VMWare low level discovery

[mcdougak]

Ok, so I'm not sure if I'm just misunderstanding or missing something but here goes ...

Documentation says I can setup a low level discovery rule for my VMWare Hosts and provides the following image as a reference ...

https://www.zabbix.com/documentation/_detail/2.2/manual/vm_monitoring/vm_hypervisor_lld.png?id=2.2%3Amanual%3Avm_monitoring

When I try to create a new discovery rule I do not see the majority of the options, I only have the basic: Name, Discovery by proxy, IP range, Delay, Checks, Device uniqueness critera and enabled.

This is a new clean install on Ubuntu of 2.2.

[tchjts1]

I am in the process of bringing up a 2.2.1 test instance today. I will let you know what I see.

[mcdougak]

Thank you for your help. I was finally able to find what I was looking for. The user guide or manual is not very clear at time where some of the items are in relation to the various menus.

I found that what I had to do first was create my host (under configuration / hosts) , then I was able to add a discovery rule.

I was assuming this low level discovery option was under Configuration / Discovery, and I was attempting to create a rule there.

I'm sure I'll come up with more questions along the way.

Right now I'm dealing with performance issues loading the Latest Data section. When all groups are selected the page waits about 30 seconds and then loads a blank page ... but it's fast (about 1-5 sec.) if I just select one server to look at. I've been reading up on the php tuning and have made some changes but I'm thinking my VM may be under resourced.

I just added a second VCPU and it currently has 2 GB of RAM. The status reports I have 227 hosts monitored and 7296 items on those hosts monitored with a required server performance of 140.97 new values per second.

I'm working on enabling the server monitoring to provide additional information.

[tchjts1]

... and it currently has 2 GB of RAM. .

I would suspect that is the bottleneck right there. If you have both the App server and DB server on the same VM, and only 2GB of RAM for both, you're going to run into performance issues.

Best to split them off to their own VM's and give that DB some breathing room in the RAM department. You have a rather small number of hosts you are monitoring, but I would personally not go any lower than 8GB of RAM, and I would prefer 16GB.

This is my memory usage on my DB server for our 2.0.9 setup with ~ 600 hosts and ~ 900NVPS (Ideally you won't see any swap being used, either)

You can see my experience on swap usage here: https://www.zabbix.com/forum/showthread.php?t=38575