Nagios Support Forum

Hi, I'm planning to use nagios in a big company with important safety constraints. I would like to propose a centralized solution with a unified monitoring console, integrated with external trouble ticketing and asset management software.

We're talking about 5000 servers and 7000 network devices.

Today we are using the HP suite of monitoring of our assets.

Having to migrate to nagios, which is the recommended solution on a distributed environment in different locations?

Nagios Xi in federated monitoring with remote monitoring nagios core?

Fusion with remote Nagios core or xi?

I repeat, the objective is a single centralized console integrated with external trouble ticketing and asset management tools.

Thanks

I would like to propose a centralized solution with a unified monitoring console, integrated with external trouble ticketing and asset management software.

The limitations really aren't on Nagios for this, it really depends on the integration that your other systems have available. Nagios is mostly a frame work, that allows you to script out anything you need. Event handlers are very powerful, and can be executed every time a state changes. Notifications don't necessarily need to be an email, they could also be a CURL request to an API.

We're talking about 5000 servers and 7000 network devices.

Depending on how many services under each host, you may be looking at multiple XI nodes. Are these all at the same site, or do you have multiple geographical locations?

Having to migrate to nagios, which is the recommended solution on a distributed environment in different locations?

I would look at using gearman at other locations, to report back to Nagios. You could also proxy things through check_by_ssh, or check_nrpe, but gearman will probably be the most efficient. You could assign USA hosts, to a USA host group, and then have one worker specifically for that host group.
https://assets.nagios.com/downloads/nag ... ios_XI.pdf
https://support.nagios.com/kb/article.php?id=484

Nagios Xi in federated monitoring with remote monitoring nagios core?

Could you clarify this question? If you're looking for everything to report to one machine, with distribution, gearman is probably the way to go that I mentioned above. If you're looking for a HA solution, look at what our partner Linbit offers for DRBD replication -
http://www.linbit.com/en/resources/tech ... centos-6-5

Fusion with remote Nagios core or xi?

Fusion will be able to interface multiple Core or XI machines. Not sure what else to mention here.

I repeat, the objective is a single centralized console

You could probably use Fusion to manage a few different XI machines, but it wouldn't be able to manage the XI machines from just that. You would still need to login to each machine that is used in conjunction with Fusion.

integrated with external trouble ticketing and asset management tools

I would need more information about your expectations / requirements to expand on this part.

Thanks a lot for your support !!!

We'll have multiple geographical locations and ideally we would want to have an encrypted and one-way connection from the peripheral to the central.

What are the main differences between Modgerman and federated monitoring ?

You suggest the use of virtual or physical machines ?

Best regards

carlobrd wrote:Thanks a lot for your support !!!

We'll have multiple geographical locations and ideally we would want to have an encrypted and one-way connection from the peripheral to the central.

What are the main differences between Modgerman and federated monitoring ?

You suggest the use of virtual or physical machines ?

Best regards

This day in age, I suggest virtual machines. They are easy to spin up, tear down, and snapshots are life saving IMO.

I believe a federated monitoring system would be multiple Nagios instances, where as mod_gearman would use one Nagios instance with 'workers' at each site handling different checks. Depending how large your environment is, you may require multiple XI machines though, even with gearman.

Not sure, if this would help.

Our virtual environment is all vmware. We have vROps, we have a management pack which feeds all nagios data (service check results, perf data etc) to vROps. so all our storage, network, EC2, Azure etc feeds data to vROps.

We have plugged in Nagios Log Server to vROps. So you can see logs of an event directly on vROps instead of going to LS.

You can even use ManageIQ.

So single pane for the whole environment.

Servers (Physical and Virtual) = 12,000
Others (storage,network devices, h/w etc) = 3,000

4 GEO locations

Min 8 Service checks per host
Max 21 Service Checks for some hosts

Max of 800 servers reporting to one NagiosXI.

We monitor OS,App and M/W.

With help of tagging we create a view of relationship between servers.

Thanks,
Anish

Hi @carlobrd,

I know you have some specific questions, and I think most if not all of those have been addressed.

I'd like to point out that we do have a list of some of our customers at https://www.nagios.com/users/. There are many big companies, like Shell and Sony. Your environment may not look like their environment, but there can be little doubt that Nagios will work at a Big Company.