What are the best practice searches for server & OS monitoring?

[sloshburch]

I'd like to implement some basic searches for server and OS monitoring without getting caught up in the differences between sourcetypes and field names.

I already implemented the Splunk Add-on for Microsoft Windows and the Splunk Add-on Unix and Linux but I find that enumerating each sourcetype and coalescing the common fields is feeling unnecessarily complicated.

Are there any basic searches that provide server and OS monitoring without me having to deal with the complexities of the sourcetype differences? Something akin to the Performance Model of the Common Information Model, perhaps?

[sloshburch · Accepted solution]

The answer for this question has been distributed to the following posts:

• Example of how to measure server disk usage?
• Example of how to measure server memory usage by host?
• Example of how to measure server network usage?
• Example of how to measure server CPU usage?

[sloshburch]

Replaced the answer with its new homes.

[sloshburch]

Updated "Storage Free" search to display by different mount/disks.

[sloshburch]

Added a section about Alerts and incoming link What are the best event-data inputs for basic server and os monitoring?.

[gjanders]

Does the existence of the Splunk App for Infrastructure change any of your recommendations?

[sloshburch]

@gjanders - You're spot on. That's the intent of the "Future" section here. We'll eventually update this for the metrics and Splunk App for Infrastructure which is quickly becoming a game changer in this domain.