Backup of NAS, file shares, file servers and object storage.
Post Reply
vIdaho1
Influencer
Posts: 14
Liked: 3 times
Joined: May 10, 2021 5:17 pm
Full Name: Andrew Foster
Contact:

NetApp FlexGroup Support

Post by vIdaho1 »

Hello,

We have a number of large file shares with high inode counts that we use FlexGroups for. It is almost impossible to get a Veeam NAS backup on these shares when we have to just back them up as an SMB share because Veeam doesn't support backup from storage snapshot on FlexGroups. Are there any plans to add support for this? With the growth of data and long retentions we can't really get away from using FlexGroups so getting this support in to Veeam is very important to us.

We do currently have a case open to try and figure out why the performance is so terrible when trying to backup some of these shares as a standard SMB share, but no leads yet. Case number is 05422701.
Dima P.
Product Manager
Posts: 14415
Liked: 1576 times
Joined: Feb 04, 2013 2:07 pm
Full Name: Dmitry Popov
Location: Prague
Contact:

Re: NetApp FlexGroup Support

Post by Dima P. » 1 person likes this post

Hello Andrew,

Unfortunately FlexGroups support is not planned for v12 but we keep tracking this request very attentively for next versions. As for the case, I've asked QA team to review the details.
Dima P.
Product Manager
Posts: 14415
Liked: 1576 times
Joined: Feb 04, 2013 2:07 pm
Full Name: Dmitry Popov
Location: Prague
Contact:

Re: NetApp FlexGroup Support

Post by Dima P. » 2 people like this post

Andrew,

From log review we've noticed that you are using ExaGrid as a backup destination. While it's a good backup target it's deduplication slows down all operations due to metadata files being deduplicated. There is a workaround to configureScale-out backup repository with data and metadata extents being separated: metadata extent is used to store only the meta files away from the the dedup, data extent can be your deduplication storage: Scale-Out Repository with Extents in Metadata and Data Roles

Can you please try this solution any let us know how it goes? Thank you!
vIdaho1
Influencer
Posts: 14
Liked: 3 times
Joined: May 10, 2021 5:17 pm
Full Name: Andrew Foster
Contact:

Re: NetApp FlexGroup Support

Post by vIdaho1 » 1 person likes this post

Thank you, we will give this a shot. If we are keeping 30 restore points can we assume the metadata will be ~5% the size of the source data?
Dima P.
Product Manager
Posts: 14415
Liked: 1576 times
Joined: Feb 04, 2013 2:07 pm
Full Name: Dmitry Popov
Location: Prague
Contact:

Re: NetApp FlexGroup Support

Post by Dima P. »

Andrew,

Yup, 5% is the worst case scenario, but I'd size meta extent capacity based on these 5% just in case. Thanks!
vIdaho1
Influencer
Posts: 14
Liked: 3 times
Joined: May 10, 2021 5:17 pm
Full Name: Andrew Foster
Contact:

Re: NetApp FlexGroup Support

Post by vIdaho1 »

Sorry but I was just getting around to testing this and had a question. We have a SOBR created of ExaGrid extents and we currently have image based and NAS backups going to this. Do I just need to mark all the ExaGrid extents as "Data" extents and the SSD one I will add as "Metadata" or should I mark the ExaGrid extents as both data and metadata? I noticed in my testing I just started if I only mark ExaGrid ones as data and then I seal the metadata extent my NAS backup fails (Just trying to simulate losing the SSD extent). If I mark them as both it works again, but when I unsealed the SSD extent which is marked metadata only I get an error "7/7/2022 2:53:19 PM :: Error: Failed to distribute metadata replica: available extents do not meet required replication factor of 2"
mark49808
Enthusiast
Posts: 83
Liked: 13 times
Joined: Feb 02, 2017 6:31 pm
Contact:

Re: NetApp FlexGroup Support

Post by mark49808 »

+1 to NetApp Flexgroup support!
Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests