Using the third party SNMP devices (Cisco Prime Infrastructure) to monitor the 9630 by SNMP (Network Management Card 2)

[Anonymous user]

My main purpose is trying to use the third party SNMP devices to monitor UPS' alarm and to be informed when the UPS is off line and its battery is dead.

First of all , I would like to know what difference between SNMP polling and SNMP traps.  

For SNMP traps:     If I will let the third party SNMP device to receive the SNMP traps , How do I let the third party SNMP device understand the UPS' trap  and translate those traps to a readable Alarm message?

For SNMP polling:   I already imported the UPS' MIB (file: powernet407) to the third party SNMP devices.  But I see there are so many Tables (see below picture) in the MIB.  I do not know which Table is for checking if the battery dead , which table is for checking if the UPS is dead ? 

[BillP]

Check .iso.org.dod.internet.private.enterprises.apc.products.hardware.ups.upsBattery.upsAdvBattery.upsAdvBatteryReplaceIndicator / .1.3.6.1.4.1.318.1.1.1.2.2.4 for replace battery or not and .iso.org.dod.internet.private.enterprises.apc.products.hardware.ups.upsOutput.upsBasicOutput.upsBasicOutputStatus / .1.3.6.1.4.1.318.1.1.1.4.1.1 for the status of the UPS (if it is online or on battery or any other state).

[Anonymous user]

My main purpose is trying to use the third party SNMP devices to monitor UPS' alarm and to be informed when the UPS is off line and its battery is dead.

First of all , I would like to know what difference between SNMP polling and SNMP traps.  

For SNMP traps:     If I will let the third party SNMP device to receive the SNMP traps , How do I let the third party SNMP device understand the UPS' trap  and translate those traps to a readable Alarm message?

For SNMP polling:   I already imported the UPS' MIB (file: powernet407) to the third party SNMP devices.  But I see there are so many Tables (see below picture) in the MIB.  I do not know which Table is for checking if the battery dead , which table is for checking if the UPS is dead ? 

[BillP · Accepted solution]

Traps are the notification portion of the SNMP protocol - when something happens, you get an alert that it occurred. For Network Management Cards, you can also tell it how often you want it to be delayed, repeat, and up to how many alerts.

Polling is the combination of a request and a response. You request data at the rate you specify (every minute, 5 minutes, etc) and you receive a request to that request with a value in integer, timeticks, or string depending on the information.

Typically, customers use both but it is not required. Polling is good for trending data over time where as traps are ideal to know when a condition occurs, such as replace battery. Here is another good supplement -> Trap and Polling

For a UPS device, you'll want to look at anything that begins with ups or is in the ups subtree. The PowerNet MIB contains all of the APC products in it and all of their OIDs. There are also OIDs not specific to APC information that are responded to that most SNMP devices, no matter the vendor, will respond to.

Can you clarify what type of UPS your Network Management Card is in? I can provide a helpful spreadsheet for some NMC applications that could tell you what events map to which traps so like "self test failed" or "bad battery" will identify what trap number to use.

As far as the particular tool, I am not quite sure how you can configure it to do specific actions based on traps but we can at least get you set up to receive them. Often, they have descriptions that are pretty straight forward but you may want to configure specific strings or map them to an action, i.e. log a ticket in your ticketing system.

Lastly, it may be helpful to load the APC PowerNet MIB into a MIB browser tool (such as GetIf or iReasoning) to browse the MIB and read the descriptions of the particular OIDs if your Cisco tool does not have an integrated MIB browser. Another option is to do an SNMP walk (which will GET/request all possible OIDs) in order to identify all of the possible items that you could poll. In your MIB browser, you'll find the ups subtree by navigating to .iso.org.dod.internet.private.enterprises.apc.products.hardware.ups (or .1.3.6.1.4.1.318.1.1.1) and you can actually look at all the traps at .iso.org.dod.internet.private.enterprises.apc.apc# (or .1.3.6.1.4.1.318.0) and then you'll see the different trap numbers. The trap numbers will change the last number in the numerical identifier - .1.3.6.1.4.1.318.54 - would be trap #54.