VMware vCenter Server indicates that the Dynamic Resource Scheduler (DRS) usage load is imbalanced

Symptoms

• Dynamic Resource Scheduler (DRS) usage load is not evenly balanced across hosts.
• Some hosts within a DRS Cluster show significantly higher CPU and/or memory usage compared to other hosts in the cluster.
• DRS resource distribution charts indicate that load is not distributed evenly across the hosts in the cluster.

Cause

The DRS process may decide not to initiate migrations of virtual machines to other hosts within the cluster if the cost involved outweighs the benefit obtained. Load balancing migrations have a cost but no additional benefit, if the required resources are available to virtual machines running on the host.

Resolution

The primary goal of the DRS is to provide virtual machines within a cluster with access to the required resources. Dynamic Resource Scheduler (DRS) is not designed to evenly balance resource usage across hosts in the cluster. 

High memory or CPU utilization of an individual host is not a sufficient reason to migrate virtual machines if the host is able to provide 100% of the resources required for its running virtual machines. This cluster imbalance can be safely ignored, if there is no observed performance degradation in virtual machines.

If a host within the DRS cluster is unable to provide a virtual machine with the resources it requires, then DRS migrates the virtual machine to a host that can provide the required resources.

The vpxd process becomes unresponsive after upgrading to vCenter Server 5.5

Symptoms

• The vpxd.exe process continually fails.
• You are unable to log in to vCenter Server.
• Windows Service Manager and service.msc show VMware VirtualCenter Server service as Starting.
• In Windows Task Manager, the vpxd process continually recycles.
• A 0 or 1 byte vpxd.dmp file is generated in the vCenter Server log directory, located at C:\ProgramData\VMware\VMware VirtualCenter\Logs.
• In the C:\ProgramData\VMware\VMware VirtualCenter\vpxd.log file, you see entries similar to:
  
  <YYYY-MM-DD>T<TIME> [16900 verbose 'VpxProfiler'] [1+] [ORM] Load: vim.vm.ConfigInfo, Id: 361
  <YYYY-MM-DD>T<TIME> [16900 verbose 'VpxProfiler'] [2+] [BulkLoadDbContent]
  <YYYY-MM-DD>T<TIME> [09124 trivia 'MoLock' opID=SWI-ba59244] Waiting on vm-361 (mode: EXCLUSIVE)
  <YYYY-MM-DD>T<TIME>[06624 warning 'utilvpxdDbLoad'] [ORM] Load: Missing mapping for property hardware.device.backing.backingObjectId 
  <YYYY-MM-DD>T<TIME> [06624 warning 'utilvpxdDbLoad'] [ORM] Load: Missing mapping for property hardware.device.backing.backingObjectId 
  <YYYY-MM-DD>T<TIME> [06624 info 'Default'] CoreDump: Writing minidump
  Note: The preceding log excerpts are only examples. Date, time, and environmental variables may vary depending on your environment. 

Purpose

Vpxd process being unresponsive after upgrading to vCenter Server 5.5 is a known issue affecting vCenter Server 5.5.

Cause

This issue occurs due to the thread stack size changing in vCenter Server 5.5.  When the VMware VirtualCenter Server service performs a validation on its Inventory data, there may be excessive snapshots that over running the smaller stack.

Resolution

This is a known issue affecting VMware vCenter Server 5.5. 

Currently, there is no resolution.

An alarm can be created to alert if virtual machines in the environment are running on a snapshot. For more information, see Configuring VMware vCenter Server to send alarms when virtual machines are running from snapshots (1018029).

To work around this issue, increase the threadstack size in the vpxd.cfg file:

1. Stop the vCenter Server service. For more information, see Stopping, starting, or restarting vCenter Server services (1003895).
2. From the vCenter Server, navigate to C:\ProgramData\VMware\VMware VirtualCenter\.
3. Open the vpxd.cfg file using a text editor.
4. Change the number in <ThreadStackKb></ThreadStackKb> to 1024 or add <ThreadStackSizeKb>1024</ThreadStackSizeKb> under <threadPool>:
   
   Caution: Entries within the vpxd.cfg file are case sensitive.
   
   For example:
   
   
   <vmacore>
   <threadPool>
   <TaskMax>90</TaskMax>
   <ThreadNamePrefix>vpxd</ThreadNamePrefix>
   <ThreadStackSizeKb>1024</ThreadStackSizeKb>
   </threadPool>
   <ssl>
   <useCompression>true</useCompresion>
   </ssl>
   </vmacore>
5. Save the changes and close the editor window.
6. Start the VMware VirtualCenter Server service .

Note: In some cases, the vpxd process may fail without creating a dump file or creating a 0 byte file. If you experience this issue, try one of these options: 

• Enable the User-Mode dumps of the application. For more information, see the Microsoft article Collecting User-Mode Dumps.
• Use the ADPlus tool from Microsoft. For more information, see Microsoft Knowledge Base article 286350.
  
  Note: The preceding link were correct as of June 25, 2014. If you find a link is broken, provide feedback and a VMware employee will update the link.

How to check number of cpu cores ,sockets, threads in a linux server

[root@linuxThrill ~]# lscpu
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                96
On-line CPU(s) list:   0-95
Thread(s) per core:    2
Core(s) per socket:    12
Socket(s):             4
NUMA node(s):          4
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 63
Stepping:              4
CPU MHz:               2094.895
BogoMIPS:              4189.42
Virtualization:        VT-x
L1d cache:             32K
L1i cache:             32K
L2 cache:              256K
L3 cache:              30720K
NUMA node0 CPU(s):     0-11,48-59
NUMA node1 CPU(s):     12-23,60-71
NUMA node2 CPU(s):     24-35,72-83
NUMA node3 CPU(s):     36-47,84-95
[root@dh-location-analytics1 ~]#  dmidecode -t processor | grep -i cpu
Version: Intel(R) Xeon(R) CPU E7-4830 v3 @ 2.10GHz
Version: Intel(R) Xeon(R) CPU E7-4830 v3 @ 2.10GHz
Version: Intel(R) Xeon(R) CPU E7-4830 v3 @ 2.10GHz
Version: Intel(R) Xeon(R) CPU E7-4830 v3 @ 2.10GHz

No of Sockets : 4
Core per socket: 12
Theards per core : 2
After enabling Hyper threads : 96.