Upgrade, redeploy or deleting an NSX-V Edge fails

Products

• VMware NSX for vSphere

Issue/Introduction

• Attempting to upgrade, delete, or redeploy a NSX Edge fails with this error:
  
  The operation failed for the entity with the following error message. Failed to deploy edge appliance vse-xxxxxxx-edge01 (dddddddd-ffff-eeee-cccc-55051132bb4b)-0.
   
• In the NSX manager vsm.log file, you see entries similar to:
   

  YYYY-MM-DD HH:mm:ss.335 GMT+00:00 INFO http-nio-127.0.0.1-7441-exec-87 AuditingServiceImpl:147 - [AuditLog] UserName:'vsphere.local\administrator', ModuleName:'NSX Edge Gateway', Operation:'UPGRADE', Resource:'vse-########-edge01 (#########-####-####-####-55051132bb4b)', Time:'Tue Mar 14 HH:mm:ss.330 GMT+00:00 2017', Status:'SUCCESS', Universal Object:'false'
YYYY-MM-DD HH:mm:ss.346 GMT+00:00 INFO http-nio-127.0.0.1-7441-exec-87 TaskServiceImpl:399 - TF:Scheduling Job jobdata-124401
YYYY-MM-DD HH:mm:ss.353 GMT+00:00 INFO http-nio-127.0.0.1-7441-exec-87 SchedulerService:198 - Job jobdata-124401 is scheduled and put data in to scheduleInfoTable
YYYY-MM-DD HH:mm:ss.353 GMT+00:00 INFO http-nio-127.0.0.1-7441-exec-87 Publisher:594 - Scheduled job Id jobdata-124401
YYYY-MM-DD HH:mm:ss.346 GMT+00:00 INFO pool-11-thread-1 ImmediateScheduler:35 - TF:Schedule Now Job ID jobdata-124401
YYYY-MM-DD HH:mm:ss.352 GMT+00:00 INFO taskScheduler-26 JobWorker:243 - Updating the status for jobinstance-2292841 to EXECUTING
YYYY-MM-DD HH:mm:ss.363 GMT+00:00 INFO taskScheduler-26 SchedulerQueueServiceImpl:64 - [TF] Created a new bucket for module default_module and total number of buckets 1
YYYY-MM-DD HH:mm:ss.363 GMT+00:00 INFO taskScheduler-26 SchedulerQueueServiceImpl:80 - The task edge.publish.task.name [id:task-241313] is added to the SchedulerQueue
YYYY-MM-DD HH:mm:ss.364 GMT+00:00 INFO pool-10-thread-1 ScheduleSynchronizer:48 - Start executing task: task-241313 and running executor threads 1
YYYY-MM-DD HH:mm:ss.388 GMT+00:00 INFO TaskFrameworkExecutor-19 PublishUtils:125 - Phase=Publishing edgeId=edge-### jobId=jobdata-124401 timeSinceStart=1084 ms message= progress='edge.message.publish;edge-###;78'
YYYY-MM-DD HH:mm:ss.428 GMT+00:00 INFO TaskFrameworkExecutor-19 PublishUtils:125 - Phase=Publishing edgeId=edge-### jobId=jobdata-124401 timeSinceStart=1124 ms message= progress='edge.appliance.preparevcconfig;edge-122;vse-########-edge01 (#########-####-####-####-55051132bb4b)'
YYYY-MM-DD HH:mm:ss.430 GMT+00:00 INFO TaskFrameworkExecutor-19 InventoryUtils:408 - Failed to find vm with ID = 'vm-###' in the inventory.
YYYY-MM-DD HH:mm:ss.431 GMT+00:00 INFO TaskFrameworkExecutor-19 InventoryUtils:415 - Discovering Vm with vmId : 'vm-###' having VC uuid '#########-####-####-####-1234567890ee'
YYYY-MM-DD HH:mm:ss.440 GMT+00:00 ERROR TaskFrameworkExecutor-19 InventoryUtils:418 - Failed to discover the vm : 'null' using vcUuId = '#########-####-####-####-1234567890ee'
YYYY-MM-DD HH:mm:ss.441 GMT+00:00 INFO TaskFrameworkExecutor-19 EdgeUtils:316 - populateSystemEvent parameters : sourceName vm-123456, morefIdOfObjectOnVc null, moduleName NSX Edge Gateway, eventCode EDGE_VM_NOT_FOUND_IN_VC_INVENTORY, severity Critical, messageParams [vm-###] eventMetaData {missing vmId=vm-123456}

Environment

VMware NSX Data Center for vSphere 6.3.x
VMware NSX Data Center for vSphere 6.2.x

Cause

This issue occurs when the vCenter server database has a different Edge VM ID than the NSX Manager database

Resolution

This is a known issue affecting NSX for vSphere 6.2.x and 6.3.x.

 

Currently, there is no resolution.

 

Contact Broadcom Support if you encounter this issue.