Take the 2-minute tour ×
Database Administrators Stack Exchange is a question and answer site for database professionals who wish to improve their database skills and learn from others in the community. It's 100% free, no registration required.

I have two purposes for this:

  1. To have an offsite backup in case of a region wide problem with Amazon Web Services.
  2. To copy production data from a production billing account to a beta billing account.

Currently it does not appear that Amazon supports either of these two use cases out of the box.

I have seen mysqldump and xtrabackup (see form post) mentioned.

I have also seen a more complicated process (documented here)

  1. A new RDS server in the source billing account (prod) is spun up off a recent backup.
  2. A new EC2 instance is spun up that has access to the RDS server in step 1.
  3. mysqldump is used to make a backup of this database.
  4. Backup is copied to an offsite location (S3?).
  5. In a separate account and/or region, a fresh RDS server is spun up.
  6. Database dump is imported.

Tips and suggestions are appreciated.

share|improve this question
add comment

2 Answers

up vote 7 down vote accepted

The recommended way to back up RDS is with automatic backups and DB snapshots. DB snapshots are basically the same as EBS snapshots, which are stored in S3 behind the scenes, but are only available within the same region.

If you need cross-region fault tolerance (good plan!), there is no way to restore your data in another region without doing it "the hard way" from a mysqldump. Your alternatives are to back up using mysqldump (slow and terrible for any reasonable sized dataset), or set up your own EC2-based slave in another region and back that up using any available method (xtrabackup, EBS snapshots, etc). However, then you are back to managing your own MySQL instances, so you might as well abandon RDS entirely.

For my money, RDS provides absolutely no benefits in really any way and a whole lot of disadvantages in performance, flexibility, and reliability. I would ask yourself what value RDS provides to you.

share|improve this answer
    
For my current needs mysqldump may work ok. I'm not expecting massive datasets and I don't generally need the data moved too quickly. I'm hoping that RDS will support a feature to spin up a new server off a snapshot in another account. Another useful feature would be to import/export the snapshots out of Amazon using either S3 or their Import/Export server. Anyway, thanks for the answer. –  Peter Stephens Feb 1 '12 at 22:47
add comment

I had the same issue. My solution was to write a simple bash script. It is limited to a single region however.

Here's the script in question:

#!/bin/bash
NOWDATE=`date +%Y-%m-%d`
BACKUPNAME="$NOWDATE.sql.gz"

echo "Creating backup of database finances to $BACKUPNAME"
mysqldump –user=user –password=password database_name | gzip -9 > $BACKUPNAME

echo "Succesfully created database backup"

echo "Uploading backup to Amazon S3 bucket…"
/usr/local/bin/s3cmd put $BACKUPNAME s3://path /to /file/$BACKUPNAME

echo "Successfully uploaded backup to S3"

echo "Deleting backup file…"
rm $BACKUPNAME

echo "Done"
share|improve this answer
1  
can you post the script here? link-only answers are discouraged on this site. –  Jack Douglas Oct 3 '12 at 21:37
    
I highly recommend --single-transaction when running mysqldump –  user1250 Jun 11 '13 at 15:04
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.