Blob restore options

Question

Assuming 'point-in-time restore for containers', 'soft delete for blobs', 'versioning for blobs' and 'blob change feed' are all enabled, are my assumptions correct?

If a file is deleted, you restore it with versioning.
Using point-in-time restore only restores an entire container, not individual files. It overwrites the current contents of the container.
Restoring a deleted container is only possible if 'soft delete for containers' is enabled, or if garbage collection hasn't run yet.

If a storage account is geo-replicated, does the replica contain the snapshots, versioning and soft delete data of the primary?
Is there a scenario in which point-in-time restore would not work if a container / blog is corrupted?
Short of replicating one blob container to another, is point-in-time restore the best backup method for blob storage?

Trying to wrap my head around all these options, since there's no traditional backup for storage accounts like there is for VMs, etc.

Answer 1

@Joe H

If a file is deleted, you restore it with versioning.
You could restore it with versioning or soft delete. You will probably find the information here helpful.

Using point-in-time restore only restores an entire container, not individual files. It overwrites the current contents of the container.
This is incorrect. During the retention period, you can restore block blobs from the present state to a state at a previous point in time. See the documentation here for more info.

Restoring a deleted container is only possible if 'soft delete for containers' is enabled, or if garbage collection hasn't run yet.
Correct, though soft delete does not protect against the deletion of a storage account. See soft delete for containers for more info.

If a storage account is geo-replicated, does the replica contain the snapshots, versioning and soft delete data of the primary?
Yes, all data is replicated.

Is there a scenario in which point-in-time restore would not work if a container / blog is corrupted?
This should not be an issue. Azure Storage regularly verifies the integrity of data stored using cyclic redundancy checks (CRCs). If data corruption is detected, it is repaired using redundant data. Azure Storage also calculates checksums on all network traffic to detect corruption of data packets when storing or retrieving data.

Hope this helps! Let us know if you still have questions and we will be happy to help.

-------------------------------

Please don’t forget to "Accept the answer" and “up-vote” wherever the information provided helps you, this can be beneficial to other community members.