Managed database backups

Every managed database on webgrow24 is backed up automatically. This article explains how often, how long we keep the backups, how to take a manual snapshot, and how to restore from one.

Backup schedule

Backups run once per day at a random time between 02:00 and 04:00 IST. The window is randomised per database so that no single time becomes a hot spot on the cluster. The first backup fires 24 hours after the database is created.

| Plan | Retention | |---|---| | Shared Starter | 3 days | | Shared Business | 14 days | | WordPress Pro | 14 days | | VPS 1 / VPS 2 | 30 days | | Cloud 1 | 30 days | | Reseller | 30 days |

Backups older than the retention window are deleted automatically. You cannot recover a backup that has been deleted. If you need longer retention, talk to support.

What is in a backup

A full logical backup (pg_dump, mysqldump, or mongodump depending on the engine). It includes:

• All schemas and tables
• All data
• All indexes
• All users and grants

It does NOT include:

• Connection logs or query logs
• Performance schema or slow-query tables
• Temporary tables that were not committed at backup time

How to take a manual snapshot

Open the database in the dashboard. Click Backups in the side nav. Click Take snapshot. The snapshot is added to the list with a timestamp and the label "Manual". Manual snapshots are NOT subject to the retention window — they are kept until you delete them. You can have up to 10 manual snapshots per database.

How to restore from a backup

Open the database. Click Backups. Find the backup you want. Click Restore. Confirm the warning. The current database is replaced with the backup contents. The whole database is locked for the duration of the restore — for a 1 GB Postgres database this is around 90 seconds.

Restore caveats

• The restore overwrites the existing database. We do not merge.
• The restore locks writes for the duration. Read traffic continues

to serve stale data from replicas.

• Apps connected to the database will lose their connection pool

briefly. Most pools recover automatically.

• We do not currently support point-in-time recovery. The restore is

always to a full backup, not to a specific timestamp within one.

How to download a backup

Click Download on any backup row. The file is a gzipped dump in the native format for the engine. You can restore it yourself with:

``bash # Postgres gunzip -c backup.sql.gz | psql "postgres://..." # MySQL / MariaDB gunzip -c backup.sql.gz | mysql "mysql://..." # MongoDB mongorestore --gzip --archive=backup.archive.gz ``

How to delete a manual snapshot

Click Delete on the row. Confirm. The snapshot is removed immediately. We do not retain deleted snapshots.

What happens if a backup fails

You get an email the same morning with the failure reason. The most common cause is a long-running transaction that holds a lock for the duration of the backup window. The fix is usually to break the transaction into smaller pieces. If you cannot, raise a ticket and we will work with you on a backup window that avoids the conflict.

What happens during a regional outage

Our backups are stored in the same Palitana data center as the primary databases. A regional outage will affect both. Cross-region backup replication is on the v2 roadmap. For now, download your backups manually if you want an off-site copy.

If you have a backup question that this guide does not answer, open a support ticket. We respond in under an hour during business hours IST.