MSPSS: is there life after the helpdesk?

sharing solutions to uncommon IT problems

Windows 2008 R2, Vmware and Driver Locked memory

with 4 comments


Hello,
we have had serious performance issues on a server for some time now.
This server is heavily used and we always assumed the problems we were having were simply due to the coexistence of 3 memory and CPU hungry applications on the same box (SQL Server, IIS and Coldfusion).
We tried everything to tweak resources so that one process would not overcome the others.
Nothing seemed to work.
One thing that always bugged us was that the physical memory would go rapidly exhausted although we had increased the memory progressively to 16Gb.
What’s even more strange is that the sum of the memory of each process was far from the “Physical Memory in use” we saw in Task Manager (or process explorer).
I also tried to limit memory usage for both Coldfusion and SQL Server to 5Gb each with no success.
Then I found this application from sysinternals (M. Russinovich rocks): RAMMap
This app gives you a detailed overview of the memory allocation and strangely I had 10Gb allocated to “Driver Locked”.
Google is my friend and I found out that this is usually due to a misconfiguration of VMware.
Apparently in VMware there are two config for each resource:
1. The Hardware configuration: the actual configuration of the VM where you define the physical resources allocated (in my case I had 16Gb for RAM).
2. The resource allocations: here you can play with the VMWare resources to limit the resources configured in point 1.
I found out that in the resource allocations tab we had limited memory to 4096Mb which means that for as much as we added memory to the VM, the available memory for that machine would be always the same and the rest would be “locked” in the vmware tools driver.
I opened that tab (you’ll find it in the VM settings) and selected unlimited and now task manager shows 13Gb of available memory.

I hope this helps,
Roberto.

Advertisements

Written by zantoro

September 15, 2012 at 2:04 pm

4 Responses

Subscribe to comments with RSS.

  1. This was happening on one of our sql clusters. Thanks!

    andrew

    December 12, 2012 at 6:02 pm

  2. Thank you for this very valuable post. It helped us find the cause of a problem here at work.

    Patrice Guay

    December 12, 2012 at 8:11 pm

  3. Life Saver!!
    Saw this in RamMap and was continually scratching my head trying to find the culprit.
    Was scratching my heads continuously and see the the original maintainer of the server had limited its resources.

    Thanks for the share 🙂

    A.Grobb

    January 21, 2015 at 9:47 pm

  4. Thank you. I had a production system down over this issue. 25 people siting on there hands. This fixed it and put them back to work. Thank you again.

    rutrow

    July 28, 2016 at 10:13 pm


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: