DB Backup/Restore Strategies

mike-k

I know that Xano only supports full Instance backups for three days. This is helpful, but the backups revert everything, including function stacks in every branch. Curious to hear what approaches anyone is using to backup just their data and have the ability to restore in case of an issue. Some ideas our team had:

1. Sign up for direct DB access and write a process to extract the databases periodically and store them in a GCP bucket.
2. Create a task that pulls the data via the function stack, converts to CSV, then ships them off to GCP for storage.

Curious if anyone is doing either of these and if if you'd had success with it? Any other approaches out there that people have had success with?

Chris Coleman

Hi, @mike-k! Both of these options are totally valid, and the approach you take is really more of a scalability concern.

1. Direct database access will skip all function stack processing of the data and can mean you not only get the data faster, but with less impact to other running processes of your instance. Of course, this does come with an additional cost if you don't currently have direct database access on your account.
2. Using a background task for this is absolutely a valid strategy. The consideration would then be on the performance side; you'll want to see how this potentially impacts other running processes and if the process completes in a sufficient amount of time. Let us know if you run into any trouble setting this up and we'll be happy to take a look.

Something else to throw out there; I do believe we can implement more frequent or custom backup strategies with our Enterprise plans. Not sure if that's reasonable for your project, but still worth mentioning just in case! Let us know if you have any questions.