First, ensure that all cacti hosts can ping the monitored hosts; If you cannot ping, verify that the network configuration and IP Settings of the host being monitored are correct.
Second, if ping is possible, verify that the monitored host has enabled SNMPD services:
ps -ef | grep snmp
Or directly restart the SNMP service of the monitored host:
rcsnmpd restart
If the above command is not available, use these two:
service snmpd reload
service snmpd restart
Then go to the Cacti server and use root:
Snmpwalk-c Public-V 2C 192.168.124.14 –& GT; (This IP is the IP of the monitored host)
If the data information of the monitored machine can be received, it means that SNMP configuration of the monitored host has been completed and there is no error. If no data information is received from the monitored host, proceed to step 3.
Third, log in the monitored host with root and modify the CONFIGURATION file of SNMP:
vi /etc/snmp/snmpd.conf
The final configuration is as follows:
syslocation Server Room
syscontact Sysadmin (root@localhost)
Rocommunity public 127.0.0.1
agentaddress 161
rocommunity public
rwcommunity private
Trapsink 192.168.124.14 Public 162 –& GT; IP =192.168.124.14 is the monitored host IP
Then, perform the second step.

Reproduced in: https://www.cnblogs.com/tommyli/p/3214595.html