Hi everyone,
Currently, To monitor my esx hosts services I am using Nagios Core 4.4.5 with check_esx3.pl script.
But, For some past days, the check_esx3.pl script is taking so much CPU utilization. And because of this our Nagios services are being affected so much. Also, Alert emails are generated properly. Due to check_esx3.pl script CPU consuming services are taking more time to reflect in Nagios.
So please help me ASAP to fix this trouble.
Thanks in advanceome tv omegle
check_esx3.pl Script problem?
Re: check_esx3.pl Script problem?
That's a pretty CPU intensive plugin in general. You may need to run it on a dedicated machine. One way to distribute your active check work is with mod-gearman:
https://exchange.nagios.org/directory/A ... an/details
But you could just as easily spin up a dedicated VM and execute this plugin with a generic agent like NRPE or NCPA. The main benefit of mod-gearman is only realized if you're wanting to distribute an awful lot (like tens of thousands) of active checks.
https://exchange.nagios.org/directory/A ... an/details
But you could just as easily spin up a dedicated VM and execute this plugin with a generic agent like NRPE or NCPA. The main benefit of mod-gearman is only realized if you're wanting to distribute an awful lot (like tens of thousands) of active checks.
Former Nagios employee
https://www.mcapra.com/
https://www.mcapra.com/
Re: check_esx3.pl Script problem?
The VMware Perl SDK which is what the check_esx3.pl (which is an install requirement) uses to do most of the work is a beast when it comes to resources and is what is sucking up the CPU.
Current SDK
https://code.vmware.com/web/sdk/7.0/vsphere-perl
When monitoring large VMware environments having a larger resource Nagios instance including worker nodes to do most of the heavy lifting is a good practice to follow.
One possible solution is to switch to using one of the many VMware APIs to get your metric data and overcome the resource costs. There are a couple of libraries available on the Nagios Exchange that may help you out.
Current SDK
https://code.vmware.com/web/sdk/7.0/vsphere-perl
When monitoring large VMware environments having a larger resource Nagios instance including worker nodes to do most of the heavy lifting is a good practice to follow.
One possible solution is to switch to using one of the many VMware APIs to get your metric data and overcome the resource costs. There are a couple of libraries available on the Nagios Exchange that may help you out.