1 Reply Latest reply on Jul 4, 2018 6:48 AM by vswitchzero

    Split-Brain MP Enhancement

    RenxiaoChen Lurker

      NSX for vSphere 6.4.0 is characterized by the management plane (MP) playing a more active role and taking a corrective action in split-brain scenarios,

       

      Two new commands have been introduced, which the MP uses to take a corrective action:

      forceStandby

      forceUpdate

       

      I want to know , how can I use these commands ? In witch UI Interface ?

        • 1. Re: Split-Brain MP Enhancement
          vswitchzero Enthusiast
          vExpert

          Hi RenxiaoChen,

           

          Just out of curiosity, where did you read about these two commands? It's possible that some VMware documentation may need to be corrected or put into the correct context.

           

          From what I can see, these are not actually user accessible commands, but rather actions the management plane can execute in an automated manner. If a split-brain scenario is detected, the management plane - i.e. NSX Manager - can set a flag called 'ForceStandby' to true. This essentially forces a specific node to move from the HA active role to standby. Once the MP detects that the split brain situation has subsided, it can then clear the flag.

           

          From a CLI perspective, the only place I've seen where this is visible is the 'show service highavailability' command output on an HA-enabled edge:

           

          ESG01-Pri-0> show service highavailability

          Highavailability Service:

          Highavailability Status:             Active

          Highavailability State since:        2018-07-04 13:10:40.682

          Highavailability Unit Id:            0

          Highavailability Unit State:         Up

          Highavailability Admin State:        Up

          Highavailability Running Nodes:      0, 1

          Unit Poll Policy:

              Frequency:                       3.75  seconds

              Deadtime:                        15    seconds

          Highavailability Services Status:

              Healthcheck Config Channel:      Up

              Healthcheck Status Channel:      Up

          Highavailability Healthcheck Status:

              This unit [0]: Up   Active: 1

              Peer unit [1]: Up   Active: 0

                  Session via vNic_1: 169.254.1.53:169.254.1.54 Up

          Config Engine:

              HA Configuration:               Enabled

              HA Admin State:                 Up

              Config Engine Status:           Active

              ForceStandby Flag:              Disabled   <--------

          Highavailability Stateful Logical Status:

              File-Sync                       running

              Connection-Sync                 running

                xmit       xerr       rcv        rerr

                30808      0          29592      0

           

          Hope this helps.

          My blog: https://vswitchzero.com
          Follow me on Twitter: @vswitchzero