Postgres Backup with Wal-e

The following are the steps I took to setup Wal-e 0.6.2 on Ubuntu 12.04.2 LTS and Postgres 9.1.9. After following the installation instructions, every minute Wal-e will make incremental backups to Amazon S3.


$ sudo apt-get install libevent-dev python-all-dev daemontools lzop pv postgresql-client
$ sudo pip install wal-e
$ 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
$ sudo chown -R root:postgres /etc/wal-e.d

Added the following to the end of the file, /etc/postgresql/9.1/main/postgresql.conf:

wal_level = archive
archive_mode = on
archive_command = 'envdir /etc/wal-e.d/env /usr/local/bin/wal-e wal-push %p'
archive_timeout = 60

Restart postgres:

$ sudo service postgresql restart

Make a base backup:

$ sudo -u postgres bash -c "envdir /etc/wal-e.d/env /usr/local/bin/wal-e backup-push \

Other Notes

You can use the following query (as superuser) to monitor how many segments are ‘ready’ for archiving but have not yet been sent (those are ‘done’). [1]

SELECT suffix, count(*)
    FROM (
        SELECT (regexp_matches(name, E'\[0-9A-F]+\.([^\.]*)$'))[1] AS suffix
        FROM pg_ls_dir('pg_xlog/archive_status') name
    ) AS matches
    GROUP BY suffix;


[1]Thanks to Daniel Farina for the tip.