
UNSOLVED
VNX monitoring and reporting Metric threasholds
All,
Recently I have install VNX monitoring and reporting and setup alerting using the default values and was surprised at how many alerts I began to receive. The heat map displays a cool blue however many luns are reporting to frequently exceed the 20ms threshold and I receive alerts for Processor Dirty Pages. The VNX monitoring and reporting does not go into great detail around the alerts and what the actually represent and how to interpret them in order to take corrective actions.
The majority of the LUNs are servicing a VShpere environment with a small set of LUNs servicing SQL and Fileserver clusters. At this point we have not heard from the application owners of performance issues and would like to get a better understanding of these metrics in order to create thresholds that would indicate potential workload performance impact.
Detail Information on the following alerts, metrics and best practices for setting thresholds would be greatly appreciated.
Alerts
LUN Response Time
Storage Processor Dirty Pages Utilization
Storage Processor Utilization
Performance Bottleneck Metrics
Responses (1)
Solutions (0)

alkhvo
11 Posts
335
0
Posted January 23rd, 2014 06:00
Hi Wade001,
I have the same problem. I`ve checked Product Guide (which is available as Help at VNX Monitoring and reporting) and found that it is possible to select particular LUNs for response time alerting (as well as another alert definitions).
But there is only one instance of each alert definition. Another words I can select for examply 50 ms and apply it for specific LUNs. Thats all. I cant choose another response time for another LUNs.
Maybe next version will give us more flexible monitoring options.
Alex