This is part of the Availability Report not the State History, but I see it is 100% for the period. Can you see that a trap was in fact sent from the PDU if you look in /var/log/snmp/snmptt.logjbennett wrote:Ok - from the report:scottwilkerson wrote:State History is in Nagios XI under Reports -> State History
The traps never show down.Code: Select all
Host Data Host UP Down Unreachable DNT-MLP2-8-PDU-2 93.485% 6.515% 0% Service Data Host Service Ok Warning Unknown Critical DNT-MLP2-8-PDU-2 PDU Input Frequency 93.101% 0% 6.899% 0% PDU Input Voltage 93.088% 0% 6.912% 0% PDU Input Voltage 2 93.124% 0% 6.876% 0% PDU Output Voltage 93.101% 0% 6.899% 0% SNMP Traps - PDU 100% 0% 0% 0% Average 94.483% 0.000% 5.517% 0.000%
Help with SNMP traps?
-
scottwilkerson
- DevOps Engineer
- Posts: 19396
- Joined: Tue Nov 15, 2011 3:11 pm
- Location: Nagios Enterprises
- Contact:
Re: Help with SNMP traps?
Re: Help with SNMP traps?
There is nothing in /var/log/snmptt/snmptt.log on the date in question and nothing in snmpttunknown.log for the date in question either.
Re: Help with SNMP traps?
For the record, changing the snmptrapd to call snmptt and snmptthandler base upon how you have SNMPTT running. If you in the snmptt.ini you have
You should be using snmptt, if its daemon, it should be snmptthandler. This will be a game breaker if it isn't set correctly. Also, with the debug on, do you see snmptrapd messages showing up in the /var/log/messages when traps are sent?
Code: Select all
mode = standalone
Nicholas Scott
Former Nagios employee
Former Nagios employee
Re: Help with SNMP traps?
I just came across something rather interesting.
While trying to get NRDS working, I went to the Unconfigured Objects page. Once there, I saw a few IP addresses that were for some of our UPSs and PDUs. I went ahead and ran them through the unconfigured objects wizard since they were showing traps as one of the services.
Lo and behold, I'm now getting traps for these 4 items.
What's different?
Here's an example of what I'm getting for one location that DOES have an issue to report:
When I search for the UPS by the host name as I had it originally configured:
When I check the host under the IP address that was in the Unconfigured Objects wizard, the specific host in question is going back and forth between these two traps:
and
Which makes sense when I check the UPS and see the following:
While trying to get NRDS working, I went to the Unconfigured Objects page. Once there, I saw a few IP addresses that were for some of our UPSs and PDUs. I went ahead and ran them through the unconfigured objects wizard since they were showing traps as one of the services.
Lo and behold, I'm now getting traps for these 4 items.
What's different?
Here's an example of what I'm getting for one location that DOES have an issue to report:
When I search for the UPS by the host name as I had it originally configured:
Code: Select all
SNMP Traps - UPS Ok 31d 3h 52m 36s 1/5 2013-04-01 21:25:30 OK: TRAP RESET
UPS Battery Status Check Critical 2h 16m 56s 10/10 2013-04-01 21:15:15 SNMP CRITICAL - *3*Code: Select all
xxx.xxx.xxx.xxx SNMP Traps Warning 1h 0m 22s 1/1 2013-04-01 21:30:13 SEVERE: The UPS batteries are low and will soon be exhausted. The UPS batteries are low and will soon be exhausted / enterprises.2468.1.2.1.2 ():The UPS batteries are low and will soon be exhaustedCode: Select all
xxx.xxx.xxx.xxx SNMP Traps Ok 1h 0m 55s 1/1 2013-04-01 21:29:08 INFORMATION: The UPS has return from a low battery condition. The UPS has return from a low battery condition / enterprises.2468.1.2.1.2 ():The UPS has return from a low battery conditionCode: Select all
Battery Capacity Remaining (%) 0I think you might be onto something here. My system was set to standalone. Newbie question: I'm not sure how to check how I have snmptt running? I'm not sure if I have it set-up as a daemon or not. I just followed the documentation for the automated install provided here: http://assets.nagios.com/downloads/nagi ... ith_XI.pdfnscott wrote:For the record, changing the snmptrapd to call snmptt and snmptthandler base upon how you have SNMPTT running. If you in the snmptt.ini you have
You should be using snmptt, if its daemon, it should be snmptthandler. This will be a game breaker if it isn't set correctly. Also, with the debug on, do you see snmptrapd messages showing up in the /var/log/messages when traps are sent?Code: Select all
mode = standalone
-
scottwilkerson
- DevOps Engineer
- Posts: 19396
- Joined: Tue Nov 15, 2011 3:11 pm
- Location: Nagios Enterprises
- Contact:
Re: Help with SNMP traps?
My guess was that these systems are sending the traps with the IP as the hostname and these IP's weren't in XI yet.jbennett wrote:Lo and behold, I'm now getting traps for these 4 items.
What's different?
If XI receives a passive result (Trap or otherwise) if the host has not been configured yet, it will place the item in unconfigured objects.
Re: Help with SNMP traps?
Be sure to check out our Knowledgebase for helpful articles and solutions!
Re: Help with SNMP traps?
I have a field where I can set a host name on the PDUs. This is set on all PDUs, but it doesn't necessarially match the host name in Nagios.scottwilkerson wrote:My guess was that these systems are sending the traps with the IP as the hostname and these IP's weren't in XI yet.jbennett wrote:Lo and behold, I'm now getting traps for these 4 items.
What's different?
If XI receives a passive result (Trap or otherwise) if the host has not been configured yet, it will place the item in unconfigured objects.
I do have all PDUs listed in Nagios with a host name as a location identifier, not just an IP. Are you saying that you think this might be the issue?
What's getting me is that I haven't seen any other devices show up as unconfigured. We have like 600 of these devices and only 4 of them have shown up?
-
scottwilkerson
- DevOps Engineer
- Posts: 19396
- Joined: Tue Nov 15, 2011 3:11 pm
- Location: Nagios Enterprises
- Contact:
Re: Help with SNMP traps?
Yes, it all depends on what the PDU sends the trap as, if that matches it should show up.jbennett wrote: I do have all PDUs listed in Nagios with a host name as a location identifier, not just an IP. Are you saying that you think this might be the issue?
They will only show up if the device sends a known trap, if it hasn't sent a trap or the trap is unknown, it will not show up.jbennett wrote:What's getting me is that I haven't seen any other devices show up as unconfigured. We have like 600 of these devices and only 4 of them have shown up?
Re: Help with SNMP traps?
It sounds like I would need to go back through and adjust all of these devices to be the same as the address field and utilize the Description or Display Name field for the location name then. Am I understanding that correctly?scottwilkerson wrote:Yes, it all depends on what the PDU sends the trap as, if that matches it should show up.jbennett wrote: I do have all PDUs listed in Nagios with a host name as a location identifier, not just an IP. Are you saying that you think this might be the issue?
-
scottwilkerson
- DevOps Engineer
- Posts: 19396
- Joined: Tue Nov 15, 2011 3:11 pm
- Location: Nagios Enterprises
- Contact:
Re: Help with SNMP traps?
That sounds correct, Nagios expects the trap to be sent from what it has listed in the host_name field