Presently, we're using NConf for configuration and Nagios core for our central portal and for distributed nodes.
NConf allows us to specify which hosts and services are active on any given distributed node; we push out the Nagios configs to the distributed nodes via scp. The distributed nodes (each running Nagios core) submit their passive check results to the central portal.
Does the Nagios CCM or NagiosXI at all have this capability? This is a critical requirement for us before making the move to NagiosXI. Thanks!
NagiosXI and distributed monitoring
-
slansing
- Posts: 7698
- Joined: Mon Apr 23, 2012 4:28 pm
- Location: Travelling through time and space...
Re: NagiosXI and distributed monitoring
Please take a look at the following document:
http://assets.nagios.com/downloads/gene ... utions.pdf
In addition, Nagios XI has the ability to receive check results from multiple Nagios instances so that you can forward this information to one central XI server, this is one option featured out of the box.
http://assets.nagios.com/downloads/gene ... utions.pdf
In addition, Nagios XI has the ability to receive check results from multiple Nagios instances so that you can forward this information to one central XI server, this is one option featured out of the box.
Re: NagiosXI and distributed monitoring
I have read that, thanks. My hope was there was some way within the web interface to specify where (on which nodes) checks should be executed. It would appear that the only way to do this is to use something like mod_gearman and configure it to execute checks based on hosts or hostgroups.
-
slansing
- Posts: 7698
- Joined: Mon Apr 23, 2012 4:28 pm
- Location: Travelling through time and space...
Re: NagiosXI and distributed monitoring
Are you speaking in regards to switching service checks on and off per Host "node"? Or mirroring out a installation and selecting which nagios XI server to check a specific Host/Service from? While you could do such things with Mod Gearman, it is likely that you may be able to do the same through the XI interface out of the box, we just need to pinpoint exactly what you are looking to do "as technical as possible."
Re: NagiosXI and distributed monitoring
The desired architecture (and what we have now) is:
- Central Node
-- NO active checks are run here
-- Runs Nagios core and only accepts data (passive check results) from Distributed Nodes
-- Maintains the central repository of configuration data for all Distributed Nodes
-- Pushes configuration data to the Distributed Nodes
- Distributed Nodes
-- Run Nagios core and execute active checks, sending results back to the Central Node
-- Receive configuration data from the Central Node
This architecture is important to us because many of the Distributed Nodes reside at customer locations. Having a Distributed Node locally at these locations provides for more accurate and efficient monitoring (low-latency) of a customer environment and does not leave monitoring dependent on the Internet link between "us" and the customer location. It also eliminates the need for the customer to open several ports from our location over the Internet for monitoring purposes. From the customer's perspective, we only require ingress SSH and egress NSCA or NRDP.
From looking at the XI configuration interface I don't see at least natively a way to define Distributed Nodes or to select which hosts and services should be executed (active) on a particular node. This feature is available within NConf (they call Distributed Nodes "collectors").
Thanks for your help and please let me know if you require further clarification.
- Central Node
-- NO active checks are run here
-- Runs Nagios core and only accepts data (passive check results) from Distributed Nodes
-- Maintains the central repository of configuration data for all Distributed Nodes
-- Pushes configuration data to the Distributed Nodes
- Distributed Nodes
-- Run Nagios core and execute active checks, sending results back to the Central Node
-- Receive configuration data from the Central Node
This architecture is important to us because many of the Distributed Nodes reside at customer locations. Having a Distributed Node locally at these locations provides for more accurate and efficient monitoring (low-latency) of a customer environment and does not leave monitoring dependent on the Internet link between "us" and the customer location. It also eliminates the need for the customer to open several ports from our location over the Internet for monitoring purposes. From the customer's perspective, we only require ingress SSH and egress NSCA or NRDP.
From looking at the XI configuration interface I don't see at least natively a way to define Distributed Nodes or to select which hosts and services should be executed (active) on a particular node. This feature is available within NConf (they call Distributed Nodes "collectors").
Thanks for your help and please let me know if you require further clarification.
Re: NagiosXI and distributed monitoring
You could use nrdp or nsca to forward the checks from a core server to an XI box.
NRDP:
http://exchange.nagios.org/directory/Ad ... or/details
http://assets.nagios.com/downloads/nagi ... erview.pdf
NSCA:
http://exchange.nagios.org/directory/Ad ... or/details
http://nagios.sourceforge.net/download/ ... _Setup.pdf
NRDP:
http://exchange.nagios.org/directory/Ad ... or/details
http://assets.nagios.com/downloads/nagi ... erview.pdf
NSCA:
http://exchange.nagios.org/directory/Ad ... or/details
http://nagios.sourceforge.net/download/ ... _Setup.pdf
Former Nagios employee
"It is turtles. All. The. Way. Down. . . .and maybe an elephant or two."
VI VI VI - The editor of the Beast!
Come to the Dark Side.
"It is turtles. All. The. Way. Down. . . .and maybe an elephant or two."
VI VI VI - The editor of the Beast!
Come to the Dark Side.
Re: NagiosXI and distributed monitoring
This is what we do now and I realize we can also do that with XI. It's really the configuration component that I'm concerned about. NConf natively supported the central configuration model I describe above but I don't see that NagiosXI supports such a model natively.
-
slansing
- Posts: 7698
- Joined: Mon Apr 23, 2012 4:28 pm
- Location: Travelling through time and space...
Re: NagiosXI and distributed monitoring
I do not believe what you are asking for is directly available, as far as pushing the configuration of one server, to another.. However.. you can import the configuration files of one XI server to another, or back one up and restore it to a new system.. this in effect clones the first server and allows you to set up distributed networks, or failover systems. This is mostly because of the database re-work that XI went through. I believe this is what you are asking for?
Re: NagiosXI and distributed monitoring
Have you looked at nrds? http://exchange.nagios.org/directory/Ad ... 29/details
Former Nagios employee
"It is turtles. All. The. Way. Down. . . .and maybe an elephant or two."
VI VI VI - The editor of the Beast!
Come to the Dark Side.
"It is turtles. All. The. Way. Down. . . .and maybe an elephant or two."
VI VI VI - The editor of the Beast!
Come to the Dark Side.
Re: NagiosXI and distributed monitoring
I think this answers my question. However, I think using either NRDS and mod_gearman (or combination thereof) and some heavy scripting we can redesign our monitoring system to accommodate XI. Thanks.slansing wrote:I do not believe what you are asking for is directly available, as far as pushing the configuration of one server, to another..