Community
- Command Central
- MVP Program
- Monthly Mission
- Blogs
- Groups
- Events
- Media Vault
Products
- Observability
- Network Management
- Application Management
- IT Security
- IT Service Management
- System Management
- Database Management
Content Exchange
- SolarWinds Platform
- Server & Application Monitor
- Database Performance Analyzer
- Server Configuration Monitor
- Network Performance Monitor
- Network Configuration Manager
- SQL Sentry
- Web Help Desk
Free Tools & Trials
Store

How to monitor disk seconds per read ?

Will the below WMI Monitor collect the average disk seconds per read off all drives on a cluster? or do I need to set the monitor for each drive if I want to alert from the following

If > 100 MS 2 times out of 2 in 2 minutes?

select AvgDisksecPerRead from Win32_PerfFormattedData_PerfDisk_PhysicalDisk

Really don't want to have to do each drive but if that is what I need to do then I will just trying to save some time by not selecting a Name ="X" in the monitor.

Thanks in advance

Find more posts tagged with

Accepted answers

All comments

profzoom1

Any help on this?

I am being told by my SQL DBA's that the SQL Page Reads per Second and SQL Page Writes per Second in the SQL Server 2005 Database Template in APM 3.0 are not anything to worry about since those can burst and go way above the MS recommended statistics in the alert.

What they want are disk seconds per read/write which I am able to get from select AvgDisksecPerRead from Win32_PerfFormattedData_PerfDisk_PhysicalDisk.

Question is do I need to create this WMI Monitor for Each logical drive or will this alert on any of the logicals going above a certain number.

When using the above WMI monitor on a busy SQL server I am returning numbers that are like 5824572 to -482478 while within the OS when running the monitor on the logical drive it returns actual seconds is there a calculation that needs to be done to get the data into something that can be read easily?

ShamChauthani

I just looked up the the PhysicalDisk class on one of my test machines and found that as long as you are using the PerfFormattedData class then you should get the "cooked" values in milliseconds. On most of my tests I saw that values to be 0.

On the machine where you are getting those values, could you run WBEMTEST and run the query you are using in APM and see what values you are getting. Please double check that you are using the FormattedData and not the RawData counters.

That said you should be able to monitor the _Total instance and get the average across all drives on the machine. That approach looses some granularity because if you have a lot of drives a spike in access time on one of the drives will get averaged out across all other drives and may not trigger the alert you are looking for, but it can be done.