Starting with Zerto Virtual Replication (ZVR) 3.1, journal sizing is flexible, both to enable dynamically growing the journal and also to shrink the journal size. This article will outline the changes in journal provisioning behavior.
Prior to ZVR 3.1, each VM journal was thin-provisioned, and could grow to the VM Max Journal Size value specified for the VM in the VPG configuration. In ZVR 3.1, while this is still true, if the journal grows to 80% of the current total journal volumes? sizes, a new volume is added to increase the journal size up until the hard limit specified. Where the VRA is not on a host ESX/ESXi that is version 5.0 or less, when the journal size shrinks to 50% of the previous size, the biggest unused journal volume is removed. With VMFS disks the journal can also shrink by reclaiming unused space. With NFS disks the journal can shrink only by removing unused volumes.
If a journal is located on a VMFS datastore, the flexible journaling mechanism behaves as follows:
Grow the used journal volume (increase space used on a VMDK)
Grow the provisioned journal volume (add additional VMDKs to the journal)
Shrink the used journal volume (reclaim space used on a VMDK)
Shrink the provisioned journal volume (remove VMDKs from the journal)
If a journal is located on an NFS datastore, the flexible journaling mechanism behaves as follows: