HV Agentless: Change of protection schedule triggers base image (4039985)

Changing the Protection Schedule for any agentlessly protected machine on a Hyper-V host will result in a failed base image followed by a "bad record" error.

After changing the protection schedule for any machine using agentless protection on a Hyper-V host, a base image is forced instead of an incremental, and the base image will fail with:

Replay.Core.Contracts.Transfer.CoreTransferFailedException: The transfer of one or more volume images from a protected machine to the Core has failed ---> System.AggregateException: One or more errors occurred. ---> Replay.Core.Contracts.Transfer.CoreVolumeTransferFailedException: The transfer failed for volume 'Hard disk 0\Volume 0 (System Reserved)' ---> System.AggregateException: One or more errors occurred. ---> System.InvalidOperationException: Sequence contains more than one element at System.Linq.Enumerable.SingleOrDefault[TSource](IEnumerable{{1 source) at

This base image will create duplicate records in the repository, which result in:

Replay.Core.Contracts.DedupeVolumeManager.DedupeVolumeManagerException: A bad record has been read. There may be a problem with the repository or the underlying storage device

To resolve this issue, please contact Support and open a service request. There are two solutions:

Solution 1

If you have changed the protection schedule and have the base image and bad record errors:

1. Contact support. The solution requires a pair of patches to be applied by a Support Engineer.
2. First patch: Applied to the core only, removes the base image with bad record chain.
3. Second patch: Applied to the core and all Hyper-V hosts, prevents the issue from happening again.
4. Patches do not require a reboot.

NOTE: After this solution is applied, the next backup on the affected machine will be a base image.

Solution 2

If you have not experienced the error and would like the patches to prevent the issue from happening:

1. Contact support. Patches will be provided and can be self applied. They do not require a Support Engineer.
2. One patch is for the core only.
3. One patch is for the Hyper-V hosts only.
4. Patches do not require a reboot and will not trigger a base image.