I am being asked to monitor Nutanix devices in Check_mk.
Now I only have them configured with the ping check, but they need more details, do you know if there is an agent for those nutanix devices? or can it be configured by SNMP?
I found information on the Setup icon in Check_mk but none have worked for me.
The plugin mentioned by @cjcox is a little bit old and can have problems with 2.0.
Some months before i startet to write a CMK2 version of a Nutanix special agent with plugins.
Personally i would not try to use SNMP with all modern products. Nearly all the products have some type of API and should this should be the preferred way to query data.
@VIX one important question - what exactly from the huge Nutanix stack do you want to monitor. My plugin covers only the “normal” Prism API (hosts, vm, storage).
@Andreas : For now they only intend to monitor devices and servers.
Right now they are only monitoring the Ping.
I assume that what is in the GitHub link should be added to the plugins path on the Check_mk server?
Path: /usr/lib/check_mk_agent/plugins
You only need to install the “mkp” file with “cmk -P install nutanix_prism-2.1.mkp”.
Then you can configure the special agent inside the “Setup/WATO” menu.
Thanks for the information.
I have not been able to install the .mkp file as it tells me that the command is not found.
I add an image.
Could you help me with this issue, please?
The name of your site is “cmkadmin”? I don’t think so.
If you do anything inside your monitoring site the first thing you do is a su sitename
to change to the site user account.
Thanks for your help.
I was able to do the installation coen the user of the site.
Set up some rules from Wato but they haven’t worked.
I had the following configured in the properties: No API integrations, no Checkmk agent, but it is only with ping.
I modified it to (Configure API Integrations, not Checkmk agent) and added three checks, 1. check_mk, 2. check_mk discovery and 3. Prism Alerts.
1 and 2 are in critical status and 3 remains in PEND status.
I clicked on the TEST option and in the Agent section it throws an error, I attach it.
Do I need to do something else to make it work?
Do I have to add something in the .mkp file?
Hi Andreas, thanks for your assistance.
I have installed the new file and it still has errors, which I attach.
Do you think it’s the file again or do I have to do something on my side?
I did the test on a single Device and it worked, I have to monitor 65 more devices but when I modify the rule so that it hits all my folder called NUTANIX and not the test device, it does not work on the rest of the devices. It still works on the test.
I attached pictures of the errors.
AGENT ERROR:
API Error:Your request timed out after 110 seconds. This issue may be related to a local configuration problem or a request which works with a too large number of objects. But if you think this issue is a bug, please send a crash report.
This error is on the remaining 65 devices
I applied a Discovery and nothing changed. I modified the rule and instead of applying to the folder I only wrote the name of some devices but it didn’t work either, only in the deviuce of the test there have been no changes.
You 65 different Nutanix Prism server? The agent is only used against the single Prism server in your environment. All the data it gets is then stored as piggyback data for the found objects.
@Andreas: Thank you very much for your patience and help in this configuration.
Thanks to your help I have managed to monitor Nutanix devices in Check_mk.
After installing mkp just activate the rule from Check_mk, the rule is called Nutanix Prism.
In it, it asks you for data such as the port, a username and password, these data are provided by the person in charge of these devices on behalf of the client.
I directed them to my folder where I saved the devices and in the explicit hosts part I added those that are clustered and now in all of them I can see what they are monitoring.
On the part of the client that reviews Nutanix, I configure the user and password that I provide in each cluster
Hi Andreas,
Thanks for the wonderfull Nutanix MKP. It gives us much more info than the included one. There is one problem with it in our enviroment. We have four clusters. The clusters that don’t have any “Prism Alerts” are ok. But, as soon as a cluster has “Prism Alerts”, we get an crash for that check.