Stuck at setting up ELK for monitoring

Elastic Stack Kibana
1

I'm new to ELK stack and trying to set it up to monitor my servers and services, but when i installed everything and installed metricbeat when i go to kibana to see the data that sent by metricbeat i overwhelmed by huge number of statistics and available filed which most of them are useless and empty like the following picture:

image
image957×549 56.3 KB

over 6k available fields and in the dashboard area:

lots of dashboards which are empty because this server doesn't have anything on it yet, and i disabled every module on metric beat just the linux module is enabled,

clearly i miss something, can anyone direct me to the correct way

thanks

2

Hi @Aamira Welcome to the community.

Couple questions/ clarification

What version of Elastic stack?

How did you install?

Apologies, it is not exactly clear What hosts and services are you trying to monitor?

The elastic components themselves and servers?

Or different servers, general host metrics like load CPU RAM?

What exactly are you trying to monitor...

3

Hi @stephenb thanks for the reply

The version i installed is version 8.11.1 and i installed it using the elastic RHEL repository on my linux server.

I want to set this server as my in-house monitoring server and monitor all of my VMs and the services on them like HTTP services and databases, but i want to start with simply monitoring a linux server metrics like cpu usage and memory usage, disk usage and etc.

can you point me to a documentation or a guide to help me set it up correcly ? i cant find the desired metrics from 6K available fileds :)))

4

I would perhaps suggest following the metricbeat quick start guide

You do not need to pick the fields they will be picked for you. You can adjust the metric sets.

The default system module will collect most of the host metrics that you will probably be interested in.

Please follow the steps very closely

The System module collects the host metrics

You can look at the other modules for database metrics.

If you want to do HTTP monitoring, typically you would use heartbeat for that.

Of course there's the whole new fleet and elastic agent, but you did not indicate that you had an installed that

1 Like
5

Thanks my friend

i followed the documentation that you suggested and it worked for me, but i'm still wondering why there has to be alot of predefiend dashboards and fields with no data in it, for example i can see over 10 dashboards about kubernetes and other tools which i dont have installed on the server or anywhere else and i have to find the system host dashboard to see the statistics between all of them. anyway thank you

6

It's just the way beats load assets... I wouldn't spend too much time trying to figure it out. It's just the legacy way of doing things.

Beats pretty much loads all the assets. You can remove what you don't want (be careful )the new elastic agent is more granule

1 Like
7

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.