4 postgres — PostgreSQL database server
12 postgres is the PostgreSQL database server. In order for a client
13 application to access a database it connects (over a network or
14 locally) to a running postgres instance. The postgres instance then
15 starts a separate server process to handle the connection.
17 One postgres instance always manages the data of exactly one database
18 cluster. A database cluster is a collection of databases that is stored
19 at a common file system location (the “data area”). More than one
20 postgres instance can run on a system at one time, so long as they use
21 different data areas and different communication ports (see below).
22 When postgres starts it needs to know the location of the data area.
23 The location must be specified by the -D option or the PGDATA
24 environment variable; there is no default. Typically, -D or PGDATA
25 points directly to the data area directory created by initdb. Other
26 possible file layouts are discussed in Section 19.2.
28 By default postgres starts in the foreground and prints log messages to
29 the standard error stream. In practical applications postgres should be
30 started as a background process, perhaps at boot time.
32 The postgres command can also be called in single-user mode. The
33 primary use for this mode is during bootstrapping by initdb. Sometimes
34 it is used for debugging or disaster recovery; note that running a
35 single-user server is not truly suitable for debugging the server,
36 since no realistic interprocess communication and locking will happen.
37 When invoked in single-user mode from the shell, the user can enter
38 queries and the results will be printed to the screen, but in a form
39 that is more useful for developers than end users. In the single-user
40 mode, the session user will be set to the user with ID 1, and implicit
41 superuser powers are granted to this user. This user does not actually
42 have to exist, so the single-user mode can be used to manually recover
43 from certain kinds of accidental damage to the system catalogs.
47 postgres accepts the following command-line arguments. For a detailed
48 discussion of the options consult Chapter 19. You can save typing most
49 of these options by setting up a configuration file. Some (safe)
50 options can also be set from the connecting client in an
51 application-dependent way to apply only for that session. For example,
52 if the environment variable PGOPTIONS is set, then libpq-based clients
53 will pass that string to the server, which will interpret it as
54 postgres command-line options.
59 Sets the number of shared buffers for use by the server
60 processes. The default value of this parameter is chosen
61 automatically by initdb. Specifying this option is equivalent to
62 setting the shared_buffers configuration parameter.
65 Sets a named run-time parameter. The configuration parameters
66 supported by PostgreSQL are described in Chapter 19. Most of the
67 other command line options are in fact short forms of such a
68 parameter assignment. -c can appear multiple times to set
72 Prints the value of the named run-time parameter, and exits.
73 (See the -c option above for details.) This returns values from
74 postgresql.conf, modified by any parameters supplied in this
75 invocation. It does not reflect parameters supplied when the
78 This can be used on a running server for most parameters.
79 However, the server must be shut down for some runtime-computed
80 parameters (e.g., shared_memory_size,
81 shared_memory_size_in_huge_pages, and wal_segment_size).
83 This option is meant for other programs that interact with a
84 server instance, such as pg_ctl, to query configuration
85 parameter values. User-facing applications should instead use
86 SHOW or the pg_settings view.
89 Sets the debug level. The higher this value is set, the more
90 debugging output is written to the server log. Values are from 1
91 to 5. It is also possible to pass -d 0 for a specific session,
92 which will prevent the server log level of the parent postgres
93 process from being propagated to this session.
96 Specifies the file system location of the database configuration
97 files. See Section 19.2 for details.
100 Sets the default date style to “European”, that is DMY ordering
101 of input date fields. This also causes the day to be printed
102 before the month in certain date output formats. See Section 8.5
103 for more information.
106 Disables fsync calls for improved performance, at the risk of
107 data corruption in the event of a system crash. Specifying this
108 option is equivalent to disabling the fsync configuration
109 parameter. Read the detailed documentation before using this!
112 Specifies the IP host name or address on which postgres is to
113 listen for TCP/IP connections from client applications. The
114 value can also be a comma-separated list of addresses, or * to
115 specify listening on all available interfaces. An empty value
116 specifies not listening on any IP addresses, in which case only
117 Unix-domain sockets can be used to connect to the server.
118 Defaults to listening only on localhost. Specifying this option
119 is equivalent to setting the listen_addresses configuration
123 Allows remote clients to connect via TCP/IP (Internet domain)
124 connections. Without this option, only local connections are
125 accepted. This option is equivalent to setting listen_addresses
126 to * in postgresql.conf or via -h.
128 This option is deprecated since it does not allow access to the
129 full functionality of listen_addresses. It's usually better to
130 set listen_addresses directly.
133 Specifies the directory of the Unix-domain socket on which
134 postgres is to listen for connections from client applications.
135 The value can also be a comma-separated list of directories. An
136 empty value specifies not listening on any Unix-domain sockets,
137 in which case only TCP/IP sockets can be used to connect to the
138 server. The default value is normally /tmp, but that can be
139 changed at build time. Specifying this option is equivalent to
140 setting the unix_socket_directories configuration parameter.
143 Enables secure connections using SSL. PostgreSQL must have been
144 compiled with support for SSL for this option to be available.
145 For more information on using SSL, refer to Section 18.9.
148 Sets the maximum number of client connections that this server
149 will accept. The default value of this parameter is chosen
150 automatically by initdb. Specifying this option is equivalent to
151 setting the max_connections configuration parameter.
154 Specifies the TCP/IP port or local Unix domain socket file
155 extension on which postgres is to listen for connections from
156 client applications. Defaults to the value of the PGPORT
157 environment variable, or if PGPORT is not set, then defaults to
158 the value established during compilation (normally 5432). If you
159 specify a port other than the default port, then all client
160 applications must specify the same port using either
161 command-line options or PGPORT.
164 Print time information and other statistics at the end of each
165 command. This is useful for benchmarking or for use in tuning
166 the number of buffers.
169 Specifies the base amount of memory to be used by sorts and hash
170 tables before resorting to temporary disk files. See the
171 description of the work_mem configuration parameter in
176 Print the postgres version and exit.
179 Sets a named run-time parameter; a shorter form of -c.
182 This option dumps out the server's internal configuration
183 variables, descriptions, and defaults in tab-delimited COPY
184 format. It is designed primarily for use by administration
189 Show help about postgres command line arguments, and exit.
191 Semi-Internal Options
193 The options described here are used mainly for debugging purposes, and
194 in some cases to assist with recovery of severely damaged databases.
195 There should be no reason to use them in a production database setup.
196 They are listed here only for use by PostgreSQL system developers.
197 Furthermore, these options might change or be removed in a future
198 release without notice.
200 -f { s | i | o | b | t | n | m | h }
201 Forbids the use of particular scan and join methods: s and i
202 disable sequential and index scans respectively, o, b and t
203 disable index-only scans, bitmap index scans, and TID scans
204 respectively, while n, m, and h disable nested-loop, merge and
205 hash joins respectively.
207 Neither sequential scans nor nested-loop joins can be disabled
208 completely; the -fs and -fn options simply discourage the
209 optimizer from using those plan types if it has any other
213 Allows the structure of system tables to be modified. This is
217 Ignore system indexes when reading system tables, but still
218 update the indexes when modifying the tables. This is useful
219 when recovering from damaged system indexes.
221 -t pa[rser] | pl[anner] | e[xecutor]
222 Print timing statistics for each query relating to each of the
223 major system modules. This option cannot be used together with
227 This option is for debugging problems that cause a server
228 process to die abnormally. The ordinary strategy in this
229 situation is to notify all other server processes that they must
230 terminate, by sending them SIGQUIT signals. With this option,
231 SIGABRT will be sent instead, resulting in production of core
235 Specifies the version number of the frontend/backend protocol to
236 be used for a particular session. This option is for internal
240 A delay of this many seconds occurs when a new server process is
241 started, after it conducts the authentication procedure. This is
242 intended to give an opportunity to attach to the server process
245 Options for Single-User Mode
247 The following options only apply to the single-user mode (see
248 Single-User Mode below).
251 Selects the single-user mode. This must be the first argument on
255 Specifies the name of the database to be accessed. This must be
256 the last argument on the command line. If it is omitted it
257 defaults to the user name.
260 Echo all commands to standard output before executing them.
263 Use semicolon followed by two newlines, rather than just
264 newline, as the command entry terminator.
267 Send all server log output to filename. This option is only
268 honored when supplied as a command-line option.
273 Default character encoding used by clients. (The clients can
274 override this individually.) This value can also be set in the
278 Default data directory location
281 Default value of the DateStyle run-time parameter. (The use of
282 this environment variable is deprecated.)
285 Default port number (preferably set in the configuration file)
289 A failure message mentioning semget or shmget probably indicates you
290 need to configure your kernel to provide adequate shared memory and
291 semaphores. For more discussion see Section 18.4. You might be able to
292 postpone reconfiguring your kernel by decreasing shared_buffers to
293 reduce the shared memory consumption of PostgreSQL, and/or by reducing
294 max_connections to reduce the semaphore consumption.
296 A failure message suggesting that another server is already running
297 should be checked carefully, for example by using the command
298 $ ps ax | grep postgres
301 $ ps -ef | grep postgres
303 depending on your system. If you are certain that no conflicting server
304 is running, you can remove the lock file mentioned in the message and
307 A failure message indicating inability to bind to a port might indicate
308 that that port is already in use by some non-PostgreSQL process. You
309 might also get this error if you terminate postgres and immediately
310 restart it using the same port; in this case, you must simply wait a
311 few seconds until the operating system closes the port before trying
312 again. Finally, you might get this error if you specify a port number
313 that your operating system considers to be reserved. For example, many
314 versions of Unix consider port numbers under 1024 to be “trusted” and
315 only permit the Unix superuser to access them.
319 The utility command pg_ctl can be used to start and shut down the
320 postgres server safely and comfortably.
322 If at all possible, do not use SIGKILL to kill the main postgres
323 server. Doing so will prevent postgres from freeing the system
324 resources (e.g., shared memory and semaphores) that it holds before
325 terminating. This might cause problems for starting a fresh postgres
328 To terminate the postgres server normally, the signals SIGTERM, SIGINT,
329 or SIGQUIT can be used. The first will wait for all clients to
330 terminate before quitting, the second will forcefully disconnect all
331 clients, and the third will quit immediately without proper shutdown,
332 resulting in a recovery run during restart.
334 The SIGHUP signal will reload the server configuration files. It is
335 also possible to send SIGHUP to an individual server process, but that
336 is usually not sensible.
338 To cancel a running query, send the SIGINT signal to the process
339 running that command. To terminate a backend process cleanly, send
340 SIGTERM to that process. See also pg_cancel_backend and
341 pg_terminate_backend in Section 9.28.2 for the SQL-callable equivalents
342 of these two actions.
344 The postgres server uses SIGQUIT to tell subordinate server processes
345 to terminate without normal cleanup. This signal should not be used by
346 users. It is also unwise to send SIGKILL to a server process — the main
347 postgres process will interpret this as a crash and will force all the
348 sibling processes to quit as part of its standard crash-recovery
353 The -- options will not work on FreeBSD or OpenBSD. Use -c instead.
354 This is a bug in the affected operating systems; a future release of
355 PostgreSQL will provide a workaround if this is not fixed.
359 To start a single-user mode server, use a command like
360 postgres --single -D /usr/local/pgsql/data other-options my_database
362 Provide the correct path to the database directory with -D, or make
363 sure that the environment variable PGDATA is set. Also specify the name
364 of the particular database you want to work in.
366 Normally, the single-user mode server treats newline as the command
367 entry terminator; there is no intelligence about semicolons, as there
368 is in psql. To continue a command across multiple lines, you must type
369 backslash just before each newline except the last one. The backslash
370 and adjacent newline are both dropped from the input command. Note that
371 this will happen even when within a string literal or comment.
373 But if you use the -j command line switch, a single newline does not
374 terminate command entry; instead, the sequence
375 semicolon-newline-newline does. That is, type a semicolon immediately
376 followed by a completely empty line. Backslash-newline is not treated
377 specially in this mode. Again, there is no intelligence about such a
378 sequence appearing within a string literal or comment.
380 In either input mode, if you type a semicolon that is not just before
381 or part of a command entry terminator, it is considered a command
382 separator. When you do type a command entry terminator, the multiple
383 statements you've entered will be executed as a single transaction.
385 To quit the session, type EOF (Control+D, usually). If you've entered
386 any text since the last command entry terminator, then EOF will be
387 taken as a command entry terminator, and another EOF will be needed to
390 Note that the single-user mode server does not provide sophisticated
391 line-editing features (no command history, for example). Single-user
392 mode also does not do any background processing, such as automatic
393 checkpoints or replication.
397 To start postgres in the background using default values, type:
398 $ nohup postgres >logfile 2>&1 </dev/null &
400 To start postgres with a specific port, e.g., 1234:
403 To connect to this server using psql, specify this port with the -p
407 or set the environment variable PGPORT:
411 Named run-time parameters can be set in either of these styles:
412 $ postgres -c work_mem=1234
413 $ postgres --work-mem=1234
415 Either form overrides whatever setting might exist for work_mem in
416 postgresql.conf. Notice that underscores in parameter names can be
417 written as either underscore or dash on the command line. Except for
418 short-term experiments, it's probably better practice to edit the
419 setting in postgresql.conf than to rely on a command-line switch to set
begriffs.com / projects / ai-pg / blob