<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>34.1. Database Connection Control Functions</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="libpq.html" title="Chapter 34. libpq - C Library" /><link rel="next" href="libpq-status.html" title="34.2. Connection Status Functions" /></head><body><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">34.1. Database Connection Control Functions</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="libpq.html" title="Chapter 34. libpq - C Library">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="libpq.html" title="Chapter 34. libpq - C Library">Up</a></td><th width="60%" align="center">Chapter 34. <span xmlns="http://www.w3.org/1999/xhtml" class="application">libpq</span> - C Library</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 11.5 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="libpq-status.html" title="34.2. Connection Status Functions">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="LIBPQ-CONNECT"><div class="titlepage"><div><div><h2 class="title" style="clear: both">34.1. Database Connection Control Functions</h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="libpq-connect.html#LIBPQ-CONNSTRING">34.1.1. Connection Strings</a></span></dt><dt><span class="sect2"><a href="libpq-connect.html#LIBPQ-PARAMKEYWORDS">34.1.2. Parameter Key Words</a></span></dt></dl></div><p>
The following functions deal with making a connection to a
<span class="productname">PostgreSQL</span> backend server. An
application program can have several backend connections open at
one time. (One reason to do that is to access more than one
database.) Each connection is represented by a
<code class="structname">PGconn</code><a id="id-1.7.3.8.2.3" class="indexterm"></a> object, which
is obtained from the function <code class="function">PQconnectdb</code>,
<code class="function">PQconnectdbParams</code>, or
<code class="function">PQsetdbLogin</code>. Note that these functions will always
return a non-null object pointer, unless perhaps there is too
little memory even to allocate the <code class="structname">PGconn</code> object.
The <code class="function">PQstatus</code> function should be called to check
the return value for a successful connection before queries are sent
via the connection object.
</p><div class="warning"><h3 class="title">Warning</h3><p>
If untrusted users have access to a database that has not adopted a
<a class="link" href="ddl-schemas.html#DDL-SCHEMAS-PATTERNS" title="5.8.6. Usage Patterns">secure schema usage pattern</a>,
begin each session by removing publicly-writable schemas from
<code class="varname">search_path</code>. One can set parameter key
word <code class="literal">options</code> to
value <code class="literal">-csearch_path=</code>. Alternately, one can
issue <code class="literal">PQexec(<em class="replaceable"><code>conn</code></em>, "SELECT
pg_catalog.set_config('search_path', '', false)")</code> after
connecting. This consideration is not specific
to <span class="application">libpq</span>; it applies to every interface for
executing arbitrary SQL commands.
</p></div><p>
</p><div class="warning"><h3 class="title">Warning</h3><p>
On Unix, forking a process with open libpq connections can lead to
unpredictable results because the parent and child processes share
the same sockets and operating system resources. For this reason,
such usage is not recommended, though doing an <code class="function">exec</code> from
the child process to load a new executable is safe.
</p></div><p>
</p><div class="note"><h3 class="title">Note</h3><p>
On Windows, there is a way to improve performance if a single
database connection is repeatedly started and shutdown. Internally,
libpq calls <code class="function">WSAStartup()</code> and <code class="function">WSACleanup()</code> for connection startup
and shutdown, respectively. <code class="function">WSAStartup()</code> increments an internal
Windows library reference count which is decremented by <code class="function">WSACleanup()</code>.
When the reference count is just one, calling <code class="function">WSACleanup()</code> frees
all resources and all DLLs are unloaded. This is an expensive
operation. To avoid this, an application can manually call
<code class="function">WSAStartup()</code> so resources will not be freed when the last database
connection is closed.
</p></div><p>
</p><div class="variablelist"><dl class="variablelist"><dt id="LIBPQ-PQCONNECTDBPARAMS"><span class="term"><code class="function">PQconnectdbParams</code><a id="id-1.7.3.8.2.12.1.1.2" class="indexterm"></a></span></dt><dd><p>
Makes a new connection to the database server.
</p><pre class="synopsis">
PGconn *PQconnectdbParams(const char * const *keywords,
const char * const *values,
int expand_dbname);
</pre><p>
</p><p>
This function opens a new database connection using the parameters taken
from two <code class="symbol">NULL</code>-terminated arrays. The first,
<code class="literal">keywords</code>, is defined as an array of strings, each one
being a key word. The second, <code class="literal">values</code>, gives the value
for each key word. Unlike <code class="function">PQsetdbLogin</code> below, the parameter
set can be extended without changing the function signature, so use of
this function (or its nonblocking analogs <code class="function">PQconnectStartParams</code>
and <code class="function">PQconnectPoll</code>) is preferred for new application
programming.
</p><p>
The currently recognized parameter key words are listed in
<a class="xref" href="libpq-connect.html#LIBPQ-PARAMKEYWORDS" title="34.1.2. Parameter Key Words">Section 34.1.2</a>.
</p><p>
When <code class="literal">expand_dbname</code> is non-zero, the
<em class="parameter"><code>dbname</code></em> key word value is allowed to be recognized
as a connection string. Only the first occurrence of
<em class="parameter"><code>dbname</code></em> is expanded this way, any subsequent
<em class="parameter"><code>dbname</code></em> value is processed as plain database name. More
details on the possible connection string formats appear in
<a class="xref" href="libpq-connect.html#LIBPQ-CONNSTRING" title="34.1.1. Connection Strings">Section 34.1.1</a>.
</p><p>
The passed arrays can be empty to use all default parameters, or can
contain one or more parameter settings. They should be matched in length.
Processing will stop at the first <code class="symbol">NULL</code> element
in the <code class="literal">keywords</code> array.
</p><p>
If any parameter is <code class="symbol">NULL</code> or an empty string, the corresponding
environment variable (see <a class="xref" href="libpq-envars.html" title="34.14. Environment Variables">Section 34.14</a>) is checked.
If the environment variable is not set either, then the indicated
built-in defaults are used.
</p><p>
In general key words are processed from the beginning of these arrays in index
order. The effect of this is that when key words are repeated, the last processed
value is retained. Therefore, through careful placement of the
<em class="parameter"><code>dbname</code></em> key word, it is possible to determine what may
be overridden by a <em class="parameter"><code>conninfo</code></em> string, and what may not.
</p></dd><dt id="LIBPQ-PQCONNECTDB"><span class="term"><code class="function">PQconnectdb</code><a id="id-1.7.3.8.2.12.2.1.2" class="indexterm"></a></span></dt><dd><p>
Makes a new connection to the database server.
</p><pre class="synopsis">
PGconn *PQconnectdb(const char *conninfo);
</pre><p>
</p><p>
This function opens a new database connection using the parameters taken
from the string <code class="literal">conninfo</code>.
</p><p>
The passed string can be empty to use all default parameters, or it can
contain one or more parameter settings separated by whitespace,
or it can contain a <acronym class="acronym">URI</acronym>.
See <a class="xref" href="libpq-connect.html#LIBPQ-CONNSTRING" title="34.1.1. Connection Strings">Section 34.1.1</a> for details.
</p></dd><dt id="LIBPQ-PQSETDBLOGIN"><span class="term"><code class="function">PQsetdbLogin</code><a id="id-1.7.3.8.2.12.3.1.2" class="indexterm"></a></span></dt><dd><p>
Makes a new connection to the database server.
</p><pre class="synopsis">
PGconn *PQsetdbLogin(const char *pghost,
const char *pgport,
const char *pgoptions,
const char *pgtty,
const char *dbName,
const char *login,
const char *pwd);
</pre><p>
</p><p>
This is the predecessor of <code class="function">PQconnectdb</code> with a fixed
set of parameters. It has the same functionality except that the
missing parameters will always take on default values. Write <code class="symbol">NULL</code> or an
empty string for any one of the fixed parameters that is to be defaulted.
</p><p>
If the <em class="parameter"><code>dbName</code></em> contains
an <code class="symbol">=</code> sign or has a valid connection <acronym class="acronym">URI</acronym> prefix, it
is taken as a <em class="parameter"><code>conninfo</code></em> string in exactly the same way as
if it had been passed to <code class="function">PQconnectdb</code>, and the remaining
parameters are then applied as specified for <code class="function">PQconnectdbParams</code>.
</p></dd><dt id="LIBPQ-PQSETDB"><span class="term"><code class="function">PQsetdb</code><a id="id-1.7.3.8.2.12.4.1.2" class="indexterm"></a></span></dt><dd><p>
Makes a new connection to the database server.
</p><pre class="synopsis">
PGconn *PQsetdb(char *pghost,
char *pgport,
char *pgoptions,
char *pgtty,
char *dbName);
</pre><p>
</p><p>
This is a macro that calls <code class="function">PQsetdbLogin</code> with null pointers
for the <em class="parameter"><code>login</code></em> and <em class="parameter"><code>pwd</code></em> parameters. It is provided
for backward compatibility with very old programs.
</p></dd><dt id="LIBPQ-PQCONNECTSTARTPARAMS"><span class="term"><code class="function">PQconnectStartParams</code><a id="id-1.7.3.8.2.12.5.1.2" class="indexterm"></a><br /></span><span class="term"><code class="function">PQconnectStart</code><a id="id-1.7.3.8.2.12.5.2.2" class="indexterm"></a><br /></span><span class="term"><code class="function">PQconnectPoll</code><a id="id-1.7.3.8.2.12.5.3.2" class="indexterm"></a></span></dt><dd><p>
<a id="id-1.7.3.8.2.12.5.4.1.1" class="indexterm"></a>
Make a connection to the database server in a nonblocking manner.
</p><pre class="synopsis">
PGconn *PQconnectStartParams(const char * const *keywords,
const char * const *values,
int expand_dbname);
PGconn *PQconnectStart(const char *conninfo);
PostgresPollingStatusType PQconnectPoll(PGconn *conn);
</pre><p>
</p><p>
These three functions are used to open a connection to a database server such
that your application's thread of execution is not blocked on remote I/O
whilst doing so. The point of this approach is that the waits for I/O to
complete can occur in the application's main loop, rather than down inside
<code class="function">PQconnectdbParams</code> or <code class="function">PQconnectdb</code>, and so the
application can manage this operation in parallel with other activities.
</p><p>
With <code class="function">PQconnectStartParams</code>, the database connection is made
using the parameters taken from the <code class="literal">keywords</code> and
<code class="literal">values</code> arrays, and controlled by <code class="literal">expand_dbname</code>,
as described above for <code class="function">PQconnectdbParams</code>.
</p><p>
With <code class="function">PQconnectStart</code>, the database connection is made
using the parameters taken from the string <code class="literal">conninfo</code> as
described above for <code class="function">PQconnectdb</code>.
</p><p>
Neither <code class="function">PQconnectStartParams</code> nor <code class="function">PQconnectStart</code>
nor <code class="function">PQconnectPoll</code> will block, so long as a number of
restrictions are met:
</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p>
The <code class="literal">hostaddr</code> parameter must be used appropriately
to prevent DNS queries from being made. See the documentation of
this parameter in <a class="xref" href="libpq-connect.html#LIBPQ-PARAMKEYWORDS" title="34.1.2. Parameter Key Words">Section 34.1.2</a> for details.
</p></li><li class="listitem"><p>
If you call <code class="function">PQtrace</code>, ensure that the stream object
into which you trace will not block.
</p></li><li class="listitem"><p>
You must ensure that the socket is in the appropriate state
before calling <code class="function">PQconnectPoll</code>, as described below.
</p></li></ul></div><p>
</p><p>
To begin a nonblocking connection request,
call <code class="function">PQconnectStart</code>
or <code class="function">PQconnectStartParams</code>. If the result is null,
then <span class="application">libpq</span> has been unable to allocate a
new <code class="structname">PGconn</code> structure. Otherwise, a
valid <code class="structname">PGconn</code> pointer is returned (though not
yet representing a valid connection to the database). Next
call <code class="literal">PQstatus(conn)</code>. If the result
is <code class="symbol">CONNECTION_BAD</code>, the connection attempt has already
failed, typically because of invalid connection parameters.
</p><p>
If <code class="function">PQconnectStart</code>
or <code class="function">PQconnectStartParams</code> succeeds, the next stage
is to poll <span class="application">libpq</span> so that it can proceed with
the connection sequence.
Use <code class="function">PQsocket(conn)</code> to obtain the descriptor of the
socket underlying the database connection.
(Caution: do not assume that the socket remains the same
across <code class="function">PQconnectPoll</code> calls.)
Loop thus: If <code class="function">PQconnectPoll(conn)</code> last returned
<code class="symbol">PGRES_POLLING_READING</code>, wait until the socket is ready to
read (as indicated by <code class="function">select()</code>, <code class="function">poll()</code>, or
similar system function).
Then call <code class="function">PQconnectPoll(conn)</code> again.
Conversely, if <code class="function">PQconnectPoll(conn)</code> last returned
<code class="symbol">PGRES_POLLING_WRITING</code>, wait until the socket is ready
to write, then call <code class="function">PQconnectPoll(conn)</code> again.
On the first iteration, i.e. if you have yet to call
<code class="function">PQconnectPoll</code>, behave as if it last returned
<code class="symbol">PGRES_POLLING_WRITING</code>. Continue this loop until
<code class="function">PQconnectPoll(conn)</code> returns
<code class="symbol">PGRES_POLLING_FAILED</code>, indicating the connection procedure
has failed, or <code class="symbol">PGRES_POLLING_OK</code>, indicating the connection
has been successfully made.
</p><p>
At any time during connection, the status of the connection can be
checked by calling <code class="function">PQstatus</code>. If this call returns <code class="symbol">CONNECTION_BAD</code>, then the
connection procedure has failed; if the call returns <code class="function">CONNECTION_OK</code>, then the
connection is ready. Both of these states are equally detectable
from the return value of <code class="function">PQconnectPoll</code>, described above. Other states might also occur
during (and only during) an asynchronous connection procedure. These
indicate the current stage of the connection procedure and might be useful
to provide feedback to the user for example. These statuses are:
</p><div class="variablelist"><dl class="variablelist"><dt id="LIBPQ-CONNECTION-STARTED"><span class="term"><code class="symbol">CONNECTION_STARTED</code></span></dt><dd><p>
Waiting for connection to be made.
</p></dd><dt id="LIBPQ-CONNECTION-MADE"><span class="term"><code class="symbol">CONNECTION_MADE</code></span></dt><dd><p>
Connection OK; waiting to send.
</p></dd><dt id="LIBPQ-CONNECTION-AWAITING-RESPONSE"><span class="term"><code class="symbol">CONNECTION_AWAITING_RESPONSE</code></span></dt><dd><p>
Waiting for a response from the server.
</p></dd><dt id="LIBPQ-CONNECTION-AUTH-OK"><span class="term"><code class="symbol">CONNECTION_AUTH_OK</code></span></dt><dd><p>
Received authentication; waiting for backend start-up to finish.
</p></dd><dt id="LIBPQ-CONNECTION-SSL-STARTUP"><span class="term"><code class="symbol">CONNECTION_SSL_STARTUP</code></span></dt><dd><p>
Negotiating SSL encryption.
</p></dd><dt id="LIBPQ-CONNECTION-SETENV"><span class="term"><code class="symbol">CONNECTION_SETENV</code></span></dt><dd><p>
Negotiating environment-driven parameter settings.
</p></dd><dt id="LIBPQ-CONNECTION-CHECK-WRITABLE"><span class="term"><code class="symbol">CONNECTION_CHECK_WRITABLE</code></span></dt><dd><p>
Checking if connection is able to handle write transactions.
</p></dd><dt id="LIBPQ-CONNECTION-CONSUME"><span class="term"><code class="symbol">CONNECTION_CONSUME</code></span></dt><dd><p>
Consuming any remaining response messages on connection.
</p></dd></dl></div><p>
Note that, although these constants will remain (in order to maintain
compatibility), an application should never rely upon these occurring in a
particular order, or at all, or on the status always being one of these
documented values. An application might do something like this:
</p><pre class="programlisting">
switch(PQstatus(conn))
{
case CONNECTION_STARTED:
feedback = "Connecting...";
break;
case CONNECTION_MADE:
feedback = "Connected to server...";
break;
.
.
.
default:
feedback = "Connecting...";
}
</pre><p>
</p><p>
The <code class="literal">connect_timeout</code> connection parameter is ignored
when using <code class="function">PQconnectPoll</code>; it is the application's
responsibility to decide whether an excessive amount of time has elapsed.
Otherwise, <code class="function">PQconnectStart</code> followed by a
<code class="function">PQconnectPoll</code> loop is equivalent to
<code class="function">PQconnectdb</code>.
</p><p>
Note that when <code class="function">PQconnectStart</code>
or <code class="function">PQconnectStartParams</code> returns a non-null
pointer, you must call <code class="function">PQfinish</code> when you are
finished with it, in order to dispose of the structure and any
associated memory blocks. This must be done even if the connection
attempt fails or is abandoned.
</p></dd><dt id="LIBPQ-PQCONNDEFAULTS"><span class="term"><code class="function">PQconndefaults</code><a id="id-1.7.3.8.2.12.6.1.2" class="indexterm"></a></span></dt><dd><p>
Returns the default connection options.
</p><pre class="synopsis">
PQconninfoOption *PQconndefaults(void);
typedef struct
{
char *keyword; /* The keyword of the option */
char *envvar; /* Fallback environment variable name */
char *compiled; /* Fallback compiled in default value */
char *val; /* Option's current value, or NULL */
char *label; /* Label for field in connect dialog */
char *dispchar; /* Indicates how to display this field
in a connect dialog. Values are:
"" Display entered value as is
"*" Password field - hide value
"D" Debug option - don't show by default */
int dispsize; /* Field size in characters for dialog */
} PQconninfoOption;
</pre><p>
</p><p>
Returns a connection options array. This can be used to determine
all possible <code class="function">PQconnectdb</code> options and their
current default values. The return value points to an array of
<code class="structname">PQconninfoOption</code> structures, which ends
with an entry having a null <code class="structfield">keyword</code> pointer. The
null pointer is returned if memory could not be allocated. Note that
the current default values (<code class="structfield">val</code> fields)
will depend on environment variables and other context. A
missing or invalid service file will be silently ignored. Callers
must treat the connection options data as read-only.
</p><p>
After processing the options array, free it by passing it to
<code class="function">PQconninfoFree</code>. If this is not done, a small amount of memory
is leaked for each call to <code class="function">PQconndefaults</code>.
</p></dd><dt id="LIBPQ-PQCONNINFO"><span class="term"><code class="function">PQconninfo</code><a id="id-1.7.3.8.2.12.7.1.2" class="indexterm"></a></span></dt><dd><p>
Returns the connection options used by a live connection.
</p><pre class="synopsis">
PQconninfoOption *PQconninfo(PGconn *conn);
</pre><p>
</p><p>
Returns a connection options array. This can be used to determine
all possible <code class="function">PQconnectdb</code> options and the
values that were used to connect to the server. The return
value points to an array of <code class="structname">PQconninfoOption</code>
structures, which ends with an entry having a null <code class="structfield">keyword</code>
pointer. All notes above for <code class="function">PQconndefaults</code> also
apply to the result of <code class="function">PQconninfo</code>.
</p></dd><dt id="LIBPQ-PQCONNINFOPARSE"><span class="term"><code class="function">PQconninfoParse</code><a id="id-1.7.3.8.2.12.8.1.2" class="indexterm"></a></span></dt><dd><p>
Returns parsed connection options from the provided connection string.
</p><pre class="synopsis">
PQconninfoOption *PQconninfoParse(const char *conninfo, char **errmsg);
</pre><p>
</p><p>
Parses a connection string and returns the resulting options as an
array; or returns <code class="symbol">NULL</code> if there is a problem with the connection
string. This function can be used to extract
the <code class="function">PQconnectdb</code> options in the provided
connection string. The return value points to an array of
<code class="structname">PQconninfoOption</code> structures, which ends
with an entry having a null <code class="structfield">keyword</code> pointer.
</p><p>
All legal options will be present in the result array, but the
<code class="literal">PQconninfoOption</code> for any option not present
in the connection string will have <code class="literal">val</code> set to
<code class="literal">NULL</code>; default values are not inserted.
</p><p>
If <code class="literal">errmsg</code> is not <code class="symbol">NULL</code>, then <code class="literal">*errmsg</code> is set
to <code class="symbol">NULL</code> on success, else to a <code class="function">malloc</code>'d error string explaining
the problem. (It is also possible for <code class="literal">*errmsg</code> to be
set to <code class="symbol">NULL</code> and the function to return <code class="symbol">NULL</code>;
this indicates an out-of-memory condition.)
</p><p>
After processing the options array, free it by passing it to
<code class="function">PQconninfoFree</code>. If this is not done, some memory
is leaked for each call to <code class="function">PQconninfoParse</code>.
Conversely, if an error occurs and <code class="literal">errmsg</code> is not <code class="symbol">NULL</code>,
be sure to free the error string using <code class="function">PQfreemem</code>.
</p></dd><dt id="LIBPQ-PQFINISH"><span class="term"><code class="function">PQfinish</code><a id="id-1.7.3.8.2.12.9.1.2" class="indexterm"></a></span></dt><dd><p>
Closes the connection to the server. Also frees
memory used by the <code class="structname">PGconn</code> object.
</p><pre class="synopsis">
void PQfinish(PGconn *conn);
</pre><p>
</p><p>
Note that even if the server connection attempt fails (as
indicated by <code class="function">PQstatus</code>), the application should call <code class="function">PQfinish</code>
to free the memory used by the <code class="structname">PGconn</code> object.
The <code class="structname">PGconn</code> pointer must not be used again after
<code class="function">PQfinish</code> has been called.
</p></dd><dt id="LIBPQ-PQRESET"><span class="term"><code class="function">PQreset</code><a id="id-1.7.3.8.2.12.10.1.2" class="indexterm"></a></span></dt><dd><p>
Resets the communication channel to the server.
</p><pre class="synopsis">
void PQreset(PGconn *conn);
</pre><p>
</p><p>
This function will close the connection
to the server and attempt to reestablish a new
connection to the same server, using all the same
parameters previously used. This might be useful for
error recovery if a working connection is lost.
</p></dd><dt id="LIBPQ-PQRESETSTART"><span class="term"><code class="function">PQresetStart</code><a id="id-1.7.3.8.2.12.11.1.2" class="indexterm"></a><br /></span><span class="term"><code class="function">PQresetPoll</code><a id="id-1.7.3.8.2.12.11.2.2" class="indexterm"></a></span></dt><dd><p>
Reset the communication channel to the server, in a nonblocking manner.
</p><pre class="synopsis">
int PQresetStart(PGconn *conn);
PostgresPollingStatusType PQresetPoll(PGconn *conn);
</pre><p>
</p><p>
These functions will close the connection to the server and attempt to
reestablish a new connection to the same server, using all the same
parameters previously used. This can be useful for error recovery if a
working connection is lost. They differ from <code class="function">PQreset</code> (above) in that they
act in a nonblocking manner. These functions suffer from the same
restrictions as <code class="function">PQconnectStartParams</code>, <code class="function">PQconnectStart</code>
and <code class="function">PQconnectPoll</code>.
</p><p>
To initiate a connection reset, call
<code class="function">PQresetStart</code>. If it returns 0, the reset has
failed. If it returns 1, poll the reset using
<code class="function">PQresetPoll</code> in exactly the same way as you
would create the connection using <code class="function">PQconnectPoll</code>.
</p></dd><dt id="LIBPQ-PQPINGPARAMS"><span class="term"><code class="function">PQpingParams</code><a id="id-1.7.3.8.2.12.12.1.2" class="indexterm"></a></span></dt><dd><p>
<code class="function">PQpingParams</code> reports the status of the
server. It accepts connection parameters identical to those of
<code class="function">PQconnectdbParams</code>, described above. It is not
necessary to supply correct user name, password, or database name
values to obtain the server status; however, if incorrect values
are provided, the server will log a failed connection attempt.
</p><pre class="synopsis">
PGPing PQpingParams(const char * const *keywords,
const char * const *values,
int expand_dbname);
</pre><p>
The function returns one of the following values:
</p><div class="variablelist"><dl class="variablelist"><dt id="LIBPQ-PQPINGPARAMS-PQPING-OK"><span class="term"><code class="literal">PQPING_OK</code></span></dt><dd><p>
The server is running and appears to be accepting connections.
</p></dd><dt id="LIBPQ-PQPINGPARAMS-PQPING-REJECT"><span class="term"><code class="literal">PQPING_REJECT</code></span></dt><dd><p>
The server is running but is in a state that disallows connections
(startup, shutdown, or crash recovery).
</p></dd><dt id="LIBPQ-PQPINGPARAMS-PQPING-NO-RESPONSE"><span class="term"><code class="literal">PQPING_NO_RESPONSE</code></span></dt><dd><p>
The server could not be contacted. This might indicate that the
server is not running, or that there is something wrong with the
given connection parameters (for example, wrong port number), or
that there is a network connectivity problem (for example, a
firewall blocking the connection request).
</p></dd><dt id="LIBPQ-PQPINGPARAMS-PQPING-NO-ATTEMPT"><span class="term"><code class="literal">PQPING_NO_ATTEMPT</code></span></dt><dd><p>
No attempt was made to contact the server, because the supplied
parameters were obviously incorrect or there was some client-side
problem (for example, out of memory).
</p></dd></dl></div><p>
</p></dd><dt id="LIBPQ-PQPING"><span class="term"><code class="function">PQping</code><a id="id-1.7.3.8.2.12.13.1.2" class="indexterm"></a></span></dt><dd><p>
<code class="function">PQping</code> reports the status of the
server. It accepts connection parameters identical to those of
<code class="function">PQconnectdb</code>, described above. It is not
necessary to supply correct user name, password, or database name
values to obtain the server status; however, if incorrect values
are provided, the server will log a failed connection attempt.
</p><pre class="synopsis">
PGPing PQping(const char *conninfo);
</pre><p>
</p><p>
The return values are the same as for <code class="function">PQpingParams</code>.
</p></dd></dl></div><p>
</p><div class="sect2" id="LIBPQ-CONNSTRING"><div class="titlepage"><div><div><h3 class="title">34.1.1. Connection Strings</h3></div></div></div><a id="id-1.7.3.8.3.2" class="indexterm"></a><a id="id-1.7.3.8.3.3" class="indexterm"></a><p>
Several <span class="application">libpq</span> functions parse a user-specified string to obtain
connection parameters. There are two accepted formats for these strings:
plain <code class="literal">keyword = value</code> strings
and URIs. URIs generally follow
<a class="ulink" href="https://tools.ietf.org/html/rfc3986" target="_top">RFC
3986</a>, except that multi-host connection strings are allowed
as further described below.
</p><div class="sect3" id="id-1.7.3.8.3.5"><div class="titlepage"><div><div><h4 class="title">34.1.1.1. Keyword/Value Connection Strings</h4></div></div></div><p>
In the first format, each parameter setting is in the form
<code class="literal">keyword = value</code>. Spaces around the equal sign are
optional. To write an empty value, or a value containing spaces, surround it
with single quotes, e.g., <code class="literal">keyword = 'a value'</code>. Single
quotes and backslashes within
the value must be escaped with a backslash, i.e., <code class="literal">\'</code> and
<code class="literal">\\</code>.
</p><p>
Example:
</p><pre class="programlisting">
host=localhost port=5432 dbname=mydb connect_timeout=10
</pre><p>
</p><p>
The recognized parameter key words are listed in <a class="xref" href="libpq-connect.html#LIBPQ-PARAMKEYWORDS" title="34.1.2. Parameter Key Words">Section 34.1.2</a>.
</p></div><div class="sect3" id="id-1.7.3.8.3.6"><div class="titlepage"><div><div><h4 class="title">34.1.1.2. Connection URIs</h4></div></div></div><p>
The general form for a connection <acronym class="acronym">URI</acronym> is:
</p><pre class="synopsis">
postgresql://[user[:password]@][netloc][:port][,...][/dbname][?param1=value1&...]
</pre><p>
</p><p>
The <acronym class="acronym">URI</acronym> scheme designator can be either
<code class="literal">postgresql://</code> or <code class="literal">postgres://</code>. Each
of the <acronym class="acronym">URI</acronym> parts is optional. The following examples
illustrate valid <acronym class="acronym">URI</acronym> syntax uses:
</p><pre class="programlisting">
postgresql://
postgresql://localhost
postgresql://localhost:5433
postgresql://localhost/mydb
postgresql://user@localhost
postgresql://user:secret@localhost
postgresql://other@localhost/otherdb?connect_timeout=10&application_name=myapp
postgresql://host1:123,host2:456/somedb?target_session_attrs=any&application_name=myapp
</pre><p>
Components of the hierarchical part of the <acronym class="acronym">URI</acronym> can also
be given as parameters. For example:
</p><pre class="programlisting">
postgresql:///mydb?host=localhost&port=5433
</pre><p>
</p><p>
Percent-encoding may be used to include symbols with special meaning in any
of the <acronym class="acronym">URI</acronym> parts, e.g. replace <code class="literal">=</code> with
<code class="literal">%3D</code>.
</p><p>
Any connection parameters not corresponding to key words listed in <a class="xref" href="libpq-connect.html#LIBPQ-PARAMKEYWORDS" title="34.1.2. Parameter Key Words">Section 34.1.2</a> are ignored and a warning message about them
is sent to <code class="filename">stderr</code>.
</p><p>
For improved compatibility with JDBC connection <acronym class="acronym">URI</acronym>s,
instances of parameter <code class="literal">ssl=true</code> are translated into
<code class="literal">sslmode=require</code>.
</p><p>
The host part may be either host name or an IP address. To specify an
IPv6 host address, enclose it in square brackets:
</p><pre class="synopsis">
postgresql://[2001:db8::1234]/database
</pre><p>
</p><p>
The host component is interpreted as described for the parameter <a class="xref" href="libpq-connect.html#LIBPQ-CONNECT-HOST">host</a>. In particular, a Unix-domain socket
connection is chosen if the host part is either empty or starts with a
slash, otherwise a TCP/IP connection is initiated. Note, however, that the
slash is a reserved character in the hierarchical part of the URI. So, to
specify a non-standard Unix-domain socket directory, either omit the host
specification in the URI and specify the host as a parameter, or
percent-encode the path in the host component of the URI:
</p><pre class="programlisting">
postgresql:///dbname?host=/var/lib/postgresql
postgresql://%2Fvar%2Flib%2Fpostgresql/dbname
</pre><p>
</p><p>
It is possible to specify multiple host components, each with an optional
port component, in a single URI. A URI of the form
<code class="literal">postgresql://host1:port1,host2:port2,host3:port3/</code>
is equivalent to a connection string of the form
<code class="literal">host=host1,host2,host3 port=port1,port2,port3</code>. Each
host will be tried in turn until a connection is successfully established.
</p></div><div class="sect3" id="LIBPQ-MULTIPLE-HOSTS"><div class="titlepage"><div><div><h4 class="title">34.1.1.3. Specifying Multiple Hosts</h4></div></div></div><p>
It is possible to specify multiple hosts to connect to, so that they are
tried in the given order. In the Keyword/Value format, the <code class="literal">host</code>,
<code class="literal">hostaddr</code>, and <code class="literal">port</code> options accept a comma-separated
list of values. The same number of elements must be given in each
option that is specified, such
that e.g. the first <code class="literal">hostaddr</code> corresponds to the first host name,
the second <code class="literal">hostaddr</code> corresponds to the second host name, and so
forth. As an exception, if only one <code class="literal">port</code> is specified, it
applies to all the hosts.
</p><p>
In the connection URI format, you can list multiple <code class="literal">host:port</code> pairs
separated by commas, in the <code class="literal">host</code> component of the URI.
</p><p>
In either format, a single host name can translate to multiple network
addresses. A common example of this is a host that has both an IPv4 and
an IPv6 address.
</p><p>
When multiple hosts are specified, or when a single host name is
translated to multiple addresses, all the hosts and addresses will be
tried in order, until one succeeds. If none of the hosts can be reached,
the connection fails. If a connection is established successfully, but
authentication fails, the remaining hosts in the list are not tried.
</p><p>
If a password file is used, you can have different passwords for
different hosts. All the other connection options are the same for every
host in the list; it is not possible to e.g. specify different
usernames for different hosts.
</p></div></div><div class="sect2" id="LIBPQ-PARAMKEYWORDS"><div class="titlepage"><div><div><h3 class="title">34.1.2. Parameter Key Words</h3></div></div></div><p>
The currently recognized parameter key words are:
</p><div class="variablelist"><dl class="variablelist"><dt id="LIBPQ-CONNECT-HOST"><span class="term"><code class="literal">host</code></span></dt><dd><p>
Name of host to connect to.<a id="id-1.7.3.8.4.2.1.1.2.1.1" class="indexterm"></a>
If a host name begins with a slash, it specifies Unix-domain
communication rather than TCP/IP communication; the value is the
name of the directory in which the socket file is stored.
The default behavior when <code class="literal">host</code> is
not specified, or is empty, is to connect to a Unix-domain
socket<a id="id-1.7.3.8.4.2.1.1.2.1.3" class="indexterm"></a> in
<code class="filename">/tmp</code> (or whatever socket directory was specified
when <span class="productname">PostgreSQL</span> was built). On machines without
Unix-domain sockets, the default is to connect to <code class="literal">localhost</code>.
</p><p>
A comma-separated list of host names is also accepted, in which case
each host name in the list is tried in order; an empty item in the
list selects the default behavior as explained above. See
<a class="xref" href="libpq-connect.html#LIBPQ-MULTIPLE-HOSTS" title="34.1.1.3. Specifying Multiple Hosts">Section 34.1.1.3</a> for details.
</p></dd><dt id="LIBPQ-CONNECT-HOSTADDR"><span class="term"><code class="literal">hostaddr</code></span></dt><dd><p>
Numeric IP address of host to connect to. This should be in the
standard IPv4 address format, e.g., <code class="literal">172.28.40.9</code>. If
your machine supports IPv6, you can also use those addresses.
TCP/IP communication is
always used when a nonempty string is specified for this parameter.
</p><p>
Using <code class="literal">hostaddr</code> instead of <code class="literal">host</code> allows the
application to avoid a host name look-up, which might be important
in applications with time constraints. However, a host name is
required for GSSAPI or SSPI authentication
methods, as well as for <code class="literal">verify-full</code> SSL
certificate verification. The following rules are used:
</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p>
If <code class="literal">host</code> is specified
without <code class="literal">hostaddr</code>, a host name lookup occurs.
(When using <code class="function">PQconnectPoll</code>, the lookup occurs
when <code class="function">PQconnectPoll</code> first considers this host
name, and it may cause <code class="function">PQconnectPoll</code> to block
for a significant amount of time.)
</p></li><li class="listitem"><p>
If <code class="literal">hostaddr</code> is specified without <code class="literal">host</code>,
the value for <code class="literal">hostaddr</code> gives the server network address.
The connection attempt will fail if the authentication
method requires a host name.
</p></li><li class="listitem"><p>
If both <code class="literal">host</code> and <code class="literal">hostaddr</code> are specified,
the value for <code class="literal">hostaddr</code> gives the server network address.
The value for <code class="literal">host</code> is ignored unless the
authentication method requires it, in which case it will be
used as the host name.
</p></li></ul></div><p>
Note that authentication is likely to fail if <code class="literal">host</code>
is not the name of the server at network address <code class="literal">hostaddr</code>.
Also, when both <code class="literal">host</code> and <code class="literal">hostaddr</code>
are specified, <code class="literal">host</code>
is used to identify the connection in a password file (see
<a class="xref" href="libpq-pgpass.html" title="34.15. The Password File">Section 34.15</a>).
</p><p>
A comma-separated list of <code class="literal">hostaddr</code> values is also
accepted, in which case each host in the list is tried in order.
An empty item in the list causes the corresponding host name to be
used, or the default host name if that is empty as well. See
<a class="xref" href="libpq-connect.html#LIBPQ-MULTIPLE-HOSTS" title="34.1.1.3. Specifying Multiple Hosts">Section 34.1.1.3</a> for details.
</p><p>
Without either a host name or host address,
<span class="application">libpq</span> will connect using a
local Unix-domain socket; or on machines without Unix-domain
sockets, it will attempt to connect to <code class="literal">localhost</code>.
</p></dd><dt id="LIBPQ-CONNECT-PORT"><span class="term"><code class="literal">port</code></span></dt><dd><p>
Port number to connect to at the server host, or socket file
name extension for Unix-domain
connections.<a id="id-1.7.3.8.4.2.1.3.2.1.1" class="indexterm"></a>
If multiple hosts were given in the <code class="literal">host</code> or
<code class="literal">hostaddr</code> parameters, this parameter may specify a
comma-separated list of ports of the same length as the host list, or
it may specify a single port number to be used for all hosts.
An empty string, or an empty item in a comma-separated list,
specifies the default port number established
when <span class="productname">PostgreSQL</span> was built.
</p></dd><dt id="LIBPQ-CONNECT-DBNAME"><span class="term"><code class="literal">dbname</code></span></dt><dd><p>
The database name. Defaults to be the same as the user name.
In certain contexts, the value is checked for extended
formats; see <a class="xref" href="libpq-connect.html#LIBPQ-CONNSTRING" title="34.1.1. Connection Strings">Section 34.1.1</a> for more details on
those.
</p></dd><dt id="LIBPQ-CONNECT-USER"><span class="term"><code class="literal">user</code></span></dt><dd><p>
<span class="productname">PostgreSQL</span> user name to connect as.
Defaults to be the same as the operating system name of the user
running the application.
</p></dd><dt id="LIBPQ-CONNECT-PASSWORD"><span class="term"><code class="literal">password</code></span></dt><dd><p>
Password to be used if the server demands password authentication.
</p></dd><dt id="LIBPQ-CONNECT-PASSFILE"><span class="term"><code class="literal">passfile</code></span></dt><dd><p>
Specifies the name of the file used to store passwords
(see <a class="xref" href="libpq-pgpass.html" title="34.15. The Password File">Section 34.15</a>).
Defaults to <code class="filename">~/.pgpass</code>, or
<code class="filename">%APPDATA%\postgresql\pgpass.conf</code> on Microsoft Windows.
(No error is reported if this file does not exist.)
</p></dd><dt id="LIBPQ-CONNECT-CONNECT-TIMEOUT"><span class="term"><code class="literal">connect_timeout</code></span></dt><dd><p>
Maximum wait for connection, in seconds (write as a decimal integer,
e.g. <code class="literal">10</code>). Zero, negative, or not specified means
wait indefinitely. The minimum allowed timeout is 2 seconds, therefore
a value of <code class="literal">1</code> is interpreted as <code class="literal">2</code>.
This timeout applies separately to each host name or IP address.
For example, if you specify two hosts and <code class="literal">connect_timeout</code>
is 5, each host will time out if no connection is made within 5
seconds, so the total time spent waiting for a connection might be
up to 10 seconds.
</p></dd><dt id="LIBPQ-CONNECT-CLIENT-ENCODING"><span class="term"><code class="literal">client_encoding</code></span></dt><dd><p>
This sets the <code class="varname">client_encoding</code>
configuration parameter for this connection. In addition to
the values accepted by the corresponding server option, you
can use <code class="literal">auto</code> to determine the right
encoding from the current locale in the client
(<code class="envar">LC_CTYPE</code> environment variable on Unix
systems).
</p></dd><dt id="LIBPQ-CONNECT-OPTIONS"><span class="term"><code class="literal">options</code></span></dt><dd><p>
Specifies command-line options to send to the server at connection
start. For example, setting this to <code class="literal">-c geqo=off</code> sets the
session's value of the <code class="varname">geqo</code> parameter to
<code class="literal">off</code>. Spaces within this string are considered to
separate command-line arguments, unless escaped with a backslash
(<code class="literal">\</code>); write <code class="literal">\\</code> to represent a literal
backslash. For a detailed discussion of the available
options, consult <a class="xref" href="runtime-config.html" title="Chapter 19. Server Configuration">Chapter 19</a>.
</p></dd><dt id="LIBPQ-CONNECT-APPLICATION-NAME"><span class="term"><code class="literal">application_name</code></span></dt><dd><p>
Specifies a value for the <a class="xref" href="runtime-config-logging.html#GUC-APPLICATION-NAME">application_name</a>
configuration parameter.
</p></dd><dt id="LIBPQ-CONNECT-FALLBACK-APPLICATION-NAME"><span class="term"><code class="literal">fallback_application_name</code></span></dt><dd><p>
Specifies a fallback value for the <a class="xref" href="runtime-config-logging.html#GUC-APPLICATION-NAME">application_name</a> configuration parameter.
This value will be used if no value has been given for
<code class="literal">application_name</code> via a connection parameter or the
<code class="envar">PGAPPNAME</code> environment variable. Specifying
a fallback name is useful in generic utility programs that
wish to set a default application name but allow it to be
overridden by the user.
</p></dd><dt id="LIBPQ-KEEPALIVES"><span class="term"><code class="literal">keepalives</code></span></dt><dd><p>
Controls whether client-side TCP keepalives are used. The default
value is 1, meaning on, but you can change this to 0, meaning off,
if keepalives are not wanted. This parameter is ignored for
connections made via a Unix-domain socket.
</p></dd><dt id="LIBPQ-KEEPALIVES-IDLE"><span class="term"><code class="literal">keepalives_idle</code></span></dt><dd><p>
Controls the number of seconds of inactivity after which TCP should
send a keepalive message to the server. A value of zero uses the
system default. This parameter is ignored for connections made via a
Unix-domain socket, or if keepalives are disabled.
It is only supported on systems where <code class="symbol">TCP_KEEPIDLE</code> or
an equivalent socket option is available, and on Windows; on other
systems, it has no effect.
</p></dd><dt id="LIBPQ-KEEPALIVES-INTERVAL"><span class="term"><code class="literal">keepalives_interval</code></span></dt><dd><p>
Controls the number of seconds after which a TCP keepalive message
that is not acknowledged by the server should be retransmitted. A
value of zero uses the system default. This parameter is ignored for
connections made via a Unix-domain socket, or if keepalives are disabled.
It is only supported on systems where <code class="symbol">TCP_KEEPINTVL</code> or
an equivalent socket option is available, and on Windows; on other
systems, it has no effect.
</p></dd><dt id="LIBPQ-KEEPALIVES-COUNT"><span class="term"><code class="literal">keepalives_count</code></span></dt><dd><p>
Controls the number of TCP keepalives that can be lost before the
client's connection to the server is considered dead. A value of
zero uses the system default. This parameter is ignored for
connections made via a Unix-domain socket, or if keepalives are disabled.
It is only supported on systems where <code class="symbol">TCP_KEEPCNT</code> or
an equivalent socket option is available; on other systems, it has no
effect.
</p></dd><dt id="LIBPQ-CONNECT-TTY"><span class="term"><code class="literal">tty</code></span></dt><dd><p>
Ignored (formerly, this specified where to send server debug output).
</p></dd><dt id="LIBPQ-CONNECT-REPLICATION"><span class="term"><code class="literal">replication</code></span></dt><dd><p>
This option determines whether the connection should use the
replication protocol instead of the normal protocol. This is what
PostgreSQL replication connections as well as tools such as
<span class="application">pg_basebackup</span> use internally, but it can
also be used by third-party applications. For a description of the
replication protocol, consult <a class="xref" href="protocol-replication.html" title="53.4. Streaming Replication Protocol">Section 53.4</a>.
</p><p>
The following values, which are case-insensitive, are supported:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term">
<code class="literal">true</code>, <code class="literal">on</code>,
<code class="literal">yes</code>, <code class="literal">1</code>
</span></dt><dd><p>
The connection goes into physical replication mode.
</p></dd><dt><span class="term"><code class="literal">database</code></span></dt><dd><p>
The connection goes into logical replication mode, connecting to
the database specified in the <code class="literal">dbname</code> parameter.
</p></dd><dt><span class="term">
<code class="literal">false</code>, <code class="literal">off</code>,
<code class="literal">no</code>, <code class="literal">0</code>
</span></dt><dd><p>
The connection is a regular one, which is the default behavior.
</p></dd></dl></div><p>
</p><p>
In physical or logical replication mode, only the simple query protocol
can be used.
</p></dd><dt id="LIBPQ-CONNECT-SSLMODE"><span class="term"><code class="literal">sslmode</code></span></dt><dd><p>
This option determines whether or with what priority a secure
<acronym class="acronym">SSL</acronym> TCP/IP connection will be negotiated with the
server. There are six modes:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">disable</code></span></dt><dd><p>
only try a non-<acronym class="acronym">SSL</acronym> connection
</p></dd><dt><span class="term"><code class="literal">allow</code></span></dt><dd><p>
first try a non-<acronym class="acronym">SSL</acronym> connection; if that
fails, try an <acronym class="acronym">SSL</acronym> connection
</p></dd><dt><span class="term"><code class="literal">prefer</code> (default)</span></dt><dd><p>
first try an <acronym class="acronym">SSL</acronym> connection; if that fails,
try a non-<acronym class="acronym">SSL</acronym> connection
</p></dd><dt><span class="term"><code class="literal">require</code></span></dt><dd><p>
only try an <acronym class="acronym">SSL</acronym> connection. If a root CA
file is present, verify the certificate in the same way as
if <code class="literal">verify-ca</code> was specified
</p></dd><dt><span class="term"><code class="literal">verify-ca</code></span></dt><dd><p>
only try an <acronym class="acronym">SSL</acronym> connection, and verify that
the server certificate is issued by a trusted
certificate authority (<acronym class="acronym">CA</acronym>)
</p></dd><dt><span class="term"><code class="literal">verify-full</code></span></dt><dd><p>
only try an <acronym class="acronym">SSL</acronym> connection, verify that the
server certificate is issued by a
trusted <acronym class="acronym">CA</acronym> and that the requested server host name
matches that in the certificate
</p></dd></dl></div><p>
See <a class="xref" href="libpq-ssl.html" title="34.18. SSL Support">Section 34.18</a> for a detailed description of how
these options work.
</p><p>
<code class="literal">sslmode</code> is ignored for Unix domain socket
communication.
If <span class="productname">PostgreSQL</span> is compiled without SSL support,
using options <code class="literal">require</code>, <code class="literal">verify-ca</code>, or
<code class="literal">verify-full</code> will cause an error, while
options <code class="literal">allow</code> and <code class="literal">prefer</code> will be
accepted but <span class="application">libpq</span> will not actually attempt
an <acronym class="acronym">SSL</acronym>
connection.<a id="id-1.7.3.8.4.2.1.19.2.2.10" class="indexterm"></a>
</p></dd><dt id="LIBPQ-CONNECT-REQUIRESSL"><span class="term"><code class="literal">requiressl</code></span></dt><dd><p>
This option is deprecated in favor of the <code class="literal">sslmode</code>
setting.
</p><p>
If set to 1, an <acronym class="acronym">SSL</acronym> connection to the server
is required (this is equivalent to <code class="literal">sslmode</code>
<code class="literal">require</code>). <span class="application">libpq</span> will then refuse
to connect if the server does not accept an
<acronym class="acronym">SSL</acronym> connection. If set to 0 (default),
<span class="application">libpq</span> will negotiate the connection type with
the server (equivalent to <code class="literal">sslmode</code>
<code class="literal">prefer</code>). This option is only available if
<span class="productname">PostgreSQL</span> is compiled with SSL support.
</p></dd><dt id="LIBPQ-CONNECT-SSLCOMPRESSION"><span class="term"><code class="literal">sslcompression</code></span></dt><dd><p>
If set to 1, data sent over SSL connections will be compressed. If
set to 0, compression will be disabled. The default is 0. This
parameter is ignored if a connection without SSL is made.
</p><p>
SSL compression is nowadays considered insecure and its use is no
longer recommended. <span class="productname">OpenSSL</span> 1.1.0 disables
compression by default, and many operating system distributions
disable it in prior versions as well, so setting this parameter to on
will not have any effect if the server does not accept compression.
On the other hand, <span class="productname">OpenSSL</span> before 1.0.0
does not support disabling compression, so this parameter is ignored
with those versions, and whether compression is used depends on the
server.
</p><p>
If security is not a primary concern, compression can improve
throughput if the network is the bottleneck. Disabling compression
can improve response time and throughput if CPU performance is the
limiting factor.
</p></dd><dt id="LIBPQ-CONNECT-SSLCERT"><span class="term"><code class="literal">sslcert</code></span></dt><dd><p>
This parameter specifies the file name of the client SSL
certificate, replacing the default
<code class="filename">~/.postgresql/postgresql.crt</code>.
This parameter is ignored if an SSL connection is not made.
</p></dd><dt id="LIBPQ-CONNECT-SSLKEY"><span class="term"><code class="literal">sslkey</code></span></dt><dd><p>
This parameter specifies the location for the secret key used for
the client certificate. It can either specify a file name that will
be used instead of the default
<code class="filename">~/.postgresql/postgresql.key</code>, or it can specify a key
obtained from an external <span class="quote">“<span class="quote">engine</span>”</span> (engines are
<span class="productname">OpenSSL</span> loadable modules). An external engine
specification should consist of a colon-separated engine name and
an engine-specific key identifier. This parameter is ignored if an
SSL connection is not made.
</p></dd><dt id="LIBPQ-CONNECT-SSLROOTCERT"><span class="term"><code class="literal">sslrootcert</code></span></dt><dd><p>
This parameter specifies the name of a file containing SSL
certificate authority (<acronym class="acronym">CA</acronym>) certificate(s).
If the file exists, the server's certificate will be verified
to be signed by one of these authorities. The default is
<code class="filename">~/.postgresql/root.crt</code>.
</p></dd><dt id="LIBPQ-CONNECT-SSLCRL"><span class="term"><code class="literal">sslcrl</code></span></dt><dd><p>
This parameter specifies the file name of the SSL certificate
revocation list (CRL). Certificates listed in this file, if it
exists, will be rejected while attempting to authenticate the
server's certificate. The default is
<code class="filename">~/.postgresql/root.crl</code>.
</p></dd><dt id="LIBPQ-CONNECT-REQUIREPEER"><span class="term"><code class="literal">requirepeer</code></span></dt><dd><p>
This parameter specifies the operating-system user name of the
server, for example <code class="literal">requirepeer=postgres</code>.
When making a Unix-domain socket connection, if this
parameter is set, the client checks at the beginning of the
connection that the server process is running under the specified
user name; if it is not, the connection is aborted with an error.
This parameter can be used to provide server authentication similar
to that available with SSL certificates on TCP/IP connections.
(Note that if the Unix-domain socket is in
<code class="filename">/tmp</code> or another publicly writable location,
any user could start a server listening there. Use this parameter
to ensure that you are connected to a server run by a trusted user.)
This option is only supported on platforms for which the
<code class="literal">peer</code> authentication method is implemented; see
<a class="xref" href="auth-peer.html" title="20.9. Peer Authentication">Section 20.9</a>.
</p></dd><dt id="LIBPQ-CONNECT-KRBSRVNAME"><span class="term"><code class="literal">krbsrvname</code></span></dt><dd><p>
Kerberos service name to use when authenticating with GSSAPI.
This must match the service name specified in the server
configuration for Kerberos authentication to succeed. (See also
<a class="xref" href="gssapi-auth.html" title="20.6. GSSAPI Authentication">Section 20.6</a>.)
</p></dd><dt id="LIBPQ-CONNECT-GSSLIB"><span class="term"><code class="literal">gsslib</code></span></dt><dd><p>
GSS library to use for GSSAPI authentication. Only used on Windows.
Set to <code class="literal">gssapi</code> to force libpq to use the GSSAPI
library for authentication instead of the default SSPI.
</p></dd><dt id="LIBPQ-CONNECT-SERVICE"><span class="term"><code class="literal">service</code></span></dt><dd><p>
Service name to use for additional parameters. It specifies a service
name in <code class="filename">pg_service.conf</code> that holds additional connection parameters.
This allows applications to specify only a service name so connection parameters
can be centrally maintained. See <a class="xref" href="libpq-pgservice.html" title="34.16. The Connection Service File">Section 34.16</a>.
</p></dd><dt id="LIBPQ-CONNECT-TARGET-SESSION-ATTRS"><span class="term"><code class="literal">target_session_attrs</code></span></dt><dd><p>
If this parameter is set to <code class="literal">read-write</code>, only a
connection in which read-write transactions are accepted by default
is considered acceptable. The query
<code class="literal">SHOW transaction_read_only</code> will be sent upon any
successful connection; if it returns <code class="literal">on</code>, the connection
will be closed. If multiple hosts were specified in the connection
string, any remaining servers will be tried just as if the connection
attempt had failed. The default value of this parameter,
<code class="literal">any</code>, regards all connections as acceptable.
</p></dd></dl></div><p>
</p></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="libpq.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="libpq.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="libpq-status.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">Chapter 34. <span class="application">libpq</span> - C Library </td><td width="20%" align="center"><a accesskey="h" href="index.html">Home</a></td><td width="40%" align="right" valign="top"> 34.2. Connection Status Functions</td></tr></table></div></body></html>
