Server 2016 file server dedup

[johnhk]

Hi,
we have enable dedup for server 2016 (1.6TB), but Veeam still backup the full size (3.6TB).
Do it normal?

Thanks.

[veremin]

Have you enabled it inside backed up Virtual Machine?

[johnhk]

Yes , Enable inline data deduplication

[veremin]

Inline data deduplication is a feature of backup server, not Windows Server 2016. So, where exactly did you enable the said feature?

If you enabled in on a backup server, what is the resulting size of backup file?

Otherwise, if data deduplication was enabled inside backed up Virtual Machine, then, that should not reduce the size of backup file, since Veeam Backup & Replication works on image level and doesn't care about data reduction technologies present inside the guest OS.

Thanks.

[johnhk]

I've enable it on file server only , not the backup server.
So Veeam backup image level, VHD or vmdk present size instead on inside guest OS size right?
Thanks.

[veremin]

Correct, more details can be found here. Actually, usage of deduplication inside Guest will result in increased size of incremental backups, as deduplication affects dramatically number of blocks changed. Thanks

[nmdange]

One thing to keep in mind, if you have dynamically expanding vhd/vmdk, the file may not shrink when you remove data within the VM. You might have to do a compact operation on the host to shrink the physical vhd/vmdk file.

[ITP-Stan]

I would not recommend deduplication on the fileserver.

[mcclans]

Why on earth would you not recommend dedup on the file server?

[mkaec]

Deduplication has been awesome on our file servers. The average space savings is 33%. But, it is not good on the servers backed up by Veeam because garbage collection dramatically increases the changed blocks, and thus the size of the incremental backups. We have dedup active on some servers being backed up by Windows Server backup and it has been working great there.

[ITP-Stan]

Because of corruption of files, but maybe this was a one-off and for normal filesizes there never was a problem.

I’ve already received the official confirmation from Microsoft that this is the know issue (ID 10165851) which is scheduled to be addressed in the next Windows Server 2016 servicing update. There are actually two separate issues, both leading to file corruption when using deduplication on very large files. One issue occurs when files grow to 2.2TB or larger, and another one causes loss of checksums for files with “smaller sizes” – this is the actual wording of the official note, so I have no idea how small

[kurt]

I believe Microsoft already released patches to fix those issues.