Solved

HA and DRS balance

Posted on 2015-02-02
14
93 Views
Last Modified: 2015-03-06
Have a cluster and both DRS and HA re enabled.
I forced one esxi host to enter into maintenance mode and noticed all the vm were migrated to other esxi host.
I rebooted the the esxi host after applying the patches.
I was expecting few vms to come back to this esxi host to balance the cluster, but it did not happen.
My assumption was that resources in the cluster will be balanced and vm will be motioned to balance the cluster if there is a requirement?
Does not it happen automatically, does it need manual intervention?
0
Comment
Question by:sara2000
[X]
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
  • 4
  • 4
  • 3
  • +1
14 Comments
 
LVL 28

Assisted Solution

by:asavener
asavener earned 143 total points
ID: 40584623
Depends on the DRS settings.  DRS can be configured to be manual or automatic, and the automatic settings can be configured.  Right-click the cluster, select Edit Settings, and then click on "VMware DRS."

Check your settings, and how aggressive the migration threshold is.  

If it is set to "manual" and you want to move some machines, you can left-click the cluster, select the DRS tab, and then select "Run DRS."  It will then give recommendations which you can choose to apply.

(If DRS is set to Automatic, then selecting "Run DRS" may result in machines being automatically vMotioned to other hosts.)
0
 
LVL 28

Expert Comment

by:asavener
ID: 40584635
And HA is for automatically restarting VMs when the host fails, so it will not get activated when a host gracefully enters maintenance mode.
0
 
LVL 121

Assisted Solution

by:Andrew Hancock (VMware vExpert / EE MVE^2)
Andrew Hancock (VMware vExpert / EE MVE^2) earned 213 total points
ID: 40584642
Give it a while, and eventually it will move VMs. (based on the number of VMs, and how heavy they are using resources!)

BUT DRS is not loading balancing, so do not expect it to be 50% CPU on Host A, 50 % CPU on Host B, and 50% Memory on Host A ot 50 % Memory on Hots B.

Also do you have it set to FULLY Automatic ?
0
Free learning courses: Active Directory Deep Dive

Get a firm grasp on your IT environment when you learn Active Directory best practices with Veeam! Watch all, or choose any amount, of this three-part webinar series to improve your skills. From the basics to virtualization and backup, we got you covered.

 
LVL 62

Assisted Solution

by:gheist
gheist earned 144 total points
ID: 40584997
DRS is polled every 5 minutes. You can force it with "RUN DRS" button in cluster DRS panel
Unless you set DRS to highest level no migrations will happen unless some host is under resource pressure.
0
 

Author Comment

by:sara2000
ID: 40585017
DRS is on  automatic. Strange one esxi host is always complaining (alerts) about host memory .
I did "RUN DRS" nothing happened? should it move VMs out of this esxi if this esxi  has memory issue?
0
 
LVL 62

Expert Comment

by:gheist
ID: 40585025
Can you post authentic host memory complaint?
And a probe in vcenter alerts that triggers it?
0
 
LVL 28

Expert Comment

by:asavener
ID: 40585109
Are you clearing the alert, and then it alerts again?  The alert icon will continue to be displayed until the alarm is cleared.
0
 
LVL 62

Accepted Solution

by:
gheist earned 144 total points
ID: 40585117
Would be nice to know it it is hardware alert or overcommit alert.....
0
 

Author Comment

by:sara2000
ID: 40606611
Gheist:
How do i find whether it is hardware alert or overcommit alert?
0
 
LVL 121

Assisted Solution

by:Andrew Hancock (VMware vExpert / EE MVE^2)
Andrew Hancock (VMware vExpert / EE MVE^2) earned 213 total points
ID: 40606677
Alerts Tab, at the root of your vCenter Server.

It will still be there, unless you've cleared it.

The alert will be triggered also on the host, a Red Alert on the host.

see here....

Alerts
The is a HA and DRS Cluster, and quite simply, there is *NOT* enough Memory in the hosts!

DRS and HA will not prevent this, they just move VMs around hosts!

You will still get memory warning's and cpu, if you do not have enough resources.

Time to either add more memory to ALL your hosts, or reduce the memory in use by your VMs if possible. But if it's all active, time to update your memory in all your hosts!
0
 
LVL 28

Assisted Solution

by:asavener
asavener earned 143 total points
ID: 40606722
Try selecting the host, and then going to the "hardware status" tab.  You should see any hardware errors there.

ESXi will inspect the BIOS log to see if there were any POST errors logged.  If it sees any (even if the errors have been resolved), it will generate an alert.  To make sure the error doesn't recur, you have to reboot, go into the BIOS, and clear the log.


That's if there's a hardware issue.
0
 

Author Comment

by:sara2000
ID: 40606791
I see very often on one esxi host as on Adrew's screen shot , not on all esxi hosts in the cluster.
I do clear that alert from time to time then it come back.
I was expecting a vm/vms should be moved out of that particular esxi to free out memory in both HA and DRS are enabled. It does not happen my case.
I am wondering why that esxi host get memory alert but vms are not moving ?
Does this means that other esxi hosts do not have enough memory to accept the vm?
0
 
LVL 62

Expert Comment

by:gheist
ID: 40606795
CAN YOU LOOK AT HARDWARE STATUS
Or you think we will poke around for another month?
0
 
LVL 121

Assisted Solution

by:Andrew Hancock (VMware vExpert / EE MVE^2)
Andrew Hancock (VMware vExpert / EE MVE^2) earned 213 total points
ID: 40606820
It's also happening, in in that screenshot. HA and DRS are enabled, a host is down, Hosts are alerting with memory issues, because both are at 95%-100%, and DRS has no where to more the VM - so what do we do....

Yes, it means you've ran out of resources, like I have in that cluster, because a machine has failed!

Add more resources, either another server, with CPU and Memory, or add more CPU or Memory to hosts.

What do you think the problem is here:- ?

Alerts1
and DRS is not moving my VMs, why?

and is this what you are observing, if you are clearing events, what action are you taking, if you just keep clearing them.......
0

Featured Post

Three Reasons Why Backup is Strategic

Backup is strategic to your business because your data is strategic to your business. Without backup, your business will fail. This white paper explains why it is vital for you to design and immediately execute a backup strategy to protect 100 percent of your data.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

In this article, I will show you HOW TO: Create your first Windows Virtual Machine on a VMware vSphere Hypervisor 6.5 (ESXi 6.5) Host Server, the Windows OS we will install is Windows Server 2016.
In this article, I will show you HOW TO: Install VMware Tools for Windows on a VMware Windows virtual machine on a VMware vSphere Hypervisor 6.5 (ESXi 6.5) Host Server, using the VMware Host Client. The virtual machine has Windows Server 2016 instal…
Teach the user how to rename, unmount, delete and upgrade VMFS datastores. Open vSphere Web Client: Rename VMFS and NFS datastores: Upgrade VMFS-3 volume to VMFS-5: Unmount VMFS datastore: Delete a VMFS datastore:
Teach the user how to convert virtaul disk file formats and how to rename virtual machine files on datastores. Open vSphere Web Client: Review VM disk settings: Migrate VM to new datastore with a thick provisioned (lazy zeroed) disk format: Rename a…

631 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question