WAL-E is a program designed to perform continuous archiving of PostgreSQL WAL files and manage the use of pg_start_backup and pg_stop_backup.

To correspond on using WAL-E or to collaborate on its development, do not hesitate to send mail to the mailing list at wal-e@googlegroups.com. Github issues are also currently being used to track known problems, so please feel free to submit those.

WAL-E has four critical operators:

• backup-fetch
• backup-push
• wal-fetch
• wal-push

Of these, the "push" operators send things to S3, and "fetch" operators get things from S3. "wal" operators send/get write ahead log, and "backup" send/get a hot backup of the base database that WAL segments can be applied to.

All of these operators work in a context of three important environment-variable based settings:

• AWS_ACCESS_KEY_ID
• AWS_SECRET_ACCESS_KEY
• WALE_S3_PREFIX

With the exception of AWS_SECRET_ACCESS_KEY, all of these can be specified as arguments as well. The AWS* variables are the standard access-control keying system provided by Amazon.

The WALE_S3_PREFIX can be thought of as a context whereby this program operates on a single database cluster at a time. Generally, for any one database the WALE_S3_PREFIX will be the same between all four operators. This context-driven approach attempts to help users avoid errors such as one database overwriting the WAL segments of another, as long as the WALE_S3_PREFIX is set uniquely for each database.

• lzop
• psql
• mbuffer

This software is most frequently used with Python 2.6+. It will probably never support Python 2.5 or below because of the much more useful timeout semantics with sockets in Python 2.6+. Still, if you feel strongly about supporting Python 2.5, please send mail to the mailing list (see the Introduction).

This software also has Python dependencies; installing with setup.py will attempt to resolve them:

• gevent>=0.13
• boto>=2.0
• argparse, if not on Python 2.7

Pushing a base backup to S3:

$ AWS_SECRET_ACCESS_KEY=... wal-e         \
  -k AWS_ACCESS_KEY_ID                \
  --s3-prefix=s3://some-bucket/directory/or/whatever  \
  backup-push /var/lib/my/database

Sending a WAL segment to S3:

$ AWS_SECRET_ACCESS_KEY=... wal-e         \
  -k AWS_ACCESS_KEY_ID                \
  --s3-prefix=s3://some-bucket/directory/or/whatever  \
  wal-push /var/lib/my/database/pg_xlog/WAL_SEGMENT_LONG_HEX

It is generally recommended that one use some sort of environment variable management with WAL-E: working with it this way is less verbose, less prone to error, and less likely to expose secret information in logs.

At this time, AWS_SECRET_KEY is the only secret value, and recording it frequently in logs is not recommended. The tool has never and should never accept secret information in argv to avoid process table security problems. However, the user running PostgreSQL (typically 'postgres') must be able to run a program that can access this secret information, as part of its archive_command.

envdir, part of the daemontools package is one recommended approach to setting environment variables. One can prepare an envdir-compatible directory like so:

# Assumption: the group is trusted to read secret information
$ umask u=rwx,g=rx,o=
$ mkdir -p /etc/wal-e.d/env
$ echo "secret-key-content" > /etc/wal-e.d/env/AWS_SECRET_ACCESS_KEY
$ echo "access-key" > /etc/wal-e.d/env/AWS_ACCESS_KEY_ID
$ echo 's3://some-bucket/directory/or/whatever' > \
  /etc/wal-e.d/env/WALE_S3_PREFIX
$ chown -R root:postgres /etc/wal-e.d

After having done this preparation, it is possible to run WAL-E commands much more simply, with less risk of accidentally using incorrect values:

$ envdir /etc/wal-e.d/env wal-e backup-push ...
$ envdir /etc/wal-e.d/env wal-e wal-push ...

envdir is conveniently combined with the archive_command functionality used by PostgreSQL to enable continuous archiving. To enable continuous archiving, one needs to edit postgresql.conf and restart the server. The important settings to enable continuous archiving are related here:

wal_level = archive # hot_standby in 9.0 is also acceptable
archive_mode = on
archive_command = 'envdir /etc/wal-e.d/env wal-e wal-push %p'
archive_timeout = 60

Every segment archived will be noted in the PostgreSQL log.

A base backup (via backup-push) can be uploaded at any time, but this must be done at least once in order to perform a restoration. It must be done again if any WAL segment was not correctly uploaded: point in time recovery will not be able to continue if there are any gaps in the WAL segments.

Pulling a base backup from S3:

$ sudo -u postgres bash -c                          \
"envdir /etc/wal-e.d/pull-env wal-e         \
--s3-prefix=s3://some-bucket/directory/or/whatever  \
backup-fetch /var/lib/my/database LATEST"

This command makes use of the "LATEST" pseudo-name for a backup, which queries S3 to find the latest complete backup. Otherwise, a real name can be used:

$ sudo -u postgres bash -c                          \
"envdir /etc/wal-e.d/pull-env wal-e         \
--s3-prefix=s3://some-bucket/directory/or/whatever  \
backup-fetch                    \
/var/lib/my/database base_LONGWALNUMBER_POSITION_NUMBER"

One can find the name of available backups via the experimental backup-list operator, or using one's S3 browsing program of choice, by looking at the S3PREFIX/basebackups_NNN/... directory.

it is also likely one will need to provide a recovery.conf file, as documented in the PostgreSQL manual, to recover the base backup, as WAL files will need to be downloaded to make the hot-backup taken with backup-push. The WAL-E's wal-fetch subcommand is designed to be useful for this very purpose, as it may be used in a recovery.conf file like this:

restore_command = 'envdir /etc/wal-e.d/env wal-e wal-fetch "%f" "%p"'

These are commands that are not used expressly for backup or WAL pushing and fetching, but are important to the monitoring or maintenance of WAL-E archived databases. Unlike the critical four operators for taking and restoring backups (backup-push, backup-fetch, wal-push, wal-fetch) that must reside on the database machine, these commands can be productively run from any computer with WALE_S3_PREFIX and the necessary credentials to manipulate or read data there.

backup-list is useful for listing base backups that are complete for a given WAL-E context. Its output is subject to change, but currently it's a CSV with a one-line prepended header. Some fields are only filled in when the --detail option is passed to backup-list ¹.

Firstly, the fields that are filled in regardless of if --detail is passed or not:

Secondly, the fields that are filled in only when --detail is passed:

delete contains additional subcommands that are used for deleting data from S3 for various reasons. These commands are organized separately because the delete subcommand itself takes options that apply to any subcommand that does deletion, such as --confirm.

All deletions are designed to be reentrant and idempotent: there are no negative consequences if one runs several deletions at once or if one resubmits the same deletion command several times, with or without canceling other deletions that may be concurrent.

These commands have a dry-run mode that is the default. The command is basically optimize to not delete data except in a very specific circumstance to avoid operator error. Should a dry-run be performed, wal-e will instead simply report every key it would otherwise delete if it was not running in dry-run mode, along with prominent HINT-lines for every key noting that nothing was actually deleted from S3.

To actually delete any data, one must pass --confirm to wal-e delete. If one passes both --dry-run and --confirm, a dry run will be performed, regardless of the order of options passed.

Currently, these kinds of deletions are supported. Examples omit environment variable configuration for clarity:

• before: Delete all backups and wal segment files before the given base-backup name. This does not include the base backup passed: it will remain a viable backup.

  Example:

  $ wal-e delete [--confirm] before base_00000004000002DF000000A6_03626144

• old-versions: Delete all backups and wal file segments with an older format. This is only intended to be run after a major WAL-E version upgrade and the subsequent base-backup. If no base backup is successfully performed first, one is more exposed to data loss until one does perform a base backup.

  Example:

  $ wal-e delete [--confirm] old-versions

• everything: Delete all backups and wal file segments in the context. This is appropriate if one is decommissioning a database and has no need for its archives.

  Example:

  $ wal-e delete [--confirm] everything

All assets pushed to S3 are run through the program "lzop" which compresses the object using the very fast lzo compression algorithm. It takes roughly 2 CPU seconds to compress a gigabyte, which when sending things to S3 at about 25MB/s occupies about 5% CPU time. Compression ratios are expected to make file sizes 50% or less of the original file size in most cases, making backups and restorations considerably faster.

Because S3 requires the Content-Length header of a stored object to be set up-front, it is necessary to completely finish compressing an entire input file and storing the compressed output in a temporary file. Thus, the temporary file directory needs to be big enough and fast enough to support this, although this tool is designed to avoid calling fsync(), so some memory can be leveraged.

Base backups first have their files consolidated into disjoint tar files of limited length to avoid the relatively large per-file S3 overhead. This has the effect of making base backups and restores much faster when many small relations and ancillary files are involved.

To reduce the read load on base backups, they are sent through the tool mbuffer first. To use this rate-limited-read mode, use the option --cluster-read-rate-limit as seen in wal-e backup-push.