Hi,
I'm using NPM to monitor our hosts in the cloud. I'm monitoring the vCenter and i noticed that sometimes it shows 2 alerts that are not true:
1. Host is down - even if the host is online and pinging fine with out inter ups. also no error on the vCenter regarding loss of connectivity or something similar
2. it shows wrong memory usage on hosts. it shows that hosts are on 100% memory usage and stays on 100% when they are 90-95 percent. I'm comparing it to the vCenter.
when i try to see polling resources one of thos hosts it gives me a message that the host is down for some reason when it's not ( happens only on nodes that faking the memory usage )
all other resources on the nodes and vCenter are shown correctly.
anyone knows what can be the issue of that? i know that a lot can cause that but for now i couldn't find any reason for that.
I tried re-adding them and still nothing. When i'm trying to pull it finishes successfully.. so frustrated 
thanks,