VMWARE 6.7 : Nagios Core vs check_mk

[emi65]

Hi expert

I'm using nagios core 4.4.5
I like to monitor esx server (VMWARE 6.7)
To implement these check I have to install on nagios server a lot of packages

1. install module Time::Piece
2. install module UUID
3. upgrade module Net::HTTP
4. upgrade module Net::INET6Glue
5. upgrade module Text::Template
6. upgrade module Socket6
7. upgrade module Try::Tiny
8. upgrade module Path::Class
9. upgrade module XML::NamespaceSupport
10. upgrade module ExtUtils::MakeMaker
11. upgrade module Exporter
12. upgrade module Test::Pod
13. upgrade module Test::Pod::Coverage
14. upgrade module Test::TrailingSpace
15. upgrade module LWP::UserAgent
16. upgrade module LWP::Protocol::https
• Copy and install file VMware-vSphere-Perl-SDK-6.5.0-7397310.x86_64.tar.gz
• tar -xzvf VMware-vSphere-Perl-SDK-6.5.0-7397310.x86_64.tar.gz
• cd vmware-vsphere-cli-distrib
• ./vmware-install.pl

.. at the end I can use check_vmware_api.pl

use set the same check in check_mk solution I don't need to install VMware-vSphere-Perl-SDK-6.5.0-7397310.x86_64.tar.gz because
the API is already integrated

Is the a simply solution in nagios in order to monitoring ESX server ?
Thanks and regards
Emilio

[gormank]

I've worked the SDK install on Nagios down to a few perl modules and the SDK, which isn't bad. It also only needs to be done when the Nagios hosts are created. I don't know of any other solution and I don't know much about check_mk at all.
You can make the SDK install non-interactive with some arguments when running the install.
As far as monitoring ESX servers, you can simplify that by monitoring via vcenter, which for one thing means you don't have to add a user on each ESX host--only a user on vcenter.

[emi65]

Thanks Gormank

If I want to follow the way to monitor the ESX server using vcenter which check Do I need to use ?

Thanks
Emilio Frini

[gormank]

I use check_vmware_api.pl but the arguments are a bit different. You need to slog through the -h output to decide which arguments to use.

[emi65]

Thanks Gormank !!!

[gormank]

Here are some checks from .cfg files with args. It may help some...

My check_esx3_host command is this:
$USER1$/check_vmware_api.pl -H $HOSTNAME$ $ARG1$ $ARG2$ $ARG3$ $ARG4$ $ARG5$

ESXi_Host_CPU.cfg: check_esx3_host!-u $USER9$ -p '$USER10$'!--command=cpu!--subcommand=usage!-w 50 -c 80 -D $USER11$ -C CMV!--timeout=15!!!add --subcommand= to arg3 for subcommands
ESXi_Host_Datastore_Usage.cfg: check_vmware_esx!-u $USER9$ -p $USER10$!-S volumes!-w 20% -c 10%!--datacenter=$USER11$ --cluster=CMV!--timeout=15 --nosession!!!
ESXi_Host_IO.cfg: check_esx3_host!-u $USER9$ -p $USER10$!--command=IO!!-D $USER11$ -C CMV!--timeout=15!!!
ESXi_Host_Issues.cfg: check_esx3_host!-u $USER9$ -p $USER10$!--command=RUNTIME!-s issues!-D $USER11$ -C CMV!--timeout=15!!!
ESXi_Host_Memory.cfg: check_esx3_host!-u $USER9$ -p $USER10$!--command=MEM!-w 80 -c 90!-D $USER11$ -C CMV!--timeout=15!!!
ESXi_Host_Networking.cfg: check_esx3_host!-u $USER9$ -p $USER10$!--command=NET!!-D $USER11$ -C CMV!--timeout=15!!!
ESXi_Host_Services.cfg: check_esx3_host!-u $USER9$ -p $USER10$!--command=SERVICE!!-D $USER11$ -C CMV!--timeout=15!!!
ESXi_Host_Status.cfg: check_esx3_host!-u $USER9$ -p $USER10$!--command=RUNTIME!!-D $USER11$ -C CMV!--timeout=15!!!

[martinhirschvogel]

Seems like you already pointed out the simple solution. In Checkmk, you don't need to install anything. It connects directly to vSphere to get the data.

[emi65]

I choose a simple solution

I monitor the ESX machine using CheckMK

I include che output of the checkMK interface in Nagios using a bash script based to lq command (api included to checkMK)

thanks
Emilio

[gormank]

I'm not sure creating another monitoring system is simpler than running a few commands one time but I'm not familiar with checkmk.