Notifying Nagios that a Node is down/disabled

Hi Timo,

 

 

Have you tried configuring LTM to send SNMP traps to the Nagios server? You can then customize the SNMP traps to ensure a message is sent when a pool member or node is disabled or marked down.

 

 

Here are some related solutions:

 

 

SOL6414: Determining which alerts are pre-configured to trigger an SNMP trap

 

https://support.f5.com/kb/en-us/solutions/public/6000/400/sol6414.html

 

 

SOL3727: Configuring custom SNMP traps

 

https://support.f5.com/kb/en-us/solutions/public/3000/700/sol3727.html

 

 

Aaron

Hi Aaron,

thank you very much.

From what i understand this is how to send traps directly to Nagios, right?

This is the way I wanted to go first.

But now i played with the check_snmp-Nagios Plugin and this is what i got so far:

nagios@mynagiosserver:~/libexec> ./check_snmp -H mybigipf5server -o 1.3.6.1.4.1.3375.2.1.4.2.0 -C public -P 2c 
 SNMP OK - "9.4.5" |

I can check the Version-String of my BigIP LTM using the Nagios-Check-Plugin. The only thing i need to know is the OID for Node-Status.

Is there some kind of "list" in which i can look for all available OIDs offered by BigIP?

Using the SNMP-Script directly on my nagios-server i dont need to write a custom trap-notification on the LTM and everything fits much better in my nagios-environment.

So far, I try to keep the monitoring-configuration on the nagios-server-side. That way all my servers are simple/stupid "clients" on which nagios performs several checks.

So, the question is: Is there a simple OID for getting the Node-Status?

Thats all i need to know so far.

Thank you!

Bye,

Timo

Well,

 

 

i think i get this wrong.

 

Is it even possible to get the information i need through this "passive" SNMP-Check instead of using SNMP-Traps?

 

 

From what I understand the LTM sends an SNMP-Trap whenever the state of a node changes, is that right?

 

 

I wonder if I can get this information using passive checks like the one that nagios performs.

 

 

Maybe someone can help me.

 

 

Timo

By default, no SNMP traps are sent from LTM. You'd need to enable SNMP traps to an SNMP monitoring station.

 

 

If you want to use Nagios to perform SNMP polling for the pool states, I think you could have it check under .1.3.6.1.4.1.3375.2.2.5.1.2.1. You can download the MIBs from the Overview page of the GUI. You can view the F5 MIBs using a standard MIB browser to find the exact OID's you want to check.

 

 

Aaron

Hi,

thanks again.

Using a MIB-Browser i discovered that i need to check OID 1.3.6.1.4.1.3375.2.4.0.12 using my SNMP-Check. The OID name is "bigipNodeDown". I think thats it.

Unfortunately, trying this command

nagios@mynagiosserver:~/libexec> ./check_snmp -H mybigipserver -o 1.3.6.1.4.1.3375.2.4.0.12 -C public -P 2c

ends in:

SNMP OK - No Such Object available on this agent at this OID |

Do i need to "allow" the Nagios-Server to get the information?

Thanks!

Bye

Timo

You can configure SNMP polling access under System >> SNMP >> Agent. You can also try testing locally first to make sure you get the expected results and then test from the Nagios host.

 

 

Aaron

Hi,

i'm getting closer:

[root@mybigipserver:Active] ~  snmpget -v 2c -c public localhost 1.3.6.1.4.1.3375.2.4.0.12 
 F5-BIGIP-COMMON-MIB::bigipNodeDown = No Such Object available on this agent at this OID 
 [root@mybigipserver:Active] ~  snmpget -v 2c -c public localhost 1.3.6.1.4.1.3375.2.4.0.4 
 F5-BIGIP-COMMON-MIB::bigipCpuTempHigh = No Such Object available on this agent at this OID

I tried theses two commands to get the NodeDown and CpuTempHigh Values. Both dont work. Whats interesting is that the MIBs are getting called correctly (F5-BIGIP-COMMON-MIB::bigipNodeDown and F5-BIGIP-COMMON-MIB::bigipCpuTempHigh).

Why dont they return any values?

For example this works (local and throug nagios as well):

[root@mybigipserver:Active] ~  snmpget -v 2c -c public localhost 1.3.6.1.4.1.3375.2.1.4.2.0 
 F5-BIGIP-SYSTEM-MIB::sysProductVersion.0 = STRING: 9.4.5 
 

Any suggestions?

Timo

Hi,

i dont get it.

i tried the following three commands:

[root@mybigipserver:Active] ~  snmpget -v 2c -c public localhost 1.3.6.1.4.1.3375.2.1.1.1.1.20.0 
 F5-BIGIP-SYSTEM-MIB::sysAttrFailoverUnitId.0 = INTEGER: 2 
 [root@mybigipserver:Active] ~  snmpget -v 2c -c public localhost 1.3.6.1.4.1.3375.2.1.7.2.1.0 
 F5-BIGIP-SYSTEM-MIB::sysHostCpuNumber.0 = INTEGER: 0 
 [root@mybigipserver:Active] ~  snmpget -v 2c -c public localhost 1.3.6.1.4.1.3375.2.1.7.3.1.0 
 F5-BIGIP-SYSTEM-MIB::sysHostDiskNumber.0 = INTEGER: 6

These commands, all of them contained in "F5-BIGIP-SYSTEM-MIB" do work. I get the values locally and through nagios as well.

There seems to be a problem with the other MIBs.

Is there a way to "enable" the missing OIDs?

Thank you!

Timo

Dear Fellows,

 

Please do help us in our study. It is very important for us to have feedback from you (who is using Nagios). I bet I would not take you more than 5 minutes, all questions are multiple choice and on a single page. We will be highly thankful to all the participants.

 

Please click on following link to answer this survey.

 

http://www.kwiksurveys.com/online-survey.php?surveyID=KBIOLF_a8c769c3

 

 

Regards

 

I get to see my status + name, with large .oid, but, I'd really like a script to ask with pool name, and to get status:

[root@bip:Active:Changes Pending] config  snmpget -v 2c -c public localhost .1.3.6.1.4.1.3375.2.2.4.3.2.1.7.18.47.86.76.65.78.49.48.51.47.118.109.100.101.98.117.103.48.50
F5-BIGIP-LOCAL-MIB::ltmNodeAddrStatusName."/VLAN103/vmdebug02" = STRING: /part1/node1
[root@bip:Active:Changes Pending] config  snmpget -v 2c -c public localhost .1.3.6.1.4.1.3375.2.2.4.3.2.1.4.18.47.86.76.65.78.49.48.51.47.118.109.100.101.98.117.103.48.50
F5-BIGIP-LOCAL-MIB::ltmNodeAddrStatusEnabledState."/part1/node1" = INTEGER: enabled(1)