1 Answer
- Newest
- Most votes
- Most comments
0
Hi Scott - The DbSnapshotIdentifier property is a "Replacement" property so whenever it is updated a new instance is created and previous instance destroyed. I'll walk through a brief scenario to hopefully explain.
- Create a cluster in a new stack without DbSnapshotIdentifier property - this will be a blank DB
- Update the stack with DbSnapshotIdentifier to restore from a snapshot - this will build a new db from the snapshot and destroy the previous instance
- Update the stack again with some other property keeping the previous DbSnapshotIdentifier value the same - no restore will be attempted as the DbSnapshotIdentifier property has not changed. It will not revert to the snapshot specified previously
- Update the stack again with a new DbSnapshotIdentifier property - this will build a new db from the new snapshot and destroy the previous instance
If you remove the DbSnapshotIdentifier property from the stack it will see that as a change and therefore build a new blank instance and destroy the previous. Therefore once you use the property once you need to keep using with either the old value to do nothing or a new value for a new restore
Relevant content
- Accepted Answerasked 5 years ago
- Accepted Answerasked 4 years ago
- AWS OFFICIALUpdated 3 months ago
- AWS OFFICIALUpdated 6 months ago
- Why is my Amazon Aurora DB cluster clone, snapshot restore, or point in time restore taking so long?AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated a year ago