Backup and Replication creating definitive snapshots

[ivanildo.galvao]

Hello friends, it's okay with everyone?


I have a client that uses Veeam 6.5, this client has lately experienced some problems where the Veeam replicas VMS doubles every so often, besides accumulating snapshots that make him lose space on datastore, Veeam does not delete all snapshots or excludes and does not consolidate disks, eg a VM with a VMDK Thick 150GB comes sometimes in the provision 300GB Datastore, when I see, I realize that the VM has several other VMDK that were created by snapshots.

Well, this customer has a continuous replication of two critical VMS, but these also are backed up twice a day at 12:00 PM and 10:00 PM, I'm starting to think this is not cool, because as the VMS are replicated continuously, then suffer daily snapshots and even comes another snapshot because of backups, is this what is causing these anomalies?

What do you think? Suggest some other format such protection VMS?


Thank you!

[foggy]

I have a client that uses Veeam 6.5, this client has lately experienced some problems where the Veeam replicas VMS doubles every so often, besides accumulating snapshots that make him lose space on datastore, Veeam does not delete all snapshots or excludes and does not consolidate disks, eg a VM with a VMDK Thick 150GB comes sometimes in the provision 300GB Datastore, when I see, I realize that the VM has several other VMDK that were created by snapshots.

Hello, Ivanildo. Could you please clarify whether you're talking about original VMs or their replicas?

Well, this customer has a continuous replication of two critical VMS, but these also are backed up twice a day at 12:00 PM and 10:00 PM, I'm starting to think this is not cool, because as the VMS are replicated continuously, then suffer daily snapshots and even comes another snapshot because of backups, is this what is causing these anomalies?

Generally, this should not cause any issues as the job that comes at the moment when the VM is being backed up or replicated will wait for the VM to be released and only then proceed to snapshotting it.

[ivanildo.galvao]

Hello, I am referring to the original VMS. They were created with Thick 150GB disks not understand why the VMware ESXi provisioned almost 300GB, as VMS possem several VMDK in sequence, type Disk1.vmdk, Disk1-0001.vmdk, Disk1-0002.vmdk etc then understood that the snapshots were causing it.

Sometimes Veeam replicas doubles, when I enter the destination ESXi server, there should be a replica of each VM, I got already seeing four replicas of each VM on the same ESXi target, then I disable replication, I delete all replicas of ESXi destination, I enable replication again within 1 year, I had to do it around 4 or 5 times.

Although, even deleting snapshots created by Veeam and consolidating the discs still one of VMS continues with almost 300GB provisioned.

Is there any latest patch from Veeam 6.5, which may be helping to solve this problem.

Thanks !

[foggy]

Hello, I am referring to the original VMS. They were created with Thick 150GB disks not understand why the VMware ESXi provisioned almost 300GB, as VMS possem several VMDK in sequence, type Disk1.vmdk, Disk1-0001.vmdk, Disk1-0002.vmdk etc then understood that the snapshots were causing it.

How are these snapshots referred in the Snapshot Manager? Were they created by Veeam B&R job (VEEAM TEMPORARY SNAPSHOT)?

Sometimes Veeam replicas doubles, when I enter the destination ESXi server, there should be a replica of each VM, I got already seeing four replicas of each VM on the same ESXi target, then I disable replication, I delete all replicas of ESXi destination, I enable replication again within 1 year, I had to do it around 4 or 5 times.

There is a probability that something changes moref ID of the source VM or replica VM (e.g., the VM is migrated or re-registered in virtual infrastructure), making Veeam B&R either treating source VM as a new one or unable to find it's replica and forcing it to start replication from scratch. In any case, I recommend contacting support with this so they could take a closer look at the actual environment, that would be way more effective.