• Back Up and Restore Deployments >
• Restore MongoDB Deployments >
• Restore Flows

Restore Flows

Introduction

When you request a restore, the Backup Daemon prepares the snapshot you receive. If you choose to restore manually, you either download the files from Ops Manager or the Backup Daemon securely copies the files to your destination server.

This page describes the flows for manual restores of scheduled snapshots and custom snapshots according to delivery option.

Restore a Backup using HTTP

Restoring a Completed Snapshot using HTTP

When you restore a snapshot using HTTP, the Backup Daemon creates and displays a download link in Ops Manager to the appropriate snapshot in snapshot storage.

When you click the download link, Ops Manager streams the snapshot archive to the destination system.

Note

This restore method requires no space on the server hosting the Backup Daemon: the file passes directly from the snapshot storage to the destination server.

Restoring a Specific Point in Time using HTTP

The HTTP point-in-time restore follows the same flow as the HTTP snapshot restore with added steps to apply the oplog .

When you restore to a point in time using HTTP, the Backup Daemon:

1. Retrieves the snapshot that immediately precedes the desired point in time.
2. Writes this snapshot to a temporary directory in the HEAD directory on server that hosts the Backup Daemon.
3. Retrieves oplog entries from the oplog store.
4. Applies the oplog entries which creates a custom snapshot from that point in time.
5. Writes the new custom snapshot back to the snapshot storage.
6. Creates and displays a download link in Ops Manager to the appropriate snapshot in snapshot storage.

When you click the download link, Ops Manager streams the snapshot archive to the destination system.

Note

Restoring to a point in time requires sufficient space in the HEAD directory on the server hosting the Backup Daemon for the snapshot files and oplog.

Restore an Archived Backup Using Secure Copy

Restoring an Archived Snapshot using Secure Copy

When you restore a snapshot using SCP, the Backup Daemon:

1. Retrieves the snapshot from the snapshot storage and writes it to its disk.
2. Copies the archive securely to the target directory on a destination server of your choosing. The format of the snapshot depends on where and how it was stored:
   • If the snapshot was stored on a blockstore, the Backup Daemon combines the database files into a snapshot archive then compresses that archive (tar.gz).
   • If the snapshot was stored on a file system store, the Backup Daemon returns a compressed snapshot archive (tar.gz) if the snapshot was compressed when the backup was taken or as an uncompressed snapshot archive (tar) if it was not.

Note

This restore method requires that you have sufficient space in the HEAD directory on the server hosting the Backup Daemon for the archive.

Restoring an Archived Snapshot from a Specific Point in Time using Secure Copy

The SCP point-in-time restore follows the same flow as the SCP snapshot restore with added steps to apply the oplog .

When you restore from a point in time using SCP, the Backup Daemon:

1. Retrieves the snapshot that immediately precedes the point in time.
2. Writes this snapshot to a temporary directory in the HEAD directory on server that hosts the Backup Daemon.
3. Retrieves all oplog entries since the last snapshot from the oplog store.
4. Applies the oplog entries which creates a custom snapshot from that point in time.
5. Combines the snapshot into an uncompressed archive (tar) or a compressed archive (tar.gz) if the snapshot was compressed when it was backed up.
6. Copies the archive securely to the target directory on a destination server of your choosing.

Note

Restoring to a point in time requires that you have sufficient space for the uncompressed snapshot files, oplog and snapshot archive in the HEAD directory on the server hosting the Backup Daemon.

Restore the Individual Database Files using Secure Copy

Restoring the Unarchived Database Files of a Snapshot using Secure Copy

When you restore the individual database files contained in a snapshot using SCP, the Backup Daemon:

1. Retrieves the snapshot files from the snapshot storage.
2. Copies the individual files securely to the target directory on a destination server of your choosing.

Note

This restore method requires no space on the server hosting the Backup Daemon: the file passes directly from the snapshot storage to the destination server.

The destination server requires sufficient space for the uncompressed files.

Restoring the Unarchived Database Files from a Specific Point in Time using Secure Copy

The SCP point-in-time restore follows the same flow as the SCP snapshot restore with added steps to apply the oplog .

When you restore individual files to a specific point in time using SCP, the Backup Daemon:

1. Retrieves the snapshot that immediately precedes the point in time.

2. Writes this snapshot to a temporary directory in the HEAD directory on server that hosts the Backup Daemon.

3. Retrieves all oplog entries since the last snapshot from the oplog store.

4. Applies the oplog entries to the temporary snapshot. This creates a custom snapshot up to requested point in time.

5. Copies the individual files securely to the target directory on a destination server of your choosing.

   The destination server requires sufficient space for the uncompressed files.

Note

Restoring to a point in time requires that you have sufficient space on the server hosting the Backup Daemon for the individual snapshot files and oplog.

The destination server requires sufficient space for the uncompressed data files.

←   Restore Overview Restore a Sharded Cluster from a Backup  →

© MongoDB, Inc 2008-present. MongoDB, Mongo, and the leaf logo are registered trademarks of MongoDB, Inc.