Error: "Unexpected VMware Cloud Director error. [ UI-####... ] Invalid UUID string: null" when migrating a VM

Products

• VMware Cloud Director
• VMware Live Recovery

Issue/Introduction

• Test failover, failover or migration fails for a VM in VMware Cloud Director Availability (VCDA)
  
  
• In the UI you get an error when trying to Test failover, failover or migrae the replication
  
  "Unexpected VMware Cloud Director error. [ UI-####... ] Invalid UUID string:null"
  
  
• On the UI in the section "topology" you can see the Datastore for the replication is N/A
  
  
• On the destination site VCDA Manager Appliance in /opt/vmware/h4/cloud/cloud.log you see entries similar to:
  
  ERROR - [UI-########-####-####-####-############-######-##-###] [job-87] com.vmware.h4.jobengine.JobExecution     : Task ########-####-####-####-############ (WorkflowInfo{type='failoverTest', resourceType='vmReplication', resourceId='########-####-####-####-############', isPrivate=false, resourceName='vapp-name'}) has failed
com.vmware.vcloud.client.exception.VcloudException: (Major code = 500, minor code = INTERNAL_SERVER_ERROR) - [ UI-########-####-####-####-############-######-##-### ] Invalid UUID string: null
        at com.vmware.vcloud.client.VcloudClient.lambda$defaultErrorDeserializer$0(VcloudClient.java:301)
...
DEBUG - [UI-########-####-####-####-############-######-##-###] [job-87] c.v.h.c.c.VcloudExceptionConverter       : Received a vCD error - message = [ UI-########-####-####-####-############-######-##-### ] Invalid UUID string: null, majorCode = 500, minorCode = INTERNAL_SERVER_ERROR
DEBUG - [UI-########-####-####-####-############-######-##-###] [job-87] c.v.h.c.c.VcloudExceptionConverter       : vCD stacktrace = java.lang.IllegalArgumentException: Invalid UUID string:
 null
 at java.base/java.util.UUID.fromString(UUID.java:215)
 at com.vmware.vcloud.api.presentation.entity.common.UuidStringType.nullSafeSet(UuidStringType.java:159)

Environment

VMware Cloud Director Availability 4.7

Cause

When the replication is created the Datastore and RT VM are not assigned correctly during the configuration of the replication in the VCDA DB.

Resolution

To resolve this issue you will have to delete the replication and re-create it following this instructions:

1. Take a note of the RT vApp UUID from the UI for the affected replication.
2. Take a note of the C4-id for the replication
3. From the VCDA destination site manager login to the VCD API:
   
   vcd -k l <vcdcell IP> 'administrator@System' ‘password’
   
   
4. From the VCDA destination site manager list the destination site RT vApp:
   
   vcd -k g vApp/vapp-<UUID>
   
   
5. Delete the replication from the UI.
6. Check if the rtvapp is deleted:
   
   vcd -k g vApp/vapp-<UUID>


7. If the RT vApp is not deleted, delete it from the VCDA manager using the VCD API:
   
   vcd -k d vApp/vapp-<UUID> 
   
   
8. Using the C4-id check if there is any folder with that name in the destination datastore, if it does exist delete it manually.
9. Create the replication for the VM again.