Showing results for 
Search instead for 
Did you mean: 
Create Post
Level 12

Monitoring Status of Dual-SUPs

We are replacing all of our distribution switches with Cisco 4507's with dual SUP 7e's. I am trying to figure out a way to monitor the status of the SUPs, and then generate alerts if the status of one of them changes. I would appreciate any suggestions.



6 Replies
Level 10

I started with this MIB for detailed module info: MIB-2-ENTITY-MIB. I did a walk on this OID, to get info about the modules.

I started with this MIB for module status: CISCO-ENTITY-FRU-CONTROL-MIB. I did a walk on this OID to get the status of the modules.

As you do a walk and see the various OIDs, you'll see that there is a pattern connected to the slot in which they are installed. I've also seen that slots with no cards will not return data, so if you configure UnDPs for the specific OIDs like I do, it'll give the message "no data", which is easy enough to filter in the alerts.

I found it easier to deal with the individual OIDs that I want to poll. For the cards in slots 3 and 4:

Slot 3 contents:

slot 3 status:

Slot 4 contents:

Slot 4 status:

The status OID actually returns a number. You can see what each number represents here Cisco SNMP Object Navigator.

I wonder if this is the best way to handle this, but it works for now. On a side note, I use this same method to poll various chassis models for module contents and status for all slots.


Thanks!!! But I put them all in the same Table with the Status, for single table display 

I am using the Tabular OID; and for Model and Serial # Display,

When you assign, search for 000 and then 0000 to get the Thousands - to make a table for your modules. (Check 1000, 2000, etc).

Next step is turning this into a full report of all my switches.. Grand way to start! thanks RomeoG

module status - 45xx.png

0 Kudos

I see. I actually use the UnDP Poller Status resource on our node details pages, which combines all UnDPs into a single table, but I guess one downside is that ALL UnDPs go int a single table. But, if you group your UnDPs like we do, it groups them in the table as well.


Let me know how you get the alerting working. I wasn't able to get alerting working the way I wanted when I was polling the table that way. I think it was due to the dynamic and tabular nature of the higher OID rather than the more specific one. I was probably doing something wrong, but it was just easier to poll the leaf OID since you always get only 1 simple value.


0 Kudos

Your Table is skinnier, my table is shorter.

Of course I could remove the Admin status, truncate the poller titles to bring the column names closer and my table would just be smaller with more data;

I have massive use for the Model and Serial #'s so adding those to the table was GRAND for me!

0 Kudos

ALL conditions must be true.

custom tabular poller = cefcModuleOperStatus    (or whatever you have renamed it to)

value NOT = ok

So when the module status is not OK, then it sends an alert.

0 Kudos


this is possible to make alert in Advanced Alert Manager.

- Click Configure Alerts

- New

- Go to Trigger condition tab

- Switch "Type of Property to Monitor" to "Custom Node Table Poller"(in case of table OID) or "Custom Node Poller"(in case of single OID)

- Click Add in bottom of the dialog to add conditions - Select Simple conditions

- Set conditions like

Poller Name is equal to entPhysicalModelName (Or other name of your poller)

Value(or Numeric Value) is equal to "Warning"(or 5)

- Replace "Warning" or "5" with value that is reported by device in case of error condition.

- If you have selected "Custom Node Table Poller" you can also add condition Label is equal to Row name.

Next go to Trigger actions tab and Add action that should happen when alert is triggered.

Also set some friendly name for the alert on General tab.