An example of Nagios fault and solution under Ubuntu12.04

The scenario is as follows: OS: Ubuntu12.04LTSx86 _ 64; Nagios version 3.2.3; Nagiosql version 3.2.0; Pnp4nagios version 0.6.13; virtual machine, quad-core, 1.5 memory; Mode: passively monitors the number of hosts: 169 Monitoring Service Quantity: 745 drawing quantity: 923 (find/var/lib/pnp4nagios/perfdata-name "*. rrd "| w

The scenario is as follows:

OS: Ubuntu 12.04 LTS x86_64;

Nagios version 3.2.3;

Nagiosql version 3.2.0;

Pnp4nagios version 0.6.13;

Virtual Machine, quad-core, 1.5 memory;

Mode: passive monitoring

Number of monitored hosts: 169

Number of monitoring services: 745

Number of drawings: 923 (find/var/lib/pnp4nagios/perfdata-name "*. rrd" | wc-l)

Drawing mode: Synchronous mode

Fault symptom: continuous delay in alarms and monitoring, but the drawing is normal, and various tuning operations are still unsolved; after the pnp4nagios drawing is disabled, the monitoring and alarm will immediately return to normal; I personally feel like monitoring and plotting are competing for system resources;

Solution: Change the pnp4nagios plotting mode to bulk-npcd.

Briefly describe how pnp4nagios works:

First: synchronous mode Chinese: synchronous mode. In my personal understanding, as long as the data is received, drawing immediately, such:

Second: bulk mode noon: Container mode. I personally understand that it will cause a timeout when converting the data to be drawn first, reaching a certain amount, and then drawing. I heard that this method will process a large amount of data at the same time; it can be mitigated by modifying specific parameters. For example:

Third: bulk mode with npcd Chinese: bulk mode managed by npcd. My personal understanding: it is equivalent to the second type, but the author's solution is like this when npcd manages plotting. See:

Type 4: Bulk Mode with npcdmod; Use nagios to run npcd in a modular manner. However, it seems that the official website does not support nagios4, and I have not done many tests. For details, refer:

Type 5: Gearman Mode .. I don't know how to translate... looking at what the official saying is very good, we can make pnp4nagios and nagios distributed on different machines. I have not had a deep understanding of it. I am interested in visiting the official Internet cafe... see:

 

For the change method, see the pnp4nagios website.

 

Nagios details: Click here
Nagios: Click here

Reading:

Network Monitor Nagios full Raiders http://www.linuxidc.com/Linux/2013-07/87067.htm

Nagios construction and configuration http://www.linuxidc.com/Linux/2013-05/84848.htm

Build a Nagios monitoring platform http://www.linuxidc.com/Linux/2011-07/38112.htm in Nginx Environment

Configure the basic Nagios System (using Nagios-3.1.2) http://www.linuxidc.com/Linux/2011-07/38129.htm on RHEL5.3

CentOS 5.5 + Nginx + Nagios monitoring end and controlled end installation Configuration Guide http://www.linuxidc.com/Linux/2011-09/44018.htm

Ubuntu 13.10 Server install Nagios Core network monitoring using http://www.linuxidc.com/Linux/2013-11/93047.htm