How much bandwidth does NagiosXI require for monitoring?

[sbaviswa]

Hi
We have around 300 network devices to be monitored using a low latency WAN connection.

Hence we would like to understand better on the bandwidth consumed for the following . . . .
1. Host checks -
2. Service checks --> check_snmp -
--> check_icmp -

Also I would appreciate if I could have information for the following type of monitoring.
1. databases checks
2. load checks

Thanks in advance
Regards
N.Viswanathan (SBA)

[abrist]

Hence we would like to understand better on the bandwidth consumed for the following . . . .
1. Host checks -
2. Service checks --> check_snmp -
--> check_icmp -

Host alive checks are icmp (ping) checks out of the box. Standard icmp echo requests are 56 bytes one way. SNMP checks are completely dependent on the type of information you are trapping.

Also I would appreciate if I could have information for the following type of monitoring.
1. databases checks
2. load checks
N.Viswanathan (SBA)

NagiosXI includes many different wizards and plugins for a whole range of databases. MSSQL, MySQL, Oracle, Mongo, etc. Is there a particular database you are interested in monitoring? Load checks can be done through snmp, nrpe and local scripts, among others. NRPE is highly extendable and plugable, so any cli utility you favor for load checks could be wrapped in a shell script, run by nrpe, and the output returned to the XI server. There may be some config time spent getting perfdata for the custom check to work correctly.

[sbaviswa]

Hi Abrist

SNMP checks are used for
1. port status capture,
2. bandwidth measure for mrtg (rrd format)

With respect to database, it will be oracle.

regards
N.Viswanathan

[slansing]

How many SNMP checks do you average you will use? Ports per switch, and amount of switches, Oracal installation sizes and what you would be checking within them etc.

[mguthrie]

Most Nagios checks poll data once every 5mn on average. So the bandwidth usage to client machines is minimal. For SNMP devices it would be a form of snmpget every 5 minutes, for other devices, a ping every 5mn.

[slansing]

It looks like you are working with sales on this, we are getting back to you via those channels, closing this and marking resolved.