Skip to main content
  • EN

Beeing a Veeam Architect as well as DataCore DCIE at the same time I was always pushing for this plugin.

After it finally happened, I took the time to write a deep dive blogpost covering all aspects:

  • Technical background, prerequisites
  • BfSS jobs, howto and caveats
  • Snapshot only jobs
  • “Instant-Sandboxing” from storage snapshots
  • Off-label-usage with unique functionality of DataCore: Using DataCore CDP rollbacks together with the plugin to recover without having had a snapshot in the first place… (cool stuff!)

Find the article in English here:
https://www.elasticsky.de/en/2020/06/veeam-storage-plugin-for-datacore-deepdive/

Might come in handy if you want to combine those great products.

 

Comments or complements welcome! :slight_smile:

 

Regards,

Michael

Thanks for sharing. Will go through it very soon.

Thanks Michael for sharing! I did quite a lot of DataCore implementations some time ago. A great product! Is this configuration fully supported by Veeam? 

Hi @vNote42. The config is fully supported as is the official plugin. Even the DataCore CDP integration I discussed in the blog is supported as it ends up to be a regular snapshot.
I used to do a DeepDive video together with DataCore in German (shouldn’t be a problem for you:wink: ).

https://www.datacore.com/de/event/deep-dive-das-optimale-backup-mit-sansymphony-und-veeam-br/

Cheers,

Mike

Thank you @Michael Melter 
great article and a very informative deep dive.

I don't have hands-on experience with DataCore up to now, but I will have a closer look at it...

BTW: what also works fine with DataCore is to present volumes as read-only to backup proxy - works for SANdirect-mode.

… I ask myself why read-only export is no common option for so many storage systems?

Hi @vNote42 and @Michael Melter , thx for sharing. With my company we implement also often Datacore, but almost exclusively in hyperconverged setup, so Datacore is implemented on a VM running on vSphere or Hyper-V and presenting the local storage as shared storage.

I’m also Veeam Certified Architect and DCIE.

In my opinion, in a hyperconverged setup the best way as a proxy is using hot-add method using a seperate VM as a proxy-server and not direct SAN connection. In my opinion there is not much difference between Datacore and VMWare vSAN or HPE StoreVirtual. By Veeam it is recommended to use hot-add method  and not SAN connection.

In a setup where Datacore is implemented on a physical server, of course SAN Direct Connection is recommended.

 

What do you guys think?

Agree, in such a HCI environment, direct-mode is not the big difference. SANdirect mode should always be discussed from a security perspective.

Even in HCI you might want to leverage snapshot only jobs or even CDP by hooking it onto the integration from your HCI VMs. Also the mode is IMHO smoother than hot-add. Many of the legends here will be aware that hot-add used to be called the “least robust” even by Veeam itself. ;)
It’s more robust today, but I always prefer direct storage access.

vSAN is different in that you don’t have the option to access the blocks directly. So hot-add is your only way to go so far.

Personally I don’t prefer to limit the proxy to read-only. We often use the mode for restoring as well as we let thin provisioning be up to the storage and not the VM. Some customers though want to have read-only for security reasons as @vNote42 mentioned already.

 

Thanks for sharing 

Comment


Terms & Conditions