Remote retrieval of system information

Question

Monitoring

Good day and Happy New Year, friends!

Once a potential employer asked a question that still interests me:

Let's say there is a server with free/debian on board, you need to check (or constantly monitor) the state of temperature and the fullness of hard drives on remote computers, starting from 50 pieces.
It is clear that in isolated cases you can connect via ssh and find out by top, but I don’t know how to do this in bulk. Tell.

Answer 1

Zabbix , Nagios , no?

Answer 2

snmp, ipmi/ilo/Dell's crap, I don't remember how it will start (if lmsensors doesn't support the chipset).
Zabbixm Nagios is already a monitoring system that runs on top of it all.

Answer 3

there was an article recently about how to read in linux. with temperature, etc.

Answer 4

Strictly speaking, the "snmp" answer simply changes the information access protocol from ssh to a simpler one and more suitable for a large number of machines.
But in general, for mass monitoring, you still need to either poll all servers (via ssh, snmp, http, your own version), or configure the servers themselves to send notifications to the central server (syslog, mail, again ssh, http, snmpv3 or something own).
There are many options, each with its own advantages and disadvantages. For example, polling servers via snmp and building rrd on one server is very convenient and normal when there are less than a hundred of them, and one central server will not be enough for a thousand servers (or it must be really powerful, especially regarding disk speed).