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.

Multiple Node Temperature Chart

I would like to create a single chart to display the temperature of several nodes.

How would I accomplish this?

The node types vary so a single UnDP would not work.

Is there a way to create a chart that displays the temperature reported under the nodes current hardware health?

  • This can be accomplished using the Custom Chart resource

    Custom Chart Resorce Temp.png

  • That gives me something but i'm not sure what.

    The number the chart shows doesn't come close to the temp shown on the node page (for Celsius or Fahrenheit)

  • just to bump this one too - i choose avg value - click preview and get:

    Screenshot_072116_105302_AM.jpg

  • You will need to provide more specificity in the resource settings as there is no one/single temperature value for a device. Devices which support Hardware Health often have multiple temperature sensors. CPU/Processor temp, Air Intake Temp, Exhaust air temp, etc. etc.. This likely explains the unusual results you are seeing when all these values are averaged together. It's also important to note that this values are retrieved from these devices and stored in the database in celsius. The values are converted to fahrenheit in the web UI for select resources based on user preference.

    pastedImage_0.png

  • so aLTeReGo​ an example would be a HP Procurve Switch, a Dell R520, and maybe a Dell Blade Centre Chassis.

    Looking at their device pages, all have the sensors - so am i best to pick say, the CPU temperature across them all? and if so, how? is this still off overall hardware health? or the node selection?

    Appreciate the help!

    Also - in the UK, so we use the correct temperature measurements anyway emoticons_happy.png

  • or, to make it easier - 16 M630 Dell Blades in one Chassis - these would have the same sensor right?

  • Anyone? this seems to be a ridiculously hard this to do! another example of the over complication or setting anything up emoticons_angry.png

  • I have a HP c7000 Enclosure, and this information is precisely what I need for triggering ALERTS (Warning / Critical), based on the results ... In my case (lattest NPM) :

    Encl.jpg

    Since you mentioned that "this values are retrieved from these devices and stored in the database", I wonder :

    - The easiest/more direct way to trigger ALERTS based on this values is doing a SWQL Query in the ALERT Manager ? Can you provide some pointers on which are the tables involved ? I was looking for easier ways, but I couldn't find an "direct" alternative using the ALERT Manager. By easier I mean finding a field accesible from the ALERT Manager related to this fields, to create the condition for the ALERT to be triggered (in the FAN case, something like "IF FAN[1-9] ARE NOT OK" -> CRITICAL ALERT)

    - Does it make sense (in order to trigger ALERTS based on this values) to create instead a UnDP Poller to gather this information ?

    Any comments or suggestions are appreciated

    Regards

    Arturo