VM bring dropped from 2016 RCT Hyper-V image backup - Error 43940

I am backing up a 2-node Server 2106 Hyper-V failover cluster, doing an incremental RCT backup nightly. One of the VMs is consistently being skipped with an error message of "Failed to get the RCT extents for Virtual machine 'server name' and will be skipped from backup. For details, please check 'winclustersvc.log'. (Log #2).

We are running Avamar version 18.2, which we upgraded to a few months ago from version 7.4. The Windows failover cluster was upgraded to Server 2016 about 3 weeks ago. The Hyper-V image backups, once we got them to work at all, had been completing successfully until 6/17, when they started dropping this VM.

An on-demand backup fails with the following messages:

2019-06-27 09:57:55 avhypervvss Error <43940>: Failed to get the RCT extents for Virtual machine 'FAMOFFSHARE ' and will be skipped from backup. For details, please check 'winclustersvc.log'. (Log #1)
2019-06-27 09:57:57 avhypervvss Error <43408>: Failed to create RCT checkpoint for the targets. (Log #1)
2019-06-27 09:57:57 avhypervvss Error <41674>: Failed to backup targets. Unable to snapshot targets. (Log #1)
2019-06-27 09:58:02 avhypervvss Error <41659>: Backup did not complete successfully (Log #1)

Disk space is not a problem on either the Hyper-V hosts or on the VM. I turned off automatic load balancing for the cluster to eliminate the possibility of the VM being migrated during the backup. There are no orphan checkpoints left behind from earlier backups. We are using the 5/16/2019 version of avhypervvss.exe that was provided by Support.

Does anyone know what is causing this or what I need to do to fix it?