vSphere 6.5 “Insufficient configured resources to satisfy the desired vSphere HA failover level on the cluster”

EDIT – Today (7/5), VMware GSS confirmed this is a bug that will be addressed in the next patch release.

Problem

Setting up a simple vSphere 6.5 (vCenter build 5178943) environment, using percentage-based HA we have been seeing the following Configuration Issue flag on the cluster (note, it is NOT an alarm!):

Insufficient configured resources to satisfy the desired vSphere HA failover level on the cluster

Screenshot:

You can see the alarms have Acknowledge/Reset options, this is a Configuration Issue message instead.

Possible Solutions

Doing some research I’ve collected the following things people said resolved this:

  • Turn on, or turn off Proactive HA and then disable HA on the cluster, then re-enable HA
  • Verify no firewall is interfering with HA
  • Use various methods to reset FDM on the hosts – disconnect/reconnect, move into another cluster, so on…
  • Set AC percentage to a different, low value (10% CPU/Memory)

In our case, for 3 clusters across 3 vCenters, none of those options worked.  Instead:

  • Disable “Override calculated failover capacity.”

That worked, as a workaround.  Unfortunately, setting ANY value (1%, 10%, 33%, 100%) for percentage based AC causes the config issue.  When I disable Override, the calculated value is 33% as expected.  So, it has to do with manually specifying a value, rather than the particular value specified.

Working Configuration, for a 3-node cluster:

This throws a configuration issue:

Hope that helps.

21 thoughts on “vSphere 6.5 “Insufficient configured resources to satisfy the desired vSphere HA failover level on the cluster”

  1. Just wanted to leave a comment to say that this helped me with my issue.

    Thanks for writing and updating this blog.

    1. VMware global support services confirmed the issue I described is a bug. Expect a resolution in some upcoming patch for vSphere (possibly vCenter, possibly ESXi, possibly both).

      1. Jeff. Can you please send me an email with the SR number. I got one of my customers with this problem and I am not able to find it

  2. In a v6.45 4 host cluster I had to increase “Host failures cluster tolerates” to 2 and then set it back to 1 to clear the warning.

  3. Can you please drop me an email with the SR number you filed. I noticed two people on VMTN Forum have the same problem, I cannot find a bug report for it and I want to make sure this particular scenario is actually covered. Thanks,

    Duncan Epping

  4. Thanks for this article, gave me a few new things to try. Disabling override is eventually what seems to have worked for the newly deployed 6.7 environment I was building.

  5. Hi,

    I was investigating this and the conclusion is the following:

    When we try to override failover capacity %; it’s needed to consider that it’s calculated automatically based on highest CPU and memory values of the host within the cluster, thus if we decrease that value lower than % calculate we are telling to AC that host capacity is less than 1 host which is the reason why the yellow banner “Insufficient configured resources to satisfy the desired vSphere HA failover level on the cluster”

    So, the conclusion here is that we can’t lower the percentage below the calculated % for less than 1 host without triggering the warning, because AC will not be able to tolerate less than 1 host failure. This now explains why when we unchecked the override failover capacity the warning goes.

    VMware is already working on creating a KB article which explains this expected behavior.

Leave a Reply

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

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

Recent Posts

Categories

Archives

GiottoPress by Enrique Chavez