Backup and recovery

This chapter describes the commands and processes you use to back up and recover your Helix Core Server. For information about multi-server backup and recovery, see the Backing up and upgrading services section under Deployment architecture.

If you are using Helix Core Extensions, see Backup of server extensions in Helix Core Extensions Developer Guide.

Warning

To reduce the risk of data loss,

  • Validate your backup procedures and follow them.

  • Take checkpoints and rotate journals on a regular basis. See Checkpoint files and Journal files.

Tip

If your site is very large (gigabytes of db.* files), creating a checkpoint might take a substantial amount of time. Therefore, consider:

  • only creating a checkpoint at the end of each work week, while rotating the journal file during your nightly backup

  • performing checkpoints on a separate instance of your Helix Core p4d Server database. See the Support Knowledgebase article, Offline checkpoints.

Version files versus database files (metadata)

Helix Core Server stores two kinds of data: versioned filesClosed and database files, which are metadataClosed.

  • Versioned files are files submitted by Helix Server users. Versioned files are stored in directory trees called depots.

    The P4ROOT environment variable represents the root directory of Helix Core Server installation. Each depot is a subdirectory under the P4ROOT directory.

    By default, the versioned files for a given depot are located in a tree of directories beneath this subdirectory.

    For an alternative location, see The server.depot.root configurable section of Set a depot location.

  • Database files contain metadata, such as the history and present state of the versioned files. Database files track changelists, opened files, client workspace specifications, branch and stream mappings, labels, and more.

    Database files appear as db.* files in the top level of the server root directory. Each db.* file contains a single, binary-encoded database table.