• EN
Question

Veeam with Nutanix that has two nodes issue

J
Userlevel 4
Badge

We have Veeam pushing a couple of Nutanix clusters.  Jobs are split between the clusters and Nutanix is replicating between the two clusters so if one goes down…

Issue that we’re seeing is when Veeam is backing up a server, snaps are being created to do the backup. Replication between the two Nutanix clusters is seeing the snap and replicating it to the second node.  Since the snap doesn’t have a retention set, Nutanix is setting the snaps for 60 day retention.  When we’re starting to deal with file servers north of 20tb, this becomes and issue.  Is there anyway to force Veeam to set a short retention on snaps or is this something Nutanix should address?

Thanks in advance.

3 comments

Pybarra
Userlevel 6
Badge +3

Since you are doing Nutanix replication to another host you should contact Nutanix support😀.

Moustafa_Hindawi
Userlevel 7
Badge +6

Hello @jdower98 

Veeam Backup for Nutanix AHV applies the configured retention policy settings to automatically remove snapshots created by jobs. If necessary, you can also remove snapshots manually.

To configure retention policy: https://helpcenter.veeam.com/docs/vbahv/userguide/snap_job_retention.html?ver=40

To remove snapshots manually: https://helpcenter.veeam.com/docs/vbahv/userguide/removing_snapshots.html?ver=40 

Moustafa Hindawi | Blogger | Nutanix Technology Champion | NCSE | NCM-MCI | VMCE | 2xVCAP-DTM & DCV | 2xVCP-NV & CMA | LinkedIn: https://www.linkedin.com/in/moustafaadelmoustafa/ | My Blog: https://cloudwhales.co.uk/author/mhindawi/
J
Userlevel 4
Badge

Locally, the snaps are removed off the storage on the node Nutanix created.  The issue lies with synchronous replication and the fact that these snaps are being replicated to a second node.  Nutanix creates the snap, replication copies the snap to the second node.  Nutanix deletes the snap off the first but not the second and there’s no expiration date so they have to be removed manually.  With hundreds of servers bouncing due to replication, its too much to keep on top of it.

Possible options - If possible, move the snaps to someplace that isn’t replicated (per Nutanix, impossible to more the default location).  Have the replicated snaps set with a expiration data - again Nutanix said can’t be done.

I’m scratching my head that we can’t be the only ones running in such a setup.  We’ve found that Hycu does the same thing in testing.  

jd

Comment


Terms & Conditions