When a client application connects to the database server, it specifies which PostgreSQL user name it wants to connect as, much the same way one logs into Linux as a particular user. Within the SQL environment, the active database user name determines access privileges to database objects. It is therefore essential to restrict which database user names a given client can connect as.
Authentication is the process by which the database server establishes the identity of the client, and by extension determines whether the client application (or the user who runs the client application) is permitted to connect with the user name that was requested.
PostgreSQL offers client authentication by (client) host and by database, with a number of different authentication methods available. Some authentication methods allow you to restrict by user name as well.
PostgreSQL database user names are logically separate from user names of the operating system in which the server runs. If all the users of a particular server also have accounts on the server's machine, it makes sense to assign database user names that match their Linux user names. However, a server that accepts remote connections may have many users who have no local account, and in such cases there need be no connection between database usernames and Linux usernames.
Client authentication is controlled by the file pg_hba.conf in the PGDATA directory, for example, /var/lib/pgsql/data/pg_hba.conf. (HBA stands for host-based authentication.) A default pg_hba.conf file is installed when the data area is initialized by initdb.
The general format of the pg_hba.conf file is of a set of records, one per line. Blank lines and lines beginning with a hash character ("#") are ignored. A record is made up of a number of fields which are separated by spaces and/or tabs. Records cannot be continued across lines.
Each record specifies a connection type, a client IP address range (if relevant for the connection type), a database name or names, and the authentication method to be used for connections matching these parameters. The first record that matches the type, client address, and requested database name of a connection attempt is used to do the authentication step. There is no "fall-through" or "backup": if one record is chosen and the authentication fails, the following records are not considered. If no record matches, the access will be denied.
A record can have one of the three formats:
local database authentication-method [ authentication-option ]
host database IP-address IP-mask authentication-method [ authentication-option ]
hostssl database IP-address IP-mask authentication-method [ authentication-option ]
|
This record pertains to connection attempts over UNIX-domain sockets.
This record pertains to connection attempts over TCP/IP networks. Note that TCP/IP connections are completely disabled unless the server is started with the -i switch or the equivalent configuration parameter is set.
This record pertains to connection attempts with SSL over TCP/IP. To make use of this option the server must be built with SSL support enabled. Furthermore, SSL must be enabled with the -l option or equivalent configuration setting when the server is started.
Specifies the database that this record applies to. The value all specifies that it applies to all databases, while the value sameuser identifies the database with the same name as the connecting user. Otherwise, this is the name of a specific PostgreSQL database.
These two fields control to which hosts a
host record applies, based on their IP
address.
The precise logic is that
must be zero for the record to match.
Specifies the method that users must use to authenticate themselves when connecting to that database. The possible choices follow, details are in the Section called Authentication Methods.
The connection is allowed unconditionally. This method allows any user that has login access to the client host to connect as any PostgreSQL user whatsoever.
The connection is rejected unconditionally. This is mostly useful to "filter out" certain hosts from a group.
The client is required to supply a password with the connection attempt which is required to match the password that was set up for the user.
An optional file name may be specified after the password keyword. This file is expected to contain a list of users that this record pertains to, and optionally alternative passwords.
The password is sent over the wire in clear text. For better protection, use the crypt method.
Like the password method, but the password is sent over the wire encrypted using a simple challenge-response protocol. This protects against incidental wire-sniffing. This is now the recommended choice for password-based authentication.
The name of a file may follow the md5 keyword. It contains a list of users who may connect using this record.
Like the md5 method but uses older crypt encryption, which is needed for pre-7.2 clients. md5 is preferred for 7.2 and later clients. The crypt method is not compatible with encrypting passwords in pg_shadow, and may fail if client and server machines have different implementations of the crypt() library routine.
Kerberos V4 is used to authenticate the user. This is available only for TCP/IP connections.
Kerberos V5 is used to authenticate the user. This is available only for TCP/IP connections.
The identity of the user as determined on login to the operating system is used by PostgreSQL to determine whether the user is allowed to connect as the requested database user. For TCP/IP connections the user's identity is determined by contacting the ident server on the client host. (Note that this is only as reliable as the remote ident server; ident authentication should never be used for remote hosts whose administrators are not trustworthy.) On operating systems supporting SO_PEERCRED requests for UNIX-domain sockets, ident authentication is possible for local connections; the system is then asked for the connecting user's identity.
On systems without SO_PEERCRED requests, ident authentication is only available for TCP/IP connections. As a workaround, it is possible to specify the localhost address 127.0.0.1 and make connections to this address.
The authentication option following the ident keyword specifies the name of an ident map that specifies which operating system users equate with which database users. See below for details.
This authentication type operates similarly to password, with the main difference that it will use PAM (Pluggable Authentication Modules) as the authentication mechanism. The authentication option following the pam keyword specifies the service name that will be passed to PAM. The default service name is postgresql. For more information about PAM, please read the Linux-PAM PAM Page (http://www.kernel.org/pub/linux/libs/pam/).
This field is interpreted differently depending on the authentication method, as described above.
Since the pg_hba.conf records are examined sequentially for each connection attempt, the order of the records is very significant. Typically, earlier records will have tight connection match parameters and weaker authentication methods, while later records will have looser match parameters and stronger authentication methods. For example, one might wish to use trust authentication for local TCP connections but require a password for remote TCP connections. In this case a record specifying trust authentication for connections from 127.0.0.1 would appear before a record specifying password authentication for a wider range of allowed client IP addresses.
The pg_hba.conf file is read on startup and when the postmaster receives a SIGHUP signal. If you edit the file on an active system, you will need to signal the postmaster (using pg_ctl reload or kill -HUP) to make it re-read the file.
An example of a pg_hba.conf file is shown below. See the Section called Authentication Methods for details on the different authentication methods.
Example 2-1. An example pg_hba.conf File
# TYPE DATABASE IP_ADDRESS MASK AUTHTYPE MAP # Allow any user on the local system to connect to any # database under any username, but only via an IP connection: host all 127.0.0.1 255.255.255.255 trust # The same, over UNIX-socket connections: local all trust # Allow any user from any host with IP address 192.168.93.x to # connect to database "template1" as the same username that # ident on that host identifies him as (typically his UNIX # username): host template1 192.168.93.0 255.255.255.0 ident sameuser # Allow a user from host 192.168.12.10 to connect to database # "template1" if the user's password in pg_shadow is correctly # supplied: host template1 192.168.12.10 255.255.255.255 crypt # In the absence of preceding "host" lines, these two lines # will reject all connection attempts from 192.168.54.1 (since # that entry will be matched first), but allow Kerberos # V5-validated connections from anywhere else on the Internet. # The zero mask means that no bits of the host IP # address are considered, so it matches any host: host all 192.168.54.1 255.255.255.255 reject host all 0.0.0.0 0.0.0.0 krb5 # Allow users from 192.168.x.x hosts to connect to any database, # if they pass the ident check. If, for example, ident says the # user is "bryanh" and he requests to connect as PostgreSQL # user "guest1", the connection is allowed if there is an entry # in pg_ident.conf for map "omicron" that says "bryanh" is # allowed to connect as "guest1": host all 192.168.0.0 255.255.0.0 ident omicron # If these are the only two lines for local connections, they allow # local users to connect only to their own databases (database named the # same as the user name), except for administrators who may connect to # all databases. The file $PGDATA/admins lists the user names who are # permitted to connect to all databases. Passwords are required in all # cases. (If you prefer to use ident authorization, an ident map can # serve a parallel purpose to the password list file used here.) local sameuser md5 local all |