I am about to open a ticket but would also like to hear what the community has to say about it. I found a few similar discussions here but they are a few years old and would like to revisit this topic. The hardware monitor is not picking up several components on our Nexus 7K/5K/2Ks. Hardware Health only shows temperature readings but not the power supplies and fans status, which are present on all of our other Cisco switches. I know we can pull the data using UnDP which would require a bit of MIB research but I imagine this would be a supported feature on the native hardware monitor. Any comments/suggestions are greatly appreciated. Thanks.
Our specs -- NPM 10.5, Nexus 7010 NX-OS v6.1(4), Nexus 5020 NX-OS v5.0(3)N2(1)
Not sure if anyone is still watching this old thread but just wanted to share my experience setting this up. I specifically was looking to alert on module status.
First, I added the the module status OID (22.214.171.124.126.96.36.199.188.8.131.52.1.2) as a new universal device poller, and under advanced options, chance format to enumeration, and manually type in the values from above. (press enter to get another line)
Then I added the physical description OID (184.108.40.206.220.127.116.11.18.104.22.168) as another new universal device poller.
Then I assigned these poller to the appropriate devices.
To add this data to the node view, you need to add the 'tabular universal device poller' resource to a view. Then under Select tabular universal device pollers for display, I checked the two previously created UNDP above.
Then for the rows I checked 22-36 (specifically just for modules) as called out above.
For 'Select the poller from which labels are taken' I just left it on the module status poller I created.
Now for the alerts, under trigger condition, for 'I want to alert on:' I put 'Custom Table Poller Currect Status'.
For scope I just left it on all objects.
For actual trigger condition, I put two conditions with AND
-Custom Node Poller, Unique Name (Custom Poller), is equal to, <name of custom poller for module status created above>.
-Custom Table Poller Current Status, Current Numeric Value, is not equal to, 2
Hope this makes sense.
(NOTE: when I was setting up the resource for the node view, I had to wait a day before it worked, because for whatever reason when I tried to assign a poller to the resource it just kept erroring out.)
I manage to create a custom SQL query for the alert variable to grab the module name from the UnDP entPhysicalDescr, alert variable as below:
I have it all displaying correctly now in my web console, thanks so much for your advice!! As regard alerts I want to alert if the supplies are anything other than OK. Have you set these up before? Did you alert on 5the numeric value returned by custom poller or the mapped to value (word)? Thanks alot!
Nexus Operational Status - NexOperStat
Nexus Power Status - NexPwrStat
These return Numeric values for status. So the alert triggers when the Status of the poller is NOT 2.
Alert trigger setup would be
Poller name = NexOperStat
Status <> 2
and then any other qualifying conditions... I use a machine type contains Nexus, or NX... some variant to pick up all my devices.
I used the above Physical Names and used this MIB to get the module card status' : 22.214.171.124.126.96.36.199.188.8.131.52.1.2
Then map your values here :
And I got this to return properly.
Hope this helps
make sure they are tabular pollers, and add the Tabular Poller resource.
Then you have to edit it, check the columns you want and then the row #'s you want to see.
If you only have the poller display resource you may see your tabular pollers, but it will show how many values you have in the table.
SolarWinds solutions are rooted in our deep connection to our user base in the THWACK® online community. More than 150,000 members are here to solve problems, share technology and best practices, and directly contribute to our product development process. Learn more today by joining now.