Nagios World Conference North America
This page has my notes form the conference.
Ethan Galstan Key Note
What is Nagios? IT Infrastructure Monitoring
- NetSync (7 years ago) -> Nagios Open Source -> Nagios Core
- NagiosXI Commercial
- Last 12 months 3.3 installations of Nagios
Why do people choose Nagios?
- Visibility – see what is happening before the end users
- Techs like it because they can tweak it and customize it
- Businesses like it because it helps reduce down time
Why is Nagiso Important
- Capable of replacing expensive solutions
- Has a big community behind it, rare for open source
- SNMP Mibs were going to cost hundreds of thousands of dollars
Security checking
- DNS resolves to the correct IP
- Web site page hasn’t been high jacked
- Log Monitoring
- Port Scans
- Motion Sensors
Examples
- Paper thickness
- Paint spraying process
- Cash Registers
- Company started in 2007
- Weather
- You can monitor anything with a digital interface
Projects under Nagios
- New PHP Open Source GUI called V-Shell by Mike Guthrie
- Naigos Mobile – Web app interface into Nagios
- Nagios CCM – Core Config Manager, borrowed from the commercial version
- Nagios BPI – for both Nagios core and XI, visualize business processes, show hierarchy of business process
- Nagios Trap Manager – to use with SNMP front end
- Nagios Montage – Easy to use distribution, most requested parts of Nagios in one install package
NRDP is the replacement for NSCA
- Uses http encoded xml
- PHP Client, Perl Client
Integrated Matt Wall as the theme
NDOUtils – the database backend for Nagios,
Adding Non-blocking, Asynchronus writes, speed the database
Nagios Ceritifications
Nagios Professional and Administrator
NagiosXI Improvements
- Added auto discovery
- Distributed Monitoring
- Scheduled Reporting
- Nagios Exchange – Needs a rating and review section, stars, downloads
Mike Guthrie Distributed Monitoring
Understanding limitations
Core Engine schedules checks
Example Monitoring Server: 1000 hosts, 4 services per host, 5 min interval
If the number of checks scheduled is greater than the number the server can perform than the checks will fall behind. Check latency will go up.
Understand Logistics – plan before you implement
"I am not as concerned about what is working as what is not working"
Classic Distributed
- Proven, flexible
- But, Must maintain multiple configs, which server issued the check?
- Automate everything, Use Templates, Read Core Docs “Object Inheritance”
DNX
worked on by William Leibzon
Mod Gearman split the checks by host group
Very easy to setup, needs exact same directory structure for plugins as main nagios server.
Nagios Fusion
Commercial Separate NagiosXI servers with a top server that has an overview of what is setup on each of the servers, has unified login with drill down from top to individual servers.
Check MK
Part of a package of several free addons, central
Offload nodustils (DB backend) to a different computer
- Does this apply to NagiosXI?
Offload the performance graphs to a different computer
- This is not yet documented
Tuning Nagios
Baseline system
NDOUtils, 2 CPU, 20 GB Disk, 1 GB RAM
Use Memcache
Auto Discovery and Hyper map for our network
Business Activity Monitoring
- Dynamic Thresholds
- Standalone process that communicates with Nagios with NSCA
- Gforge.ingby.com/gf/project/bischeck
Eesensor
Device Address: Moved
Data String that is handed to the plugin : ZN111222TF: 73.9HU:41.7%IL:187.2
Plugins for Nagios: http://72.45.239.78:81/em01/Em01_Es11_nagios_Plugin_Linux.tar.gz
sensor sensors humidity temp