Data Store Backup / Restore

Question

Hello all,

I am looking for some advice on the best way to refresh our test environment ArcGIS Portal Data Store.

We are using a cloud based solution and the test environment will be turned-off when not required and therefore will get out of sync with the production server.

I would like to be able to do a back-up of the production Data Store and restore it into the test environment when we have a significant production change that needs to be tested.

Are there any issues with this approach?

Is it possible?

Regards

Dean Howell

ChrisBeaudette · Answer

FWIW Esri Professional Services is maturing the environment-to-environment capability using the webgisdr tools and other custom code to resolve hostnames and URLs between portals to properly replicate content.  It's not productized (not sure if that's on the roadmap) but I believe they have a part-automated, part-manual process that reliably clones portal content across environments. Reach out to your Esri rep for details.

DeanHowell1 · Answer

Thanks Jonathan, very much appreciated. This will come in very handy for planning the way forward.

JonathanQuinn · Answer

You can use the exportmanageddb/importmanageddb tools within the install directory of Data Store.  The thing you'll need to consider, which may be tricky, is what about the services the data references in Server?  You'll need to back up the Server as well.  But then, you have items in the Portal that need to be created, and then other items, (such as webmaps), may have dependencies on the service items.  This gets complicated rather quickly.  Something you can look into is the WebGIS DR tool to take snapshots of your full deployment at a point in time, (data in the ArcGIS Data Store, any services published to any federated Server including the hosting Server, and any data, users, and content in the Portal).  You can then restore the snapshot to your test site.  The only caveat to this is that the tool was designed for disaster recovery, which means it's meant to either quickly repopulate a deployment with data after a disaster, or keep a standby deployment in sync with the primary.  In the event that you're just repopulating a deployment or switching to your standby, your users expect to reach their services, apps, and content using the same URLs.  That being said, where you take the backup has to have the same front-facing URLs as where you apply the backup.  The help on disaster recovery goes into a bit of detail on this.