Disabling and re-enabling HA on the cluster will probably fix the issue.
See here: http://kb.vmware.com/kb/1021173
And you should do that ASAP, because NOW your VMs are at risk already, because HA is currently not working properly.
Turning HA off and on will not affect the VMs in any way except from putting them at the risk of not being restarted automatically in case of a host failure.
- Andreas