Unsolved
This post is more than 5 years old
2 Posts
0
1088
disk letancy in VNX
how to check disk letancy in VNX7600 and ,and how it co-relate the response time.
Is there is any specific command to check the same things with the help of cli.
I want to monitor the VNX through command line with real time analysis.
Regards
Manoj -Infra specilist
Roger_Wu
2 Intern
2 Intern
•
4K Posts
0
October 16th, 2014 21:00
You can use Unisphere Performance Analyzer to get such information. However if your device doesn't have its Enabler installed, you can only gather the performancel log files but can not open it.
You can also take a look at VNX Monitoring and Reporting which is free for VNX. But I'm not sure whether it has disk latency info.
VNX Monitoring and Reporting
manojrajput22
2 Posts
0
October 16th, 2014 22:00
In M&R there is no disk letancy type of features,also performance analyzer and logs is not giving the exact figure of storage bottleneck in realtime.So my concern related to 1 sec basis output where i analyze disk as well as other bottleneck issue which can be better server in cli than GUI
Joshua_Fawcett
35 Posts
0
October 22nd, 2014 10:00
When you say disk latency are you talking about latency on the actual backend hard drives on your VNX array or are you talking about the LUNs?
There are not really command line commands that can be used to evaluate performance on the VNX at this time. It is mainly all done through M&R or Unisphere Analyzer.
JonK1
247 Posts
0
October 24th, 2014 04:00
Manoj, you can use Analyzer and adjust the realtime/archive polling intervals down from the standard 300s interval. Be aware though that more frequent polling also increases the performance impact of the performance monitoring. 60s intervals work; i have never driven it down further.
For third party products you could use SMI-S in combination with a number of tools: SolarWinds Storage Manager, vCOPs, etc...
Polling every second is unpractical: you'll end up with an immense amount of data with limited to no use. If you suspect your servers to have storage bottlenecks, monitor the response times from the server perspective; it's easier to adjust the polling interval to 5s in perfmon.