hello friends! I have the following task:
it is necessary to monitor hundreds of hosts (virtual machines). now they are monitored using Nagios, but I want to get a clear visualization of the availability of hosts.
there are a few points:
- I cannot use all *beats ELK modules (heartbeat, metricbeat, etc.): third-party software cannot be installed on all these hosts
- monitoring is needed based on icmp (ping)
are there any implementations for ELK so that in the end I can make a dashboard in Kibana with a list of all hosts and their online / offline status (and status history of each host)
You can use heartbeat installed on an external host and configured to ping your hosts, heartbeat does not need to be installed in the host it will monitor.
But this works only for availability, if you want things like CPU Usage, Disk Space etc, you need an local agent or build some custom script that will get this information and send to Elasticsearch.
this is splendid! and make the task easier. I thought heartbeat should also be installed on client machines, thanks!
Also is an option to make a script Pearl, Powershell, Python with bulk insert directly to ES or via FB/LS to check:
- service status
- disk space
The script can be very simplified version of agents which leandrojmp mentioner or use Osquery.
Also is an option to use SNMP and Nagios integration.
If you want only heartbeat, that's fine, but it's very limited.