Cheshire Cat Computing

This is probably a MRTG/RRD question but i thought i would try here first.

I had a router go offline for a couple of minutes and now i have been getting these error messages in my mail account and the graphs are not updating. and if i try to create a new cfg file with buildwan.pl or cfgmaker_host i get "unable to connect to SNMP Agent" I can ping and telnet into the router without any problems. I also have another server just running mrtg and it is updating the graphs and no error messages. both are RH7.3

Any help in trying to re-initialize the graph and/or mrtg without losing any history would be grand.


Error Messages:

Exiting subroutine via last at /usr/local/mrtg-2/bin/../lib/mrtg2/Net_SNMP_util.pm line 1780.
Exiting subroutine via last at /usr/local/mrtg-2/bin/../lib/mrtg2/Net_SNMP_util.pm line 1780.
Exiting subroutine via last at /usr/local/mrtg-2/bin/../lib/mrtg2/Net_SNMP_util.pm line 1780.
Exiting subroutine via last at /usr/local/mrtg-2/bin/../lib/mrtg2/Net_SNMP_util.pm line 1780.
Exiting subroutine via last at /usr/local/mrtg-2/bin/../lib/mrtg2/Net_SNMP_util.pm line 1780.
Use of uninitialized value in concatenation (.) or string at /usr/local/mrtg-2/bin/mrtg line 2004.
Use of uninitialized value in concatenation (.) or string at /usr/local/mrtg-2/bin/mrtg line 2004.
Use of uninitialized value in string ne at /usr/local/mrtg-2/bin/mrtg line 2082.

AND

Wednesday, 9 August 2006 at 8:18: WARNING: skipping because at least the query for ifInOctets.1 on 192.168.x.x did not succeed
Wednesday, 9 August 2006 at 8:18: WARNING: no data for ifInOctets&ifOutOctets:public@192.168.x.x. Skipping further queries for Host 192.168.x.x in this round.
Wednesday, 9 August 2006 at 8:20: ERROR: Target[192.168.x.x.serial0.18][_IN_] ' $target->[20]{$mode} ' did not eval into defined data
Wednesday, 9 August 2006 at 8:20: ERROR: Target[192.168.x.x.serial0.18][_OUT_] ' $target->[20]{$mode} ' did not eval into defined data


Jeff

This means (as it says) it cannot contact the SNMP agent. The fact that you can telnet or ping makes no difference, it's the SNMP agent you need. Are you absolutely certain that the SNMP agent is running on this server, and you have the correct community string defined? Try using snmpwalk to check that you can retrieve data from the remote device.

_________________
Steve Shipway
UNIX Systems, ITSS, University of Auckland, NZ
Woe unto them that rise up early in the morning... -- Isaiah 5:11

Thanks for the quick reply,

I am pretty sure that it is the SNMP agent on the server that is causing this.

The server gving me the error is running "UCD-snmp 4.2.4" and the server not returning any errors is using "NET-SNMP version: 5.1.1.rc1".

the only thing that i have not done yet is to reboot the server. I am having a hard time finding info on UCD-snmp - rightly so, since it is outdated - to figure out what is happening. are there any tricks or tips that might help?

Here are the results from snmpwalk on the two different servers:

working server:
~ >snmpwalk -v 1 -c public 192.168.x.x system
SNMPv2-MIB::sysDescr.0 = STRING: Cisco Internetwork Operating System Software
IOS (tm) C1700 Software (C1700-SV3Y-M), Version 12.2(4.2), MAINTENANCE INTERIM SOFTWARE
Copyright (c) 1986-2001 by cisco Systems, Inc.
Compiled Tue 21-Aug-01 07:39 by pwade
SNMPv2-MIB::sysObjectID.0 = OID: SNMPv2-SMI::enterprises.9.1.216
SNMPv2-MIB::sysUpTime.0 = Timeticks: (3885125) 10:47:31.25
SNMPv2-MIB::sysContact.0 = STRING:
SNMPv2-MIB::sysName.0 = STRING: Francis
SNMPv2-MIB::sysLocation.0 = STRING:
SNMPv2-MIB::sysServices.0 = INTEGER: 78
SNMPv2-MIB::sysORLastChange.0 = Timeticks: (0) 0:00:00.00

non working server:

[root@im root]# snmpwalk -v 1 -c public 192.168.x.x system
Timeout: No Response from 192.168.x.x


jeff

UCD SNMP is the old name for Net SNMP. I think.

The output you're getting from the non-working server would indicate that either:
1) you have the wrong IP address
2) you have the wrong community string
3) You have a firewall in the way
4) the snmp access control doesnt permit your client address
5) snmpd is not actually running
6) the intervening network is massively slow or overloaded and is dropping packets
7) snmpd is running but has a different problem - look in the syslog for error messages

_________________
Steve Shipway
UNIX Systems, ITSS, University of Auckland, NZ
Woe unto them that rise up early in the morning... -- Isaiah 5:11

I just rebooted the server and it is working now. Don't know why snmpwalk was hanging up on that one router, but restarting did the trick.

Thanks for replying to my questions.

FYI - no need to reply - unless you have any other ideas.

i just had another serial link go down on another router and the same thing is happening with this one.


By the way routers2 is fantastic.

Thanks for a great product.

Probably you are polling the device via the wrong interface.

For example, if your router has 2 interfaces, one facing you and one the serial, you need to poll the one closest to you. This may seem obvious, but sometimes people use the DNS name for the router, and it points to the IP of the far interface. Then, when the serial line goes down (ie, the remote interface) it cannot poll the router, even though the near interface will respond fine.

This may not be the problem, though

_________________
Steve Shipway
UNIX Systems, ITSS, University of Auckland, NZ
Woe unto them that rise up early in the morning... -- Isaiah 5:11