thold to monitor hd and ram usage

[temak82]

Hi,

We have THOLD setup to monitor the traffic usage and send out alerts to the users if they are bursting over the thold for 24 cycles (2 hours I think). My question is, how can we monitor the usage of hard disk space and virtual/physical memory and send out similar alerts if say 90% of the RAM is being used?

Thanks for any help!

[mcutting]

If you are using THOLD 0.4.x you can setup a template based on the HDD Host MIB data template. Using hdd_used, you can define a percentage and who gets alerted.

Once you've defined the template, you can then apply it to all hosts who use this data source. The same template will work for disks, physical and virtual memory.

Definitely works, as I do exactly the same thing.

[temak82]

mcutting,

Thank you. This may be exactly what I'm looking for! How do you define the treshhold high/low so you get the alerts? You said you use percentages?

[mcutting]

Hi.

THOLD 0.4.1 has a good wizard which helps you create thresholds. The easiest way to do this for new thresholds is to create a template and deploy this against all the hosts using Host MIB (which is what I do).

If you're looking to deploy individually, then it can be done by gloing to the data sources and clicking the links on the right (sorry, don't have Cacti access just now). Here, you can define a threshold in the same way.

Let me know if you aren't sure on the above, and I'll post some more concise instructions.

[temak82]

Thank you very much for your reply. I will check to see what our thold is.

To be honest, I have never actually created any temlates within Cacti as we just mainly use the default ones. Aside from the Thold plugin to monitor bandwidth traffic and send out alerts, we have not really modified it in any other way. If you can provide some detailed instructions on how you have done it, for example, that would be very much appreciated.

Furthermore, I just checked the DATA Templates and we already of course have the following setup:

Host MIB - CPU Utilization
Host MIB - Hard Drive Space
Host MIB - Logged in Users
Host MIB - Processes
Linux - Memory - Free
Linux - Memory - Free Swap

Do you use any Threshold CDEFs in order to monitor the used memory, and hard drive space? CPU utilization is fairly straight forward since you can just put a high treshold for CPU to be 10 and then when it is 10 for a certain amount of time, it will email us.

Also, a rather dumb question, but is there a quick way to check what version of THOLD we currently have running?

Thanks again for your help with this!

[temak82]

Just to follow up. I created the hd_used template and found a formula for a CDEF to convert the values into percentages. However, when I go to create the monitor for the user, for some reason hd_total and hd_used are both empty for "Current Value" settings. Shouldn't they show how much current HD space is being used and match that against the High and Low treshholds?

Thanks