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.

Creating Temperature Alerts

How can I setup alerts for my temperature readings on my cisco devices?

Thanks

  • i don't think there's a magic bullet for this, so it will have to be done on a per-device basis because it seems that many Cisco devices have varying tolerances to temperature, which is based on device chassis, modules, etc and each can have different thresholds. also, some Cisco devices don't support a TEMPERATURE OID table like the switch shown below but might be presented differently in UnDP. i suggest you research the Cisco device on Cisco's website and find out the temperature tolerance/threshold.

    for instance on this switch - you might build a UnDP poller for the TemperatureStatusValue and create your Advanced Alert Trigger to test on the TemperatureThreshold value, taking into consideration the TemperatureStatusDescr. it can get a little bit involved, but it is doable.

    and you'll need to consider celcius/fahrenheit. UnDP provides a TRANSFORM feature to build that conversion. i hope this helps in some little way...

  • I am having a problem getting more columns in my table to show up in the web view. How can I do that?

    Thanks.

  • as you continue to add other Pollers from the UnDp OID table and they are assigned to your Orion nodes, then as they appear in your Node Details View (web view) then select EDIT from your first newly created poller and you'll see a list of your table/pollers in the area named Select tabular Universal Device Pollers for display. Check those related to that OID group you're working on and they should show up in the web view. NOTE: i usually have to do some clean-up in the Resource View named NODE DETAILS to remove those extra pollers after making a new Poller group.

  • I got my table all setup. Now just have to figure out how to get the Temperature info to show up when I create a new alert.

  • I believe that most (all?) Cisco devices can be configured to send a trap when temperature exceeds a threshold.  That threshold is set by default on the device, but can be changed if you wish.  

    So rather than trying to set up a temperature alert on pollers, it might be easier to set up an alert based on received the 'overtemp' trap.  

    I do all my "environmental" alerts based on traps as opposed to polling for them.  I have alerts set up for temperature, fan failure, redundant power supply failure, etc. based on the trap instead of a custom poller.  It's faster, too.  A power supply loses power, gets pulled out, turned off, whatever and I'll have an email in seconds rather than waiting the 1 to 5 minutes for the device to be polled than another 1 to 5 minutes for the alert to check the status of the poller.

    Don't know if that helps, but it works for me.

  • Ten48

    good point using the traps. there's definately pros and cons to both.

    my greatest problem with traps is it's incoming trap name. how do you prepare to capture the exact trap name?

    i guess what i'm asking is, how do you anticipate the trap type or is there a repository of these for (all) Cisco devices somewhere? this has been my biggest problem to nail down with traps.

  • That is a good idea with the traps. I looked at my switch configs for the trap settings and they are not all the same between different switches. Not a big deal, it would just take a bit more to setup alerts based on the trap name.

    With the pollers I can keep historical data on my temp readings.

  • To find the correct OID, I did a MIB walk on a couple devices and just browsed through the Cisco env (environmental) MIB.  

    I also set up a test 2811 I had laying around to send traps to Orion and opened up the case and yanked off the cooling fan power cable to verify that the fan failure OID I was the one I was the one I was expecting.  After I set up the alerts, I tested further by pulling out a redundant power supply in one of my less important (out of band backups) 4506 switches to verify I was looking for the correct OID.  

    The Cisco ENV MIB is pretty universal among their different product lines.

    As user salating does, I also set up a temperature poller to keep historical information about temperatures on my switches and routers, I just don't use them to fire alerts.

  • Hello Ichance,

    I want to configure same table for my Checkpoint device not getting the option for this. Could you please give me the detail path from where i can do this. I have set up the UnDP for temperature.

    Below is the information about my environment :-

    NPM- 12.0 ,

    Orion Platform- Orion Platform 2016.1

    NCM-7.5