --  *      eth1

all this is working fine !

i'm already monitoring physical interface (eth1) of that server using mrtg/smp and now, i would like to monitore pkts/bytes passing through extra interfaces.
i searched a bit if iptables is reachable via snmp but cannot find anything.

does somebody know a way to achieve this ?

mrtg mailing list
mrtg <at> lists.oetiker.ch
Bob Reiber | 16 Mar 20:21 2014

can i change the 1 day graph to show three days

I like the 5 minute graph but want it to display 3 days worth of data. I have looked through the paramaters and don’t see anything that relates to the number of intervals that the 5 minute graph will display. Can someone point me in the right direction


Bob Reiber

BK Sales and Service

Tel: (650) 376-1122


mrtg mailing list
mrtg <at> lists.oetiker.ch
Steve Lindemann | 21 Feb 22:19 2014

MIB for Dell PowerConnect 3524?

Does anyone know where I can get the MIB for a Dell PowerConnect 3524 to 
use with MRTG?  I've been looking and haven't had any luck finding what 
I need.

In lieu of that, I'll take suggestions for monitoring thru put on the 
ports on that model switch.

Thanks all!

Steve Lindemann, MSIS                   __
Network Administrator                  //\\  ASCII Ribbon Campaign
Marmot Library Network                 \\//  against HTML/RTF email,
+1.970.242.3331 x116                   //\\  vCards & M$ attachments
Beluc | 20 Feb 14:43 2014

convert string to oid

Hi everybody,

i'm trying to apply smth that works using snmpwalk but doesn't in mrtg :

mrtg:/etc/mrtg# snmpwalk -OnE -c snmppub -v 2c\"CLUSTER1\"
. = Counter64: 428593
mrtg:/etc/mrtg# snmpwalk -OnE -c snmppub -v 2c\"CLUSTER1_PREDICTOR\"
. = Counter64: 3857472

Target[cluster1_test_failedconnections]:\"CLUSTER1\"&\"CLUSTER1_PREDICTOR\":snmppub <at>
Options[cluster1_test_failedconnections]: growright,noinfo
YLegend[cluster1_test_failedconnections]: failed connections
ShortLegend[cluster1_test_failedconnections]: conn
MaxBytes[cluster1_test_failedconnections]: 1000000
Title[cluster1_test_failedconnections]: Failed Connections -- cluster1
PageTop[cluster1_test_failedconnections]: <h1>Failed Connections --cluster1 </h1>
LegendI[cluster1_test_failedconnections]: &nbsp;Predictor:
LegendO[cluster1_test_failedconnections]: &nbsp;Predicted:

mrtg:/etc/mrtg# if [ -x /usr/bin/mrtg ] && [ -r /etc/mrtg/mrtg.cfg-ServerFarmFailedConns55 ]; then env LANG=C /usr/bin/mrtg /etc/mrtg/mrtg.cfg-ServerFarmFailedConns55 2>&1 | tee -a /var/log/mrtg/mrtg.log-ServerFarmFailedConns55 ; fi
2014-02-19 18:36:49: WARNING: Could not match host:'snmppub <at>' ref:'Descr' key:'"CLUSTER1_PREDICTOR\"'
2014-02-19 18:36:51: ERROR: Target[cluster1_test_failedconnections][_IN_] '\"CLUSTER1\"& $target->[12]{$mode} ' (kill): Can't find string terminator '"' anywhere before EOF at (eval 44) line 1.
2014-02-19 18:36:51: ERROR: Target[cluster1_test_failedconnections][_OUT_] '\"CLUSTER1\"& $target->[12]{$mode} ' (kill): Can't find string terminator '"' anywhere before EOF at (eval 45) line 1.

I hope someone will have an idea.

CLUSTER1 & CLUSTER1_PREDICTOR are serverfarms from cisco ACE30


mrtg mailing list
mrtg <at> lists.oetiker.ch
Alessandro Forghieri | 5 Feb 17:00 2014

mrtg-2.14.5-4 (Centos5): oid-mib-cache.txt not created (w/solution)

This may be an old bug - hard to say with the trac access sealed in the web site - but it's happened to me plenty and recently, and the 'net is quiet about it.
Hopefully this'll help some poor schmuck

I have seen that in the above (subject) mentioned conditions it may happen that a
clean configuration file starts acting up by spewing 
Unknown SNMP Variable errors. Others have met this as well.: see https://lists.oetiker.ch/pipermail/mrtg/2011-March/036481.html) As explained in the link, the apparent cause of the error is a missing /var/lib/mrtg/oid-mib-cache.txt file that should be automatically created and populated with name-to-oid translations loaded from the LoadMib specified files (but isn't). One unpleasant solution is creating the file by hand. The root cause I found for this is as follows. The mrtg distro apparently ships with two similarly named files: Net_SNMP_util.pm and SNMP_util.pm They are different but contain similarly named functions and it turns out that MRTG mixes up calls between the two. In this particular case, when parsing the cfg file MRTG_lib.pm calls (line 681): snmpQueue_MIB_File($mibFile) stepping through in the debugger, it is possible to see that this is

the name-to-oid translation, however, is made through SNMP_utils::toOID function. This does not know about the MIB files (they have been queued in the wrong module, above) and so it breaks. Replacing; snmpQueue_MIB_File($mibFile) with SNMP_util::snmpQueue_MIB_File($mibFile) Makes all the error go away. The reason of the mixup or of the presence of two SNMP libraries, I cannot fathom. I'd bet that one OR the other should be present and that shipping both causes the snafu, but that's me. Maybe removing one or the other fixes everything as well - or it may make it worse - I did not try that. Cheers, alf
-- -- Alessandro Forghieri Nouvelle s.r.l. via Giardini 456/C 41124 Modena (ITALY) tel. +39 059 2929338 - Fax +39 059 347606
Attachment (alf.vcf): text/x-vcard, 257 bytes
mrtg mailing list
mrtg <at> lists.oetiker.ch
Henri Stofberg | 7 Jan 12:17 2014


Hi guys.

I used to do a lot of LINUX setup and support, but it was about 5yrs 
since the last time.

So please don't shoot me, but I have forgotten o lot of it.

I now have a UBUNTU based laptop, but I want to run MRTG to monitor the 
traffic through the router.

Previously I think I did the setup via WEBMIN, but I am not to sure.

1. Can I set it up via WEBMIN?
2. Is there some friendlier interface that I can try and use?


Henri Stofberg