Migrating Monitoring Configuration's to NagiosXI

[gary80918]

We are in the procurement stage of migrating 8000 servers monitoring from HPOA to Nagios XI. During the migration process it is not reasonable to setup each server individually so per server I have all customized monitoring in csv format, example:
Filesystem csv file:
server123|Linux|Customer|in production|/|[OS,NONE,NONE]|97% 90% warning 0000-2400 *|98% 95% critical 0000-2400 *||
server123|Linux|Customer|in production|/boot|[OS,NONE,NONE]|90% 90% warning 0000-2400 *|95% 95% critical 0000-2400 *||
server123|Linux|Customer|in production|/dev/shm|[OS,NONE,NONE]|90% 90% warning 0000-2400 *|95% 95% critical 0000-2400 *||
server123|Linux|Customer|in production|/home|[OS,NONE,NONE]|97% 90% warning 0000-2400 *|98% 95% critical 0000-2400 *||
server123|Linux|Customer|in production|/msr|[OS,NONE,NONE]|90% 90% warning 0000-2400 *|95% 95% critical 0000-2400 *||
server123|Linux|Customer|in production|/opt|[OS,NONE,NONE]|97% 90% warning 0000-2400 *|98% 95% critical 0000-2400 *||
server123|Linux|Customer|in production|/opt/app1|[OS,NONE,NONE]|90% 90% warning 0000-2400 *|95% 95% critical 0000-2400 *||
server123|Linux|Customer|in production|/opt/app2|[OS,NONE,NONE]|90% 90% warning 0000-2400 *|95% 95% critical 0000-2400 *||
server123|Linux|Customer|in production|/tmp|[OS,NONE,NONE]|97% 90% warning 0000-2400 *|98% 95% critical 0000-2400 *||
server123|Linux|Customer|in production|/var|[OS,NONE,NONE]|97% 90% warning 0000-2400 *|98% 95% critical 0000-2400 *||

Process csv file:
server123||||OS|NONE|NONE|crond|critical||1-|||
server123||||OS|NONE|NONE|rpcbind|critical||1-|||
server123||||OS|NONE|NONE|python|critical||1-|||
server123||||rsyslogd|APP|rsyslogd|rsyslogd|critical||1-|||

The log monitoring csv will be in a very similar format, still working on it though.


Our current monitoring tool is HPOM and when we move to Nagios XI (Core, Log Server and Network Analyzer (fusion also)) our intent is to use the NCPA agent on Windows and Linux systems. Can anyone tell me if they have done this before, or at least tell me what files (format) to convert these csvs to?

Thank you for any help/guidance you can provide.

[Box293]

What I would do is write a script to iterate over your CSV file and then use the API to add hosts and services to Nagios XI. Full documentation on the API with examples is located via Help > API Docs > Config Reference.

[gary80918]

Thank you, that was my intent. Was hoping for a command line way to do this. Looking at the API doc now.

[Box293]

OK great. You'll find that the API method using a scripting language will be the best over time. I suggest python as I've had good success playing with it.

[gormank]

With that many (or in my opinion any number) of hosts, I'd use templates and define them with the bare minimum of info, which is hostname, address, and template.
I'd do the same for services. For example, I'd create a single disk check for Linux and apply a hostgroup containing all Linux hosts to it.
This separates host/service info from monitoring info such as check intervals, notification range, users notified, etc.
Of course, some would argue that this makes things inflexible if there's a desire to change things (an example might be monitoring interval) on a one by one basis. This is true. For hosts it isn't an issue, but for services it can be.
Still, it might not be fun to change 8000 disk check services even with the bulk tool.

[gary80918]

Thank you Gormank,

There is to much customized monitoring in this environment. We can and intend on doing what you suggested with a standard build server. However, our environment has 100's of specialized app servers each one has a LOT of custom monitoring configurations. So....

Anyway, Box293 pointed me in the right direction, and I now have what I needed to get started.

[Box293]

Excellent, let us know if you need any further help.