Where is the source bucket specification in RDS > Restore from S3?
After selecting the Restore from S3 button option in RDS > Databases, the first section of the UI is contradictory to me.
It says "S3 destination" and talks about **writing **audit logs. But clicking on the "info" link on the "prefix" field opens something that talks about creating the DB **from **the files stored in the bucket.
It strikes me that this section is really where one should enter the bucket and path for the snapshot in S3 that is being restored from?
I can't otherwise see where the S3 source location is specified.
However, if I try entering the bucket and patch of a valid snapshot it is also not working for me - even when I choose "Create a New Role" for the IAM Role, it fails with error:
**IAM role ARN value is invalid or does not include the required permissions for: S3_SNAPSHOT_INGESTION **
Which is odd, as the role should have been created specifically to perform this operation. And it looks like it has permission to read from the bucket.
Am I doing something wrong, or it this a bit broken?
It does seem a bit odd I can't choose a stored snapshot in the Exports in Amazon S3 list and then have the option to "restore from this snapshot".
PS: Sorry about the incorrect tag. It was not possible to choose a simple "RDS" tag, there were only 5 non-relevant options to choose from.
- Newest
- Most votes
- Most comments
You cant restore an RDS instance from an exported Snapshot to S3 as the data is stored in an Apache Parquet format that is compressed.
Here is a link to the documentation for restoring an RDS DB from a snapshot:
https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/USER_RestoreFromSnapshot.html
You can also export a DB snapshot to S3 but cannot restore to RDS from the S3 export. (See Exporting DB snapshot data to Amazon S3)
You can export DB snapshot data to an Amazon S3 bucket. The export process runs in the background and doesn't affect the performance of your active DB instance.
When you export a DB snapshot, Amazon RDS extracts data from the snapshot and stores it in an Amazon S3 bucket. The data is stored in an Apache Parquet format that is compressed and consistent.
You can export all types of DB snapshots—including manual snapshots, automated system snapshots, and snapshots created by the AWS Backup service. By default, all data in the snapshot is exported. However, you can choose to export specific sets of databases, schemas, or tables.
After the data is exported, you can analyze the exported data directly through tools like Amazon Athena or Amazon Redshift Spectrum. For more information on using Athena to read Parquet data, see Parquet SerDe in the Amazon Athena User Guide. For more information on using Redshift Spectrum to read Parquet data, see COPY from columnar data formats in the Amazon Redshift Database Developer Guide.
Hope this helps!
Thanks iBehr. I've checked and it looks to be the case! It seems odd to me, as an important part of any backup strategy is to be able to test that you actually can restore from the backup, and the "restore from snapshot" works completely as expected. What is actually the best practice then to be able restore from >35 days old? Back to "mysqldump"?
You’ll need to use AWS backup for backups needing to be kept for anything older than 35 days
Thanks Gary. This was totally unclear to me. Now I am pointing in the right direction with AWS Backup :-)
Relevant content
- asked a year ago
- asked a year ago
AWS OFFICIALUpdated 2 years ago- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated 2 years ago
Thanks Gary for the reply. The "Restore from S3" really does look like a function to do exactly that, although as I noticed the UI does look at bit inconsistent. Do you know what it does?