One of my VMs has recently developed issues when taking a checkpoint or backing up (via Veeam which uses checkpoints as part of the backup I believe).
If I manually run a checkpoint it gets to 9%, lingers, gets to 19% then fails with the error:
An error occurred while attempting to checkpoint the selected virtual machine(s)
Checkpoint operation failed
'VMname' could not initiate a checkpoint operation
Nothing untoward appears in Event Viewer / Hyper-V-VMMS other than an ID 18012 Error then a couple of informational alerts regarding the background merge of the failed checkpoint.
Weirdly, if I click on the VM's settings / Integration services and uncheck Backup (volume shadow copy), hit Apply then recheck it and hit OK, it'll take a snapshot quite happily. Something later seems to knock it out of whack again and checkpoints, Veeam replicas and backups fail.
I've rebooted the VM (backups are OK when it's off) but not the host yet. The only odd thing about this VM is it has a load of .vmgs files (4097kb each) in its Snapshots folder, each with a correspondingly named empty folder. e.g
DD4E1F41-B2E0-4007-8089-18C7A91967CB.vmgs
and Snaphots\DD4E1F41-B2E0-4007-8089-18C7A91967CB\
These seem to relate to times and dates of recent checkpoints/replication events.
I'm stumped. Any ideas?