VMware Cloud Community
Matt_B1
Enthusiast
Enthusiast

Site Recovery Manager (SRM) v6.0 - several recovery errors

I plan to open a support ticket but want to make these errors visible to all.  For the errors, I have replaced names but left the important content.  We had several issues with ESXi v6.0, vCenter 6.0, and SRM 6.0.  I recommend hold off on v6.0 products until VMware releases a U1 patch.

#1

Permission related blocked SRM views and certain functions not executing

When using SRM as any other user aside from administrator@vsphere.local, you run into permission related blocked views and certain functions not even initiating.  If you click around, you can sometimes get the view to populate.  In some situations, it is impossible to execute a recovery.  I have tried using AD accounts that are nested in groups with Administrator permission and even not nested.   If I log in as administrator@vsphere.local, there are no issues.

#2

"Cannot unmount volume 'Datastore Name: datastore1 VMFS uuid: <id>' because file system is busy. Correct the problem and retry the operation."

SRM fails to unmount the datastore from the protected site across all ESXi hosts.  The behavior is inconsistent in regards to which host it can't unmount from.  If I immediately rerun the recovery without making any changes, it will complete without issue.  I can also manually unmount the datastore without issue.

#3

Recovered datastore 'datastore1' is not accessible to host 'host1'.

Warns about the datastore not being acccessible on the recovery site during the recovery.   When I check the recovery cluster, the datastore is visible on all hosts.  I can simply rerun the recovery without any changes and it will work properly.

0 Kudos
1 Reply
basher
VMware Employee
VMware Employee

Hi

Opening a support ticket is always the best way to get consistent help with issues.

Let me try to ask some questions and offer some hints:

#1 Is the user granted Global Permission in SSO? You can check that by going to Administration -> Global Permissions. If not, could you please add the user there with any limited role and try again.

#2 Looks like a timeout/delay issue. GSS may be able to recommend changing some advanced settings so that SRM waits a bit longer after shutting down VMs and before attempting to unmount datastores

#3 This is the most intriguing one. Can you please look at all the tasks in the recovery site and check whether some mount datastore task has failed.

Best regards

Stefan Tsonev

Director - VMware Site Recovery Manager
0 Kudos