How often does this collect Unix/Linux server performance metrics and where does it get them from?

[andycaley1]

I'm looking to monitor unix (various flavours) server performance metrics at a deep level of granularity and wondered what is already out there. NMON may not be an option. Is there another solution written for Splunk?
What stats do you get and how often are they sampled? Looking at every 5 seconds if possible, for troubleshooting

[guilmxm]

Hello !

If neither Nmon or legacy TA*nix are options for you, I would suggest you have a look at ingesting metrics from statd or collectd:

https://www.splunk.com/blog/2017/09/26/what-s-new-in-splunk-enterprise-7-0.html

https://docs.splunk.com/Documentation/Splunk/7.0.1/Metrics/Overview

Guilhem

[adonio]

Hello there,

you can start with the Splunk TA for Nix
https://splunkbase.splunk.com/app/833/
here is a list of all the data you can grab out of the box
http://docs.splunk.com/Documentation/UnixAddOn/5.2.4/User/SourcetypesandCIMdatamodelinfo
you can fool around with different intervals on different metrics
example for inputs.conf
[script://./bin/vmstat.sh]
interval = 60
sourcetype = vmstat
source = vmstat
index = os
disabled = 1
set the interval as you wish, modify disabled to "0" and start splunking away

hope it helps

p.s. same goes for windows and windows TA