Jump to content
Welcome to our new Citrix community!
  • 0

Extremely slow I/O on all guest VMs when running mysql import on a VM.


Luc Lezon

Question

Hi guys and girls, 

 

We've seen this issue quite a few times now and I'm wondering if there is a way to either fix it with a patch or tell Xen to not give as much I/O resources to a single VM.

 

Our setup:

 

* Xen 7.0 (patched, all updates). We need to use this version for legacy reasons. Not in pool. 

* 6x 3.84TB SSD in RAID10. Local storage.

* Dell R620 with 2x 2.8GHz CPU and 512GB RAM

* 2 CentOS 6.x VMs installed from template and updated (VM1 and VM2)

* VM1 and VM2 are both idle with minimal traffic, CPU and I/O

 

When we try to import a large mysql dump on VM2, after VM2 runs out of RAM which mysql uses to setup the indexes, I/O seriously begins to effect VM1 to the point where a regular INSERT/SELECT statement in mysql takes about 1-2 minutes and queries start to stack quickly. The problem goes away immediately after the VM2 mysql import is aborted. 

 

We've seen this on quite a few servers running different RAID configurations and different drives types (ex: SAS mechanical and SAS SSD). 

 

It appears that Xen is allocating most of the read/write to a single VM and that VM is using up most of the resources preventing the other guest VMs from reading/writing data in a reasonable time. 

 

So my question is... Is there a way to fix this? Is there a patch? Or a way to tell Xen to use only 10% IO for this VM and 90% for the rest?

 

Thanks!

Joe

 

Link to comment

2 answers to this question

Recommended Posts

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...