6 Responses

  1. Newsletter: July 20, 2014 | Notes from MWhite

    […] Packet loss at Guest OS Level in ESXi when using VMXNET3 Michael wrote about this recently in this.  If you have a busy server and there are drops in the network you may need to make a change in […]

  2. Karl Wilson
    Karl Wilson at |

    I ran into a this issue when virtualizing our Exchange 2010 Environment, even adjusting the Ring Buffers with support reduced, but did not remove the large packet loss. We had hoped the previous issues with the vmxnet 3 adapater had been resolved, but alas was not the case.

    In the end we had to move back towards a e1000 adapater

    Reply
    1. Cliff
      Cliff at |

      Karl we had the same problem with our enviorment with both the Exchange and Domain Controllers. Even with the buffer change we are still seeing problems. We were worried about moving back to the e1000 with all of the other problems that were happening during that time. http://kb.vmware.com/selfservice/microsites/searc

      Reply
  3. Dominik
    Dominik at |

    Hi Michael, did you have the same problem with Linux OS (Rhel, CentOS)? Thanks

    Reply
  4. » VMware vSphere 5.5 Virtual Network Adapter Performance Long White Virtual Clouds

    […] out on top, this is why VMware made it a best practice to use VMXNET3. Even though you may have to adjust some settings to get optimal performance it is worthwhile using VMXNET3 as the default. This begs the question, why is VMXNET3 not the […]

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.