When would an administrator configure the Explicit Failover option for a portgroup created on a virtual switch

When would an administrator configure the Explicit Failover option for a portgroup created on a virtual switch with multiple uplinks?

A.
To manually configure load balancing for the portgroup

B.
To share uplinks between vSwitches

C.
When configuring iSCSI Software Initiator multipathing

D.
If the physical switch does not support etherchannel or 802.3ad

18 Comments on “When would an administrator configure the Explicit Failover option for a portgroup created on a virtual switch

  1. baher says:

    Multi-pathing with iSCSI is supported both on Hardware and Software iSCSI initiators. So, C is the correct answer as in this case it is addressing multiple NICs.

  2. Tony says:

    Use explicit failover order = Always use the highest order uplink from the list of Active adapters that passes failover detection criteria.

    So this is not true multipathing, only one link is used at a time. By elmination, A seems to be the right answer. All other load balancing policies are autmoatic, this one is manual:

    http://pubs.vmware.com/vsphere-50/index.jsp?topic=%2Fcom.vmware.vsphere.networking.doc_50%2FGUID-D5EA6315-5DCD-463E-A701-B3D8D9250FB5.html

  3. DCW says:

    I would say C based on Tony’s logic…

    “So this is not true multipathing, only one link is used at a time”, but A states “manually configure load balancing”, and using 1 link at a time is defiantly NOT load balancing….

  4. Pavel says:

    I think A is a right answer.
    For iscsi software multupathing the right thing to configure is override switch and different unused nic for portgroups

  5. Simon Mutuku says:

    The correct Answer is C.
    Explicit Failover option for a portgroup is used when configuring iSCSI Software Initiator multipathing……. T…R…U…E
    …B…U…T…Am not sure about the Multipathing bit since only one Nic/ link is used at any one time……..
    …….Perhaps the use of the word “Multipathing” shouldn’t be included in the answer.

    I.e Nic A & B can’t be accessed at the same time.

  6. karloce says:

    Seems like A is correct, please comment if you can explain:

    Use explicit failover order :Always use the highest order uplink from the list of Active adapters that passes failover detection criteria.

    Set Failover Order to specify how to distribute the work load for adapters.

    To use some adapters but reserve others for emergencies, you can set this condition using the drop-down menu to place them into groups.

    Active Adapters :
    Continue to use the adapter when the network adapter connectivity is available and active.

    Standby Adapters:
    Use this adapter if one of the active adapter’s connectivity is unavailable.

    Unused Adapters:
    Do not use this adapter.

    If you are using iSCSI Multipathing, your VMkernel interface must be configured to have one active adapter and no standby adapters. See the vSphere Storage documentation.

  7. moors says:

    The correct answer is C

    With the Explicit Failover Order load balancing algorithm in effect, you are essentially not load balancing at all! Explicit failover will utilize, for all traffic, the “highest order” uplink from the list of Active pNICs that passes the “I’m alive” test. What does the “highest order” mean? Well, it’s simply the pNIC that has been up the longest!

  8. jackal99a says:

    I do iscsi and benoit is right – supported config for iSCSI is one per kernel and you setup multipathing in the software iscsi adapter.

  9. Ytsejamer1 says:

    Benoit and Jackal99a’s answers would point you to answer C being correct. It’s kind of a bad question because you can override at any point depending on your intended use.

    In this case…when you are configuring iSCSI multipathing, you MUST override the defaults and ensure you take out any standby/active adapters and set them to unused. No ifs or buts about it…you HAVE to override it when configuring iSCSI multipathing (as indicated by the links posted by Benoit).

    It’s simply optional when you just want to override it for any other reason…so the more correct answer to me is C – When configuring iSCSI Software Initiator multipathing.

    1. Arie says:

      Ytsejamer1: “Benoit and Jackal99a’s answers would point you to answer C being correct.”

      If you would have read properly Ytsejamer1, you would have seen that Benoit stated that: “the correct answer is A”. I also believe that A is the correct answer.

  10. Rich says:

    iSCSI port binding configuration requires explicit failover selected AND round robin. Since Round robin is a multipathing protocol, iSCSI port binding IS a form of multiPathing.

    The thing that I don’t like is that they use the term ‘Port Group’, not ‘VMKernel port’. This would indicate that A is correct.

  11. Ali says:

    The last option, explicit failover order, isn’t really a “load-balancing” policy; instead, it uses the user-specific failover order. More information on the failover order is provided in the section “Configuring Failover Detection and Failover Policy.”

    Answer C


Leave a Reply to Ali Cancel reply

Your email address will not be published. Required fields are marked *