| PostgreSQL 8.0.1 Documentation | ||||
|---|---|---|---|---|
| Prev | Fast Backward | Fast Forward | Next | |
postmaster
Synopsis
postmaster [-A 0 | 1 ] [-B nbuffers] [-c name=value] [-d debug-level] [-D datadir] [-F] [-h hostname] [-i] [-k directory] [-l] [-N max-connections] [-o extra-options] [-p port] [-S] [--name=value] [-n | -s]
Description
postmaster is the PostgreSQL multiuser database server. In order for a client application to access a database it connects (over a network or locally) to a running postmaster. The postmaster then starts a separate server process ("postgres") to handle the connection. The postmaster also manages the communication among server processes.
By default the postmaster starts in the foreground and prints log messages to the standard error stream. In practical applications the postmaster should be started as a background process, perhaps at boot time.
One postmaster always manages the data from exactly one database cluster. A database cluster is a collection of databases that is stored at a common file system location (the "data area"). More than one postmaster process can run on a system at one time, so long as they use different data areas and different communication ports (see below). A data area is created with initdb.
When the postmaster starts it needs to know the location of the data area. The location must be specified by the -D option or the PGDATA environment variable; there is no default. Typically, -D or PGDATA points directly to the data area directory created by initdb. Other possible file layouts are discussed in Section 16.4.1.
Options
postmaster accepts the following command line arguments. For a detailed discussion of the options consult Section 16.4. You can also save typing most of these options by setting up a configuration file.
- -A 0|1
Enables run-time assertion checks, which is a debugging aid to detect programming mistakes. This option is only available if assertions were enabled when PostgreSQL was compiled. If so, the default is on.
- -B nbuffers
Sets the number of shared buffers for use by the server processes. The default value of this parameter is chosen automatically by initdb; refer to Section 16.4.3.1 for more information.
- -c name=value
Sets a named run-time parameter. The configuration parameters supported by PostgreSQL are described in Section 16.4. Most of the other command line options are in fact short forms of such a parameter assignment. -c can appear multiple times to set multiple parameters.
- -d debug-level
Sets the debug level. The higher this value is set, the more debugging output is written to the server log. Values are from 1 to 5.
- -D datadir
Specifies the file system location of the data directory or configuration file(s). See Section 16.4.1 for details.
- -F
Disables
fsynccalls for improved performance, at the risk of data corruption in the event of a system crash. Specifying this option is equivalent to disabling the fsync configuration parameter. Read the detailed documentation before using this!--fsync=true has the opposite effect of this option.
- -h hostname
Specifies the IP host name or address on which the postmaster is to listen for TCP/IP connections from client applications. The value can also be a space-separated list of addresses, or * to specify listening on all available interfaces. An empty value specifies not listening on any IP addresses, in which case only Unix-domain sockets can be used to connect to the postmaster. Defaults to listening only on localhost. Specifying this option is equivalent to setting the listen_addresses configuration parameter.
- -i
Allows remote clients to connect via TCP/IP (Internet domain) connections. Without this option, only local connections are accepted. This option is equivalent to setting listen_addresses to * in postgresql.conf or via -h.
This option is deprecated since it does not allow access to the full functionality of listen_addresses. It's usually better to set listen_addresses directly.
- -k directory
Specifies the directory of the Unix-domain socket on which the postmaster is to listen for connections from client applications. The default is normally /tmp, but can be changed at build time.
- -l
Enables secure connections using SSL. PostgreSQL must have been compiled with support for SSL for this option to be available. For more information on using SSL, refer to Section 16.7.
- -N max-connections
Sets the maximum number of client connections that this postmaster will accept. By default, this value is 32, but it can be set as high as your system will support. (Note that -B is required to be at least twice -N. See Section 16.5 for a discussion of system resource requirements for large numbers of client connections.) Specifying this option is equivalent to setting the max_connections configuration parameter.
- -o extra-options
The command line-style options specified in extra-options are passed to all server processes started by this postmaster. See postgres for possibilities. If the option string contains any spaces, the entire string must be quoted.
- -p port
Specifies the TCP/IP port or local Unix domain socket file extension on which the postmaster is to listen for connections from client applications. Defaults to the value of the PGPORT environment variable, or if PGPORT is not set, then defaults to the value established during compilation (normally 5432). If you specify a port other than the default port, then all client applications must specify the same port using either command-line options or PGPORT.
- -S
Specifies that the postmaster process should start up in silent mode. That is, it will disassociate from the user's (controlling) terminal, start its own process group, and redirect its standard output and standard error to /dev/null.
Using this switch discards all logging output, which is probably not what you want, since it makes it very difficult to troubleshoot problems. See below for a better way to start the postmaster in the background.
--silent-mode=false has the opposite effect of this option.
- --name=value
Sets a named run-time parameter; a shorter form of -c.
Two additional command line options are available for debugging problems that cause a server process to die abnormally. The ordinary strategy in this situation is to notify all other server processes that they must terminate and then reinitialize the shared memory and semaphores. This is because an errant server process could have corrupted some shared state before terminating. These options select alternative behaviors of the postmaster in this situation. Neither option is intended for use in ordinary operation.
These special-case options are:
- -n
postmaster will not reinitialize shared data structures. A knowledgeable system programmer can then use a debugger to examine shared memory and semaphore state.
- -s
postmaster will stop all other server processes by sending the signal SIGSTOP, but will not cause them to terminate. This permits system programmers to collect core dumps from all server processes by hand.
Environment
- PGCLIENTENCODING
Default character encoding used by clients. (The clients may override this individually.) This value can also be set in the configuration file.
- PGDATA
Default data directory location
- PGDATESTYLE
Default value of the DateStyle run-time parameter. (The use of this environment variable is deprecated.)
- PGPORT
Default port (preferably set in the configuration file)
- TZ
Server time zone
Diagnostics
A failure message mentioning semget or shmget probably indicates you need to configure your kernel to provide adequate shared memory and semaphores. For more discussion see Section 16.5.
Tip: You may be able to postpone reconfiguring your kernel by decreasing shared_buffers to reduce the shared memory consumption of PostgreSQL, and/or by reducing max_connections to reduce the semaphore consumption.
A failure message suggesting that another postmaster is already running should be checked carefully, for example by using the command
or
depending on your system. If you are certain that no conflicting postmaster is running, you may remove the lock file mentioned in the message and try again.
A failure message indicating inability to bind to a port may indicate that that port is already in use by some non-PostgreSQL process. You may also get this error if you terminate the postmaster and immediately restart it using the same port; in this case, you must simply wait a few seconds until the operating system closes the port before trying again. Finally, you may get this error if you specify a port number that your operating system considers to be reserved. For example, many versions of Unix consider port numbers under 1024 to be "trusted" and only permit the Unix superuser to access them.
Notes
If at all possible, do not use SIGKILL to kill the postmaster. Doing so will prevent postmaster from freeing the system resources (e.g., shared memory and semaphores) that it holds before terminating. This may cause problems for starting a fresh postmaster run.
To terminate the postmaster normally, the signals SIGTERM, SIGINT, or SIGQUIT can be used. The first will wait for all clients to terminate before quitting, the second will forcefully disconnect all clients, and the third will quit immediately without proper shutdown, resulting in a recovery run during restart. The SIGHUP signal will reload the server configuration files.
The utility command pg_ctl can be used to start and shut down the postmaster safely and comfortably.
The -- options will not work on FreeBSD or OpenBSD. Use -c instead. This is a bug in the affected operating systems; a future release of PostgreSQL will provide a workaround if this is not fixed.
Examples
To start postmaster in the background using default values, type:
To start postmaster with a specific port:
This command will start up postmaster communicating through the port 1234. In order to connect to this postmaster using psql, you would need to run it as
or set the environment variable PGPORT:
Named run-time parameters can be set in either of these styles:
Either form overrides whatever setting might exist for work_mem in postgresql.conf. Notice that underscores in parameter names can be written as either underscore or dash on the command line.
Tip: Except for short-term experiments, it's probably better practice to edit the setting in postgresql.conf than to rely on a command-line switch to set a parameter.
