This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.

You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Change threshold for all nodes that are specific model(s)?

jaybone over 7 years ago

NPM 11.5.3

We had an AC failure yesterday and didn't get an alert on this until over 6 hours into the event. The one thing that seems to have changed at the time of the alert was that the Intake Left value ticked up from 113.xF where it'd been for over an hour, to over 114F, just before 10pm. Max operating temp for this particular line of devices, per Cisco, is 104F. By the time we could get someone in the room this morning, the ambient temp was up to 120F.

I see where I can change the temperature threshold for the individual device, but we have dozens of these things. I'm failing to find anything regarding how to do mass-changes for something like this - using the wrong search terms, maybe.

Is there some way to tell NPM that any/all Cisco device model 29XX's intake temperature threshold should be warn at, say 99F or over 90F for X minutes, and critical at 104F?

Top Replies

blsanner over 7 years ago +2

If you go to Setting --> Manage Hardware Sensors, you can apply filters to the columns to limit it just Temperature for the hardware category: You can also filter the model under the product line column…

0 borgan over 7 years ago

On the Trigger Conditions tab of your alert, you can limit the scope to just that Machine Type. You could then configure another version of the same alert and have it apply to all other machine types if you need it. Is that what you are after?
Cancel
Vote Up 0 Vote Down

Cancel
0 blsanner over 7 years ago

If you go to Setting --> Manage Hardware Sensors, you can apply filters to the columns to limit it just Temperature for the hardware category:
You can also filter the model under the product line column:
You can then select the sensors and do a mass edit:
Cancel
Vote Up +2 Vote Down

Cancel
0 jaybone over 7 years ago in reply to blsanner

Thanks. Yeah, I saw this, but this won't help in the future when more of these are added, unless someone goes in and reconfigures the newly-added node.
Cancel
Vote Up 0 Vote Down

Cancel
0 jaybone over 7 years ago in reply to borgan

I didn't think of attacking from this angle, but yeah, this could probably work.
I'd have to know what alert to edit, though, which is a problem. As far as I've ever been able to figure out, there's no way to go back after the fact and tell which alert was actually triggered. I can look at the reports and see that, yes, at this time, on this node, some alert was triggered and generated this message, but nothing about *which* alert that was. Given that we have hundreds of alerts configured, finding the proper one to edit would take a huge amount of time. I know that if they have trigger actions that send an email with ${N=Alerting;M=AlertName} in them, that would identify the alert, but none of them have that.
Cancel
Vote Up 0 Vote Down

Cancel