Comprehensive data protection for all workloads
Post Reply
averylarry
Veteran
Posts: 268
Liked: 30 times
Joined: Mar 22, 2011 7:43 pm
Full Name: Ted
Contact:

Replication issue -- possible V13 changes to or issues with calculating digests?

Post by averylarry »

Support case 08066190

(I've had Veeam for many years and have done dozens-to-hundreds of disk expansion activities.)

I recently upgraded V12 to V13 (Windows). 13.0.1.2067

I have a concern that something changed with V13 replication jobs calculating digests. I need to understand if this is going to happen every time I expand a disk? The first time I expand a disk after the V13 upgrade? Never again because this is just some strange aberration?

I expanded a 5.8Tb disk to 6.4Tb. Backup read the entire disk (as expected) and found 42.5Gb of changes.
Replication job uses the backup repository as source and was up-to-date before the disk expansion.

The replication job "found" that 5.8Tb of data had changed on the expanded disk. There is just no way that's possible. I would have been better off just deleting the disk so it wouldn't have to calculate digests. (Note that the replication job still hasn't completed and seems to be incorrectly finding changed data on the non-expanded disks where no data was changed.)

I think this is correct:
1) Expand production disk
2) CBT reset by backup job
3) Backup job reads entire disk (calculating digests)
4) Backup compares repository data to production data (digest comparison)
5) Changed data is backed up
6) Replication job deletes snapshots
7) Replication disk expanded
8) CBT reset (I think, or maybe this doesn’t matter with replication jobs)
9) Digests calculated
10) Replica digests compared with backup repository digests
11) Matching data skipped
12) Changed data transferred

The backup correctly found 42.5Gb of changed data. But on the replication job (steps 10-12 above), it appears that none of the data matched on the resized disk (except for or excluding the 600Gb of expanded emptiness), so all data is being incorrectly seen as changed and thus being transferred.
Mildur
Product Manager
Posts: 11697
Liked: 3292 times
Joined: May 13, 2017 4:51 pm
Full Name: Fabian K.
Location: Switzerland
Contact:

Re: Replication issue -- possible V13 changes to or issues with calculating digests?

Post by Mildur »

Hello Ted,

Off the top of my head, there was no change that would lead to a full replication of the entire disk after disk expansion. I also haven’t heard about this from other customers yet.

I ran a short test in my lab for both scenarios, replica from production and backup. In both scenarios, the transferred data was well below the initial transfer size.
Initial run:
Image

After disk expansion:
Image

Maybe your production disk has too many changed data blocks compared to the base disk of the replica VM. Since all snapshots are deleted on the replica, the first replica session has to replicate all changed and new blocks from the production VM disk compared to the base disk on the replica. A Multi TB server must have many changed data blocks between base disk of a replica and now.

Please continue working with our support team on this behavior. Use the Talk to a Manager option if you believe it takes too long to get an answer.

Best,
Fabian
Product Management Analyst @ Veeam Software
Post Reply

Who is online

Users browsing this forum: Google [Bot], joecruz and 595 guests