Host vmnic down Alert and Supermetric

Use Case 

To be able to get information on which vmnic is down rather than generic vmnic down alert. 

 

Solution: 
We can create super metrics for each vmnic on the host such that if the vmnic is down then the value of this super metric is 1. 
We can now create symptoms for each vmnic if they are down. then create an alert with these symptoms with OR condition. This alert will be triggered if any vmnic is down with the name of the vmnic. 
Then, you may disable the original alert for vmnic monitoring in policy which is creating  alerts without actual vmnic names. 
Example of super metric: 
This SM will be 1 if vmnic4 is down and 0 if it's either up or doesn't exist: 
count(${this, metric=net:vmnic4|nicDriver}) && !count(${this, metric=net:vmnic4|speed}) 

Now the question is how many super metrics to create on each host. one idea is to create report to see the maximum count of vmnic used in any host and use that many SMs on each host just so that we don't miss on any vmnic on any host


Sign in to be able to add comments.

Comments 0