pgbouncer
pgbouncer — aPostgres Pro connection pooler
Synopsis
On Linux systems:
pgbouncer [ -d ] [ -R ] [ -v ] [ -uuser ] pgbouncer.ini
pgbouncer -V | -h
On Windows:
pgbouncer [ -v ] [ -uuser ] pgbouncer.ini
pgbouncer -V | -h
To usepgbouncer as a Windows service:
pgbouncer.exe --regservice pgbouncer.ini
pgbouncer.exe --unregservice pgbouncer.ini
Description
pgbouncer is aPostgres Pro connection pooler. Any target application can be connected topgbouncer as if it were aPostgres Pro server, andpgbouncer will create a connection to the actual server, or it will reuse one of its existing connections.
The aim ofpgbouncer is to lower the performance impact of opening new connections toPostgres Pro.
In order not to compromise transaction semantics for connection pooling,pgbouncer supports several types of pooling when rotating connections:
- Session pooling
Most polite method. When a client connects, a server connection will be assigned to it for the whole duration the client stays connected. When the client disconnects, the server connection will be put back into the pool. This is the default method.
- Transaction pooling
A server connection is assigned to a client only during a transaction. Whenpgbouncer notices that transaction is over, the server connection will be put back into the pool.
- Statement pooling
Most aggressive method. The server connection will be put back into the pool immediately after a query completes. Multi-statement transactions are disallowed in this mode as they would break.
The administration interface ofpgbouncer consists of some newSHOW commands available when connected to a special“virtual” databasepgbouncer.
Quick Start
Basic setup and usage is as follows.
Create a
pgbouncer.inifile. Details in thepgbouncer(5)man page. Simple example:[databases]template1 = host=localhost dbname=template1 auth_user=someuser[pgbouncer]listen_port = 6432listen_addr = localhostauth_type = md5auth_file = userlist.txtlogfile = pgbouncer.logpidfile = pgbouncer.pidadmin_users = someuser
Create a
userlist.txtfile that contains the users allowed in:"someuser" "same_password_as_in_server"
Launchpgbouncer:
$ pgbouncer -d pgbouncer.ini
Note
The above command does not work on Windows systems. Instead,pgbouncer must be launched as a service that first needs to be registered, as follows:
pgbouncer --regservice
Have your application (or the
psqlclient) connect topgbouncer instead of directly to thePostgres Pro server:$ psql -p 6432 -U someuser template1
Managepgbouncer by connecting to the special administration databasepgbouncer and issuing
SHOW HELP;to begin:$ psql -p 6432 -U someuser pgbouncerpgbouncer=# SHOW HELP;NOTICE: Console usageDETAIL: SHOW [HELP|CONFIG|DATABASES|FDS|POOLS|CLIENTS|SERVERS|SOCKETS|LISTS|VERSION|...] SET key = arg RELOAD PAUSE SUSPEND RESUME SHUTDOWN [...]
If you made changes to the
pgbouncer.inifile, you can reload it with:pgbouncer=# RELOAD;
Options
-d, --daemonRun in the background. Without it, the process will run in the foreground. In daemon mode, setting
pidfileas well aslogfileorsyslogis required. No log messages will be written tostderr after going into the background.Note
Does not work on Windows,pgbouncer needs to run as service there.
-R, --rebootNote
This option is deprecated. Instead of this option use a rolling restart with multiple pgbouncer processes listening on the same port using
so_reuseportinstead.Do an online restart. That means connecting to the running process, loading the open sockets from it, and then using them. If there is no active process, boot normally.
-uuser, --useruserSwitch to the given user on startup.
-v, --verboseIncrease verbosity. Can be used multiple times.
-q, --quietBe quiet: do not log tostderr. This does not affect logging verbosity, only thatstderr is not to be used. For use in
init.dscripts.-V, --versionShow version.
-h, --helpShow short help.
--regserviceWin32: Register to run as Windows service. The
service_nameconfiguration parameter value is used as the name to register under.--unregserviceWin32: Unregister Windows service.
Admin Console
The console is available by connecting as normal to the databasepgbouncer:
$ psql -p 6432 pgbouncer
Only users listed in the configuration parametersadmin_users orstats_users are allowed to log in to the console. (Except whenauth_mode=any, then any user is allowed in as astats_user.)
Additionally, the user name The admin console currently only supports the simple query protocol. Some drivers use the extended query protocol for all commands; these drivers will not work for this. The Shows statistics. In this and related commands, the total figures are since process start, the averages are updated every Statistics are presented per database. Total number ofSQL transactions pooled bypgbouncer. Total number ofSQL queries pooled bypgbouncer. Total volume in bytes of network traffic received bypgbouncer. Total volume in bytes of network traffic sent bypgbouncer. Total number of microseconds spent bypgbouncer when connected toPostgres Pro in a transaction, either idle in transaction or executing queries. Total number of microseconds spent bypgbouncer when actively connected toPostgres Pro, executing queries. Time spent by clients waiting for a server, in microseconds. Updated when a client connection is assigned a backend connection. Average transactions per second in last stat period. Average queries per second in last stat period. Average received (from clients) bytes per second. Average sent (to clients) bytes per second. Average transaction duration, in microseconds. Average query duration, in microseconds. Average time spent by clients waiting for a server that were assigned a backend connection within the current Subset of Subset of Like S, for server. User namepgbouncer uses to connect to server. Database name. State of thepgbouncer server connection, one of IP address ofPostgres Pro server. Port ofPostgres Pro server. Connection start address on local machine. Connection start port on local machine. When the connection was made. When last request was issued. Not used for server connections. Not used for server connections. 1 if the connection will be closed as soon as possible, because a configuration file reload or DNS update changed the connection information or Address of internal object for this connection. Used as unique ID. Address of client connection the server is paired with. PID of backend server process. In case connection is made over Unix socket and OS supports getting process ID info, its OSPID. Otherwise it's extracted from cancel packet the server sent, which should be thePID in case the server isPostgres Pro, but it's a random number in case the server is anotherpgbouncer. A string with TLS connection information, or empty if not using TLS. A string containing the C, for client. Client connected user. Database name. State of the client connection, one of IP address of the client. Source port of the client. Connection end address on local machine. Connection end port on local machine. Timestamp of connect time. Timestamp of latest client request. Current waiting time in seconds. Microsecond part of the current waiting time. Not used for clients. Address of internal object for this connection. Used as unique ID. Address of server connection the client is paired with. Process ID, in case client connects over Unix socket and OS supports getting it. A string with TLS connection information, or empty if not using TLS. A string containing the A new pool entry is made for each couple of (database, user). Database name. User name. Client connections that are either linked to server connections or are idle with no queries waiting to be processed. Client connections that have sent queries but have not yet got a server connection. Client connections that have forwarded query cancellations to the server and are waiting for the server response. Client connections that have not forwarded query cancellations to the server yet. Server connections that are linked to a client. Server connections that are currently forwarding a cancel request. Servers that normally could become idle but are waiting to do so until all in-flight cancel requests have completed that were sent to cancel a query on this server. Server connections that are unused and immediately usable for client queries. Server connections that have been idle for more than Server connections that are currently running either Server connections currently in the process of logging in. How long the first (oldest) client in the queue has waited, in seconds. If this starts increasing, then the current pool of servers does not handle requests quickly enough. The reason may be either an overloaded server or just too small of a Microsecond part of the maximum waiting time. The pooling mode in use. A new ID of the configured peer entry. Client connections that have forwarded query cancellations to the server and are waiting for the server response. Client connections that have not forwarded query cancellations to the server yet. Server connections that are currently forwarding a cancel request. Server connections currently in the process of logging in. Show following internal information, in columns (not rows): Count of databases. Count of users. Count of pools. Count of free clients. Count of used clients. Count of clients in Count of free servers. Count of used servers. Count of DNS names in the cache. Count of DNS zones in the cache. Count of in-flight DNS queries. Not used. The user name. The user's override Name of configured database entry. Hostpgbouncer connects to. Portpgbouncer connects to. Actual database namepgbouncer connects to. When the user is part of the connection string, the connection betweenpgbouncer andPostgres Pro is forced to the given user, whatever the client user. Maximum number of server connections. Minimum number of server connections. Maximum number of additional connections for this database. The database's override pool_mode, or Maximum number of allowed connections for this database, as set by Current number of connections for this database. 1 if this database is currently paused, else 0. 1 if this database is currently disabled, else 0. ID of the configured peer entry. Hostpgbouncer connects to. Portpgbouncer connects to. Maximum number of server connections that can be made to this peer. Internal command — shows list of file descriptors (FDs) in use with internal state attached to them. When the connected user has the user name This does not work on Windows. This command also blocks the internal event loop, so it should not be used whilepgbouncer is in use. File descriptor numeric value. One of User of the connection using the FD. Database of the connection using the FD. IP address of the connection using the FD, Port used by the connection using the FD. Cancel key for this connection. File descriptor for corresponding server/client. Shows low-level information about sockets or only active sockets. This includes the information shown under Show the current configuration settings, one per row, with the following columns: Configuration variable name. Configuration value. Configuration default value. Either Shows low-level information about the current sizes of various internal memory allocations. The information presented is subject to change. Show host names inDNS cache. Host name. How many seconds until next lookup. Comma separated list of addresses. Show thepgbouncer version string. Show thepgbouncer state settings. Current states are active, paused and suspended. pgbouncer tries to disconnect from all servers. Disconnecting each server connection waits for that server connection to be released according to the server pool's pooling mode (in transaction pooling mode, the transaction must complete, in statement mode, the statement must complete, and in session pooling mode the client must disconnect). The command will not return before all server connections have been disconnected. To be used at the time of database restart. If database name is given, only that database will be paused. New client connections to a paused database will wait until Reject all new client connections on the given database. Allow new client connections after a previous Close each open server connection for the given database, or all databases, after it is released (according to the pooling mode), even if its lifetime is not up yet. New server connections can be made immediately and will connect as necessary according to the pool size settings. This command is useful when the server connection setup has changed, for example to perform a gradual switchover to a new server. It is not necessary to run this command when the connection string in After this command is run, there could be an extended period where some server connections go to an old destination and some server connections go to a new destination. This is likely only sensible when switching read-only traffic between read-only replicas, or when switching between nodes of a multimaster replication setup. If all connections need to be switched at the same time, Immediately drop all client and server connections on given database. New client connections to a killed database will wait until All socket buffers are flushed andpgbouncer stops listening for data on them. The command will not return before all buffers are empty. To be used at the time ofpgbouncer online reboot. New client connections to a suspended database will wait until Resume work from previous Thepgbouncer process will exit. Thepgbouncer process will reload its configuration files and update changeable settings. This includes the main configuration file as well as the files specified by the settings pgbouncer notices when a configuration file reload changes the connection parameters of a database definition. An existing server connection to the old destination will be closed when the server connection is next released (according to the pooling mode), and new server connections will immediately use the updated connection parameters. Wait until all server connections, either of the specified database or of all databases, have cleared the Changes a configuration setting (see alsothe section called “SHOW CONFIG”). For example: (Note that this command is run on thepgbouncer admin console and setspgbouncer settings. A Reload config. Same as issuing the command Safe shutdown. Same as issuing Immediate shutdown. Same as issuing Same as issuing Same as issuing From thelibevent documentation: It is possible to disable support for By setting the environment variable The configuration file is in the Specifies the log file. For daemonization ( Note that setting Default: not set Specifies the PID file. Without Default: not set Specifies a list (comma-separated) of addresses where to listen for TCP connections. You may also use Addresses can be specified numerically (IPv4/IPv6) or by name. Default: not set Which port to listen on. Applies to both TCP and Unix sockets. Default: 6432 Specifies the location for Unix sockets. Applies to both the listening socket and server connections. If set to an empty string, Unix sockets are disabled. A value that starts with For online reboot ( Default: File system mode for Unix socket. Ignored for sockets in the abstract namespace. Not supported on Windows. Default: 0777 Group name to use for Unix socket. Ignored for sockets in the abstract namespace. Not supported on Windows. Default: not set If set, specifies the Unix user to change to after startup. Works only ifpgbouncer is started as root or if it's already running as the given user. Not supported on Windows. Default: not set Specifies when a server connection can be reused by other clients. Server is released back to pool after client disconnects. Default. Server is released back to pool after transaction finishes. Server is released back to pool after query finishes. Transactions spanning multiple statements are disallowed in this mode. Maximum number of client connections allowed. When this setting is increased, then the file descriptor limits in the operating system might also have to be increased. Note that the number of file descriptors potentially used is more than If a database user is specified in the connection string (all users connect under the same user name), the theoretical maximum is: The theoretical maximum should never be reached, unless somebody deliberately crafts a special load for it. Still, it means you should set the number of file descriptors to a safely high number. Search for Default: 100 How many server connections to allow per user/database pair. Can be overridden in the per-database configuration. Default: 20 Add more server connections to pool if below this number. Improves the behavior when the normal load suddenly comes back after a period of total inactivity. The value is effectively capped at the pool size. Default: 0 (disabled) How many additional connections to allow to a pool (see Default: 0 (disabled) If a client has not been serviced in this time,pgbouncer enables use of additional connections from the reserve pool. The 0 value disables this parameter. [seconds] Default: 5.0 Do not allow more than this many server connections per database (regardless of user). This considers thepgbouncer database that the client has connected to, not thePostgres Pro database of the outgoing connection. This can also be set per database in the Note that when you hit the limit, closing a client connection to one pool will not immediately allow a server connection to be established for another pool, because the server connection for the first pool is still open. Once the server connection closes (due to idle timeout), a new server connection will immediately be opened for the waiting pool. Default: 0 (unlimited) Do not allow more than this many server connections per user (regardless of database). This considers thepgbouncer user that is associated with a pool, which is either the user specified for the server connection or in absence of that the user the client has connected as. This can also be set per user in the Note that when you hit the limit, closing a client connection to one pool will not immediately allow a server connection to be established for another pool, because the server connection for the first pool is still open. Once the server connection closes (due to idle timeout), a new server connection will immediately be opened for the waiting pool. Default: 0 (unlimited) By default,pgbouncer reuses server connections in LIFO (last-in, first-out) manner, so that few connections get the most load. This gives best performance if you have a single server serving a database. But if there is a round-robin system behind a database address (TCP,DNS, or host list), then it is better ifpgbouncer also uses connections in that manner, thus achieving uniform load. Default: 0 By default,pgbouncer tracks If you need to specify multiple values, use a comma-separated list (e.g. Most parameters cannot be tracked this way. The only parameters that can be tracked are ones thatPostgres Pro reports to the client.Postgres Pro has an official list of parameters that it reports to the client.Postgres Pro extensions can change this list though, they can add parameters themselves that they also report, and they can start reporting already existing paremeters thatPostgres Pro does not report. Notably Citus 12.0+ causesPostgreSQL to also report The Default: By default,pgbouncer allows only parameters it can keep track of in startup packets: All other parameters will raise an error. To allow other parameters, they can be specified here, so thatpgbouncer knows that they are handled by the admin and it can ignore them. If you need to specify multiple values, use a comma-separated list (e.g. The Default: empty The peer ID used to identify thispgbouncer process in a group ofpgbouncer processes that are peered together. The Default: 0 Disable the Simple Query protocol (PQexec). Unlike the Extended Query protocol, Simple Query allows multiple queries in one packet, which allows some classes of SQL-injection attacks. Disabling it can improve security. Obviously, this means only clients that exclusively use the Extended Query protocol will stay working. Default: 0 Add the client host address and port to the application name setting set on connection start. This helps in identifying the source of bad queries, etc. This logic applies only at the start of a connection. If Default: 0 Show location of current configuration file. Changing it will makepgbouncer use another configuration file for next Default: file from command line Used on win32 service registration. Default: Alias for Sets how often the averages shown in various Default: 60 pgbouncer handles its own client authentication and has its own database of users. These settings control this. How to authenticate users. The client must connect overTLS connection with a valid client certificate. The user name is then taken from the Use MD5-based password check. This is the default authentication method. Use password check with SCRAM-SHA-256. The clear-text password is sent over the wire. Deprecated. No authentication is done. The user name must still exist in Like the The actual authentication type is loaded from Pluggable Authentication Modules (PAM) method is used to authenticate users, HBA configuration file to use when Default: not set The name of the file to load user names and passwords from. Seethe section called “Authentication File Format” for details. Most authentication types (see Default: not set If Direct access to Default: not set Query to load user's password from database. Direct access to Note that the query is run inside the target database. So if a function is used, it needs to be installed into each database. Default: Database name in the[databases] section to be used for authentication purposes. This option can be either global or overriden in the connection string if this parameter is specified. Togglessyslog on/off. On Windows, the event log is used instead. Default: 0 Under what name to send logs tosyslog. Default: Under what facility to send logs tosyslog. Possibilities: Default: Log successful logins. Default: 1 Log disconnections with reasons. Default: 1 Log error messages the pooler sends to clients. Default: 1 Write aggregated statistics into the log, every Default: 1 Increase verbosity. Mirrors the Default: 0 Comma-separated list of database users that are allowed to connect and run all commands on the console. Ignored when Default: empty Comma-separated list of database users that are allowed to connect and run read-only queries on the console. That means all Default: empty Query sent to server on connection release, before making it available to other clients. At that moment no transaction is in progress, so the value should not include The query is supposed to clean any changes made to the database session so that the next client gets the connection in a well-defined state. The default is When transaction pooling is used, the Default: Whether This setting is for working around broken setups that run applications that use session features over a transaction-pooledpgbouncer. It changes non-deterministic breakage to deterministic breakage: clients always lose their state after each transaction. Default: 0 How long to keep released connections available for immediate re-use, without running Default: 30.0 Simple do-nothing query to check if the server connection is alive. If an empty string, then sanity checking is disabled. Default: Disconnect a server in session pooling mode immediately or after the end of the current transaction if it is in If because of this setting a server connection is closed before the end of the client session, the client connection is also closed. This ensures that the client notices that the session has been interrupted. This setting makes connection configuration changes take effect sooner if session pooling and long-running sessions are used. The downside is that client sessions are liable to be interrupted by a configuration change, so client applications will need logic to reconnect and reestablish session state. But note that no transactions will be lost, because running transactions are not interrupted, only idle sessions. Default: 0 The pooler will close an unused (not currently linked to any client connection) server connection that has been connected longer than this. Setting it to 0 means the connection is to be used only once, then closed. [seconds] Default: 3600.0 If a server connection has been idle more than this many seconds it will be closed. If 0 then timeout is disabled. [seconds] Default: 600.0 If connection and login don't finish in this amount of time, the connection will be closed. [seconds] Default: 15.0 If login to the server failed, because of failure to connect or from authentication, the pooler waits this much before retrying to connect. During the waiting interval, new clients trying to connect to the failing server will get an error immediately without another connection attempt. [seconds] The purpose of this behavior is that clients don't unnecessarily queue up waiting for a server connection to become available if the server is not working. However, it also means that if a server is momentarily failing, for example during a restart or if the configuration was erroneous, then it will take at least this long until the pooler will consider connecting to it again. Planned events such as restarts should normally be managed using the Default: 15.0 If a client connects but does not manage to log in in this amount of time, it will be disconnected. Mainly needed to avoid dead connections stalling Default: 60.0 If the automatically created (via "*") database pools have been unused this many seconds, they are freed. The negative aspect of that is that their statistics are also forgotten. [seconds] Default: 3600.0 How longDNS lookups can be cached. The actualDNSTTL is ignored. [seconds] Default: 15.0 How longDNS errors and Default: 15.0 Period to check if a zone serial has changed. pgbouncer can collectDNS zones from host names (everything after first dot) and then periodically check if the zone serial changes. If it notices changes, all host names under that zone are looked up again. If any host IP changes, its connections are invalidated. Default: 0.0 (disabled) The location of a custom The parsing of the file is done by theDNS backend library, notpgbouncer, so see the library's documentation for details on allowed syntax and directives. Default: empty (use operating system defaults) TLS mode to use for connections from clients.TLS connections are disabled by default. When enabled, PlainTCP. If client requestsTLS, it's ignored. Default. If client requestsTLS, it is used. If not, plainTCP is used. If the client presents a client certificate, it is not validated. Same as The client must useTLS. If not, the client connection is rejected. If the client presents a client certificate, it is not validated. Client must useTLS with valid client certificate. Same as Private key forpgbouncer to accept client connections. Default: not set Certificate for private key. Clients can validate it. Default: not set Root certificate file to validate client certificates. Default: not set WhichTLS protocol versions are allowed. Allowed values: Default: AllowedTLS ciphers, in OpenSSL syntax. Shortcuts: Only connections usingTLS version 1.2 and lower are affected. There is currently no setting that controls the cipher choices used byTLS version 1.3 connections. Default: Elliptic Curve name to use for ECDH key exchanges. Allowed values: Default: DHE key exchange type. Allowed values: Default: TLS mode to use for connections toPostgres Pro servers. The default mode is PlainTCP.TLS is not even requested from the server. TLS connection is always requested first fromPostgres Pro. If refused, the connection will be established over plainTCP. Server certificate is not validated. Default. Connection must go overTLS. If server rejects it, plainTCP is not attempted. Server certificate is not validated. Connection must go overTLS and server certificate must be valid according to Connection must go overTLS and server certificate must be valid according to Root certificate file to validatePostgres Pro server certificates. Default: not set Private key forpgbouncer to authenticate againstPostgres Pro server. Default: not set Certificate for private key.Postgres Pro server can validate it. Default: not set WhichTLS protocol versions are allowed. Allowed values: Default: AllowedTLS ciphers, in OpenSSL syntax. Shortcuts: Only connections usingTLS version 1.2 and lower are affected. There is currently no setting that controls the cipher choices used byTLS version 1.3 connections. Default: Setting the following timeouts can cause unexpected errors. Queries running longer than that are canceled. This should be used only with a slightly smaller server-side Default: 0.0 (disabled) Maximum time queries are allowed to spend waiting for execution. If the query is not assigned to a server during that time, the client is disconnected. The 0 value disables this parameter. If this is disabled, clients will be queued indefinitely. [seconds] This setting is used to prevent unresponsive servers from grabbing up connections. It also helps when the server is down or rejects connections for any reason. Default: 120.0 Maximum time cancellation requests are allowed to spend waiting for execution. If the cancel request is not assigned to a server during that time, the client is disconnected. 0 disables. If this is disabled, cancel requests will be queued indefinitely. [seconds] This setting is used to prevent a client locking up when a cancel cannot be forwarded due to the server being down. Default: 10.0 Client connections idling longer than this many seconds are closed. This should be larger than the client-side connection lifetime settings, and only used for network problems. [seconds] Default: 0.0 (disabled) If a client has been in the“idle in transaction” state longer, it will be disconnected. [seconds] Default: 0.0 (disabled) How long to wait for buffer flush during Default: 10 Internal buffer size for packets. Affects size ofTCP packets sent and general memory usage. Actuallibpq packets can be larger than this, so no need to set it large. Default: 4096 Maximum size forPostgres Pro packets thatpgbouncer allows through. One packet is either one query or one result set row. The full result set can be larger. Default: 2147483647 The value of the Default: 128 How many times to process data on one connection, before proceeding. Without this limit, one connection with a big result set can stallpgbouncer for a long time. One loop processes one Default: 5 Specifies whether to set the socket option This setting has the desired effect on Linux. On systems that don't support the socket option at all, turning this setting on will result in an error. Eachpgbouncer instance on the same host needs different settings for at least To make sure query cancellations keep working, you should set uppgbouncer peering between the differentpgbouncer processes. For details see the Default: 0 Sets the This is currently only supported on Linux. Default: 1 on Linux, otherwise 0 Default: not set Turns on basic keepalive with OS defaults. On Linux, the system defaults aretcp_keepidle=7200,tcp_keepintvl=75,tcp_keepcnt=9. They are probably similar on other operating systems. Default: 1 Default: not set Default: not set Default: not set Sets the This is currently only supported on Linux. Default: 0 The section where the key will be taken as a database name and the value as a connection string, consisting of key=value pairs of connection parameters, described below (similar tolibpq, but the actuallibpq is not used and the set of available features is different). Example: The database name can contain characters The database name In this case, a connection topgbouncer specifying a database Such automatically created database entries are cleaned up if they stay idle longer than the time specified by the Destination database name. Default: same as client-side database name Host name or IP address to connect to. Host names are resolved at connection time, the result is cached per If the value begins with A comma-separated list of host names or addresses can be specified. In that case, connections are made in a round-robin manner. (If a host list contains host names that in turn resolve viaDNS to multiple addresses, the round-robin systems operate independently. This is an implementation dependency that is subject to change.) Note that in a list, all hosts must be available at all times: there are no mechanisms to skip unreachable hosts or to select only available hosts from a list or similar. (This is different from what a host list inlibpq means.) Also note that this only affects how the destinations of new connections are chosen. See also the setting Examples: Default: not set, meaning to use a Unix socket Default: 5432 If Otherwisepgbouncer logs into the destination database with the client user name, meaning that there will be one pool per user. If no password is specified here, the password from the Override of the global Set the maximum size of pools for this database. If not set, the Set the minimum pool size for this database. If not set, the global Set additional connections for this database. If not set, Query to be executed after a connection is established, but before allowing the connection to be used by any clients. If the query raises errors, they are logged but ignored otherwise. Set the pool mode specific to this database. If not set, the default Configure a database-wide maximum (i.e. all pools within the database will not have more than this many server connections). Ask specific Ask specific Ask specific This section contains key=value lines like where the key will be taken as a user name and the value as a list of configuration settings specific for this user. Example: Only a few settings are available here. Set the pool mode to be used for all connections from this user. If not set, the database or default Configure a maximum for the user (i.e. all pools with the user will not have more than this many server connections). This section defines the peers thatpgbouncer can forward cancellation requests to and where those cancellation requests will be routed. pgbouncer processes can be peered together in a group by defining a The section contains key=value lines like where the key will be taken as a Example: For peering to work, the Host name or IP address to connect to. Host names are resolved at connection time, the result is cached per If the value begins with Examples: Default: 6432 If not set, the Thepgbouncer configuration file can contain include directives, which specify another configuration file to read and process. This allows splitting the configuration file into physically separate parts. The include directives look like this: If the filename is not an absolute path, it is taken as relative to the current working directory. This section describes the format of the file specified by the There should be at least two fields, surrounded by double quotes. The first field is the user name and the second is either a plain-text, a MD5-hashed password, or a SCRAM secret.pgbouncer ignores the rest of the line. Double quotes in a field value can be escaped by writing two double quotes. Postgres Pro MD5-hashed password format: So user Postgres Pro SCRAM secret format: The passwords or secrets stored in the authentication file serve two purposes. First, they are used to verify the passwords of incoming client connections, if a password-based authentication method is configured. Second, they are used as the passwords for outgoing connections to the backend server, if the backend server requires password-based authentication (unless the password is specified directly in the database's connection string). The latter works if the password is stored in plain text or MD5-hashed. SCRAM secrets can only be used for logging into a server if the client authentication also uses SCRAM, thepgbouncer database definition does not specify a user name, and the SCRAM secrets are identical inpgbouncer and thePostgres Pro server (same salt and iterations, not merely the same password). This is due to an inherent security property of SCRAM: the stored SCRAM secret cannot by itself be used for deriving login credentials. The authentication file can be written by hand, but it's also useful to generate it from some other list of users and passwords. See Alternatively, use The location of theHBA file is specified by the setting The file follows the format of thePostgres Pro Supported record types: Database field: Supports User name field: Supports Address field: Supports Auth-method field: Only methods supported bypgbouncer's Small example configuration: Database examples: Example of a secure function for Example configs for 2 peeredpgbouncer processes to create a multi-corepgbouncer setup using The config for the first process: The config for the second process:pgbouncer is allowed to log in without password, if the login comes via the Unix socket and the client has same Unix useruid as the running process.Show Commands
SHOW commands output information. Each command is described below.SHOW STATS
stats_period. database total_xact_count total_query_count total_received total_sent total_xact_time total_query_time total_wait_time avg_xact_count avg_query_count avg_recv avg_sent avg_xact_time avg_query_time avg_wait_timestats_period, in microseconds (average per second within that period).SHOW STATS_TOTALS
SHOW STATS showing the total values (total_).SHOW STATS_AVERAGES
SHOW STATS showing the average values (avg_).SHOW TOTALS
SHOW STATS but aggregated across all databases.SHOW SERVERS
type user database stateactive,idle,used,tested,new,active_cancel, orbeing_canceled. addr port local_addr local_port connect_time request_time wait wait_us close_neededRECONNECT was issued. ptr link remote_pid tls application_nameapplication_name set on the linked client connection, or empty if this is not set, or if there is no linked connection.SHOW CLIENTS
type user database stateactive,waiting,active_cancel_req, orwaiting_cancel_req. addr port local_addr local_port connect_time request_time wait wait_us close_needed ptr link remote_pid tls application_nameapplication_name set by the client for this connection, or empty if this is not set.SHOW POOLS
database user cl_active cl_waiting cl_active_cancel_req cl_waiting_cancel_req sv_active sv_active_cancel sv_being_canceled sv_idle sv_usedserver_check_delay, so they needserver_check_query to run on them before they can be used again. sv_testedserver_reset_query orserver_check_query. sv_login maxwaitpool_size setting. maxwait_us pool_modeSHOW PEER_POOLS
peer_pool entry is made for each configured peer. database cl_active_cancel_req cl_waiting_cancel_req sv_active_cancel sv_loginSHOW LISTS
databases users pools free_clients used_clients login_clientslogin state. free_servers used_servers dns_names dns_zones dns_queries dns_pendingSHOW USERS
name pool_modepool_mode, orNULL if the default will be used instead.SHOW DATABASES
name host port database force_user pool_size min_pool_size reserve_pool pool_modeNULL if the default will be used instead. max_connectionsmax_db_connections, either globally or per database. current_connections paused disabledSHOW PEERS
peer_id host port pool_sizeSHOW FDS
pgbouncer, connects through the Unix socket and has the sameUID as the running process, the actual FDs are passed over the connection. This mechanism is used to do an online restart.Note
fd taskpooler,client orserver. user database addrunix if a Unix socket is used. port cancel linkNULL if idle.SHOW SOCKETS, SHOW ACTIVE_SOCKETS
SHOW CLIENTS andSHOW SERVERS as well as other more low-level information.SHOW CONFIG
key value default changeableyes orno, shows if the variable can be changed while running. Ifno, the variable can be changed only at boot-time. UseSET to change a variable at run time.SHOW MEM
SHOW DNS_HOSTS
hostname ttl addrsSHOW DNS_ZONES
SHOW VERSION
SHOW STATE
Process Controlling Commands
PAUSE [
db]RESUME is called.DISABLE
dbENABLE
dbDISABLE command.RECONNECT
dbpgbouncer.ini has been changed and reloaded (seeRELOAD) or when DNS resolution has changed, because then the equivalent of this command will be run automatically. This command is only necessary if something downstream ofpgbouncer routes the connections.PAUSE is recommended instead. To close server connections without waiting (for example, in emergency failover rather than gradual switchover scenarios), also considerKILL.KILL
dbRESUME is called.SUSPEND
RESUME is called.RESUME [
db]KILL,PAUSE, orSUSPEND command.SHUTDOWN
RELOAD
auth_file andauth_hba_file.WAIT_CLOSE [
db]close_needed state (seethe section called “SHOW SERVERS”). This can be called after aRECONNECT orRELOAD to wait until the respective configuration change has been fully activated, for example in switchover scripts.Other Commands
SET
key =argSET log_connections = 1;SET server_check_query = 'select 2';
SET command run on another database will be passed to thePostgres Pro backend like any other SQL command.)Signals
SIGHUPRELOAD on the console. SIGINTPAUSE andSHUTDOWN on the console. SIGTERMSHUTDOWN on the console. SIGUSR1PAUSE on the console. SIGUSR2RESUME on the console.Libevent Settings
epoll,kqueue,devpoll,poll, orselect by setting the environment variableEVENT_NOEPOLL,EVENT_NOKQUEUE,EVENT_NODEVPOLL,EVENT_NOPOLL orEVENT_NOSELECT, respectively.EVENT_SHOW_METHOD,libevent displays the kernel notification method that it uses.pgbouncer.ini Configuration File.ini format. Section names are between[ and]. Lines starting with; or# are taken as comments and ignored. The characters; and# are not recognized as special when they appear later in the line.Generic Settings
logfile-d), either this orsyslog has to be set. The log file is kept open, so after rotation,kill -HUP or on consoleRELOAD; should be done. On Windows, the service must be stopped and started.logfile does not by itself turn off logging tostderr. Use the command-line option-q or-d for that.pidfilepidfile set, daemonization (-d) is not allowed.listen_addr* meaning "listen on all addresses". When not set, only Unix socket connections are accepted.listen_portunix_socket_dir@ specifies that a Unix socket in the abstract namespace should be created (currently supported on Linux and Windows).-R) to work, a Unix socket needs to be configured, and it needs to be in the file-system namespace./tmp (empty on Windows)unix_socket_modeunix_socket_groupuserpool_mode session transaction statementmax_client_connmax_client_conn. If each user connects under its own username to the server, the theoretical maximum used is:max_client_conn + (max pool_size * total databases * total users)
max_client_conn + (max pool_size * total databases)
ulimit in your favorite shell man page. Note:ulimit does not apply in a Windows environment.default_pool_sizemin_pool_sizereserve_pool_sizereserve_pool_timeout). The 0 value disables this parameter.reserve_pool_timeoutmax_db_connections[databases] section.max_user_connections[users] section.server_round_robintrack_extra_parametersclient_encoding,datestyle,timezone,standard_conforming_strings andapplication_name parameters per client. To allow other parameters to be tracked, they can be specified here, so thatpgbouncer knows that they should be maintained in the client variable cache and restored in the server whenever the client becomes active.default_transaction_readonly, IntervalStyle)Note
search_path.postgres protocol allows specifying parameter settings, both directly as a parameter in the startup packet, or inside theoptions startup packet. Parameters specified using both of these methods are supported bytrack_extra_parameters. However, it's not possible to includeoptions itself intrack_extra_parameters, only the parameters contained inoptions.IntervalStyleignore_startup_parametersclient_encoding,datestyle,timezone andstandard_conforming_strings.options,extra_float_digits).postgres protocol allows specifying parameter settings, both directly as a parameter in the startup packet, or inside theoptions startup packet. Parameters specified using both of these methods are supported byignore_startup_parameters. It's even possible to includeoptions itself inignore_startup_parameters, which results in any unknown parameters contained insideoptions to be ignored.peer_idpeer_id value should be unique within a group of peeredpgbouncer processes. When set to 0,pgbouncer peering is disabled. See also[peers] Section for more information. The maximum value that can be used for thepeer_id is 16383.disable_pqexecapplication_name_add_hostapplication_name is later changed withSET,pgbouncer does not change it again.conffileRELOAD /SIGHUP.service_namepgbouncerjob_nameservice_name.stats_periodSHOW commands are updated and how often aggregated statistics are written to the log (but seelog_stats). [seconds]Authentication Settings
auth_type certCommonName field from the certificate. md5auth_file may contain both MD5-encrypted and plain-text passwords. Ifmd5 is configured and a user has a SCRAM secret, then SCRAM authentication is used automatically instead. scram-sha-256auth_file has to contain SCRAM secrets or plain-text passwords. Note that SCRAM secrets can only be used for verifying the password of a client but not for logging into a server. To be able to use SCRAM on server connections, use plain-text passwords. plain trustauth_file. anytrust method, but the user name given is ignored. Requires that all databases are configured to log in as a specific user. Additionally, the console database allows any user to log in as admin. hbaauth_hba_file. This allows different authentication methods for different access paths, for example: connections over Unix socket usepeer authentication method, connections overTCP must useTLS. pamauth_file is ignored. This method is not compatible with databases using theauth_user option. The service name reported toPAM ispgbouncer.pam is not supported in theHBA configuration file.auth_hba_fileauth_type ishba.auth_fileauth_type) require that eitherauth_file orauth_user be set; otherwise there would be no users defined.auth_userauth_user is set, then any user not specified inauth_file will be queried through theauth_query query frompg_shadow in the database, usingauth_user. The password ofauth_user will be taken fromauth_file. (Ifauth_user does not require a password, then it does not need to be defined inauth_file.)pg_shadow requires admin rights. It's preferable to use a non-superuser that calls aSECURITY DEFINER function instead.auth_querypg_shadow requires admin rights. It's preferable to use a non-superuser that calls aSECURITY DEFINER function instead.SELECT usename, passwd FROM pg_shadow WHERE usename=$1auth_dbnameLog Settings
syslogsyslog_identpgbouncer (program name)syslog_facilityauth,authpriv,daemon,user,local0-7.daemonlog_connectionslog_disconnectionslog_pooler_errorslog_statsstats_period. This can be disabled if external monitoring tools are used to grab the same data fromSHOW commands.verbose-v switch on the command line. For example, using-v -v on the command line is the same asverbose=2.Console Access Control
admin_usersauth_type isany, in which case any user name is allowed in as admin.stats_usersSHOW commands exceptSHOW FDS.Connection Sanity Checks, Timeouts
server_reset_queryABORT orROLLBACK.DISCARD ALL, which cleans everything, but that leaves the next client no pre-cached state. It can be made lighter, e.g.DEALLOCATE ALL to just drop prepared statements, if the application does not break when some state is kept around.server_reset_query is not used, because in that mode, clients must not use any session-based features, since each transaction ends up in a different connection and thus gets a different session state.DISCARD ALLserver_reset_query_alwaysserver_reset_query should be run in all pooling modes. When this setting is off (default), theserver_reset_query will be run only in pools that are in sessions-pooling mode. Connections in transaction-pooling mode should not have any need for a reset query.server_check_delayserver_check_query on it. If 0 then the query is always run.server_check_queryselect 1server_fast_closeclose_needed mode (set byRECONNECT,RELOAD that changes connection settings, or DNS change), rather than waiting for the session end. In statement or transaction pooling mode, this has no effect since that is the default behavior there.server_lifetimeserver_idle_timeoutserver_connect_timeoutserver_login_retryPAUSE command to avoid this.client_login_timeoutSUSPEND and thus online restart. [seconds]autodb_idle_timeoutdns_max_ttldns_nxdomain_ttlNXDOMAINDNS lookups can be cached. [seconds]dns_zone_check_periodresolv_confresolv.conf file. This is to allow specifying customDNS servers and perhaps other name resolution options, independent of the global operating system configuration.TLS Settings
client_tls_sslmodeclient_tls_key_file andclient_tls_cert_file must be also configured to set up the key and certificatepgbouncer uses to accept client connections. disable allow preferallow. require verify-ca verify-fullverify-ca.client_tls_key_fileclient_tls_cert_fileclient_tls_ca_fileclient_tls_protocolstlsv1.0,tlsv1.1,tlsv1.2,tlsv1.3. Shortcuts:all (tlsv1.0,tlsv1.1,tlsv1.2,tlsv1.3),secure (tlsv1.2,tlsv1.3),legacy (all).secureclient_tls_ciphersdefault/secure,compat/legacy,insecure/all,normal,fast.fastclient_tls_ecdhcurvenone (DH is disabled),auto (256-bit ECDH), curve name.autoclient_tls_dheparamsnone (DH is disabled),auto (2048-bit DH),legacy (1024-bit DH).autoserver_tls_sslmodeprefer. disable prefer require verify-caserver_tls_ca_file. Server host name is not checked against certificate. verify-fullserver_tls_ca_file. Server host name must match certificate information.server_tls_ca_fileserver_tls_key_fileserver_tls_cert_fileserver_tls_protocolstlsv1.0,tlsv1.1,tlsv1.2,tlsv1.3. Shortcuts:all (tlsv1.0,tlsv1.1,tlsv1.2,tlsv1.3),secure (tlsv1.2,tlsv1.3),legacy (all).secureserver_tls_ciphersdefault/secure,compat/legacy,insecure/all,normal,fast.fastDangerous Timeouts
query_timeoutstatement_timeout, to apply only for network problems. [seconds]query_wait_timeoutcancel_wait_timeoutclient_idle_timeoutidle_transaction_timeoutsuspend_timeoutSUSPEND or reboot (-R). A connection is dropped if the flush does not succeed. [seconds]Low-Level Network Settings
pkt_bufmax_packet_sizelisten_backlogbacklog argument forlisten(). Determines how many new unanswered connection attempts are kept in the queue. When the queue is full, further new connections are dropped.sbuf_loopcntpkt_buf amount of data. 0 means no limit.so_reuseportSO_REUSEPORT onTCP listening sockets. On some operating systems, this allows running multiplepgbouncer instances on the same host listening on the same port and having the kernel distribute the connections automatically. This option is a way to getpgbouncer to use more CPU cores. (pgbouncer is single-threaded and uses one CPU core per instance.)unix_socket_dir andpidfile, as well aslogfile if that is used. Also note that if you make use of this option, you can no longer connect to a specificpgbouncer instance via TCP/IP, which might have implications for monitoring and metrics collection.peer_id configuration option and the[peers] configuration section. There's also an example that uses peering andso_reuseport in theExamples section.tcp_defer_acceptTCP_DEFER_ACCEPT socket option; seeman 7 tcp for details. (This is a Boolean option: 1 means enabled. The actual value set if enabled is currently hardcoded to 45 seconds.)tcp_socket_buffertcp_keepalivetcp_keepcnttcp_keepidletcp_keepintvltcp_user_timeoutTCP_USER_TIMEOUT socket option. This specifies the maximum amount of time in milliseconds that transmitted data may remain unacknowledged before the TCP connection is forcibly closed. If set to 0, then operating system's default is used.Section [databases]
[databases] defines the names of the databases that clients ofpgbouncer can connect to and specifies where those connections will be routed. The section contains key=value lines likedbname = connection string
foodb = host=host1.example.com port=5432bardb = host=localhost dbname=bazdb
_0-9A-Za-z without quoting. Names that contain other chars need to be quoted with standard SQL ident quoting: double quotes where "" is taken as single quote.pgbouncer is reserved for the admin console and cannot be used as a key here.* acts as fallback database: if the exact name does not exist, its value is taken as connection string for the requested database. For example, if there is the following entry (and no other overriding entries):* = host=foo
bar will effectively behave as if the following entry exists (taking advantage of the default fordbname being the client-side database name):bar = host=foo dbname=bar
autodb_idle_timeout parameter.dbnamehostdns_max_ttl parameter. When a host name's resolution changes, existing server connections are automatically closed when they are released (according to the pooling mode), and new server connections immediately use the new resolution. IfDNS returns several results, they are used in a round-robin manner./, then a Unix socket in the file-system namespace is used. If the value begins with@, then a Unix socket in the abstract namespace is used.server_round_robin for how clients are assigned to already established server connections.host=localhosthost=127.0.0.1host=2001:0db8:85a3:0000:0000:8a2e:0370:7334host=/var/run/postgresqlhost=192.168.0.1,192.168.0.2,192.168.0.3
portuseruser= is set, all connections to the destination database will be done with the specified user, meaning that there will be only one pool for this database.passwordauth_file orauth_query will be used.auth_userauth_user setting, if specified.pool_sizedefault_pool_size is used.min_pool_sizemin_pool_size is used.reserve_poolreserve_pool_size is used.connect_querypool_modepool_mode is used.max_db_connectionsclient_encodingclient_encoding from server.datestyledatestyle from server.timezonetimezone from server.Section [users]
user1 = settings
user1 = pool_mode=session
pool_modepool_mode is used.max_user_connectionsSection [peers]
peer_id value and a[peers] section in the configs of all thepgbouncer processes. Thesepgbouncer processes can then forward cancellation requests to the process that it originated from. This is needed to make cancellations work when multiplepgbouncer processes (possibly on different servers) are behind the same TCP load balancer. Cancellation requests are sent over different TCP connections than the query they are cancelling, so a TCP load balancer might send the cancellation request connection to a different process than the one that it was meant for. By peering them these cancellation requests eventually end up at the right process.peer_id = connection string
peer_id and the value as a connection string, consisting of key=value pairs of connection parameters, described below (similar tolibpq, but the actuallibpq is not used and the set of available features is different).1 = host=host1.example.com2 = host=/tmp/pgbouncer-2 port=5555
Note
peer_id of eachpgbouncer process in the group must be unique within the peered group. And the[peers] section should contain entries for each of those peer IDs. An example can be found in theExamples section. It is allowed, but not necessary, for the[peers] section to contain thepeer_id of thepgbouncer that the config is for. Such an entry will be ignored, but it is allowed to make config management easier. Because it allows using the exact same[peers] section for multiple configs.hostdns_max_ttl parameter. If DNS returns several results, they are used in a round-robin manner. But in general it's not recommended to use a hostname that resolves to multiple IPs, because then the cancel request might still be forwarded to the wrong node and it would need to be forwarded again (which is only allowed up to three times)./, then a Unix socket in the file-system namespace is used. If the value begins with@, then a Unix socket in the abstract namespace is used.host=localhosthost=127.0.0.1host=2001:0db8:85a3:0000:0000:8a2e:0370:7334host=/var/run/pgbouncer-1
portpool_sizedefault_pool_size is used.Include Directive
%include
filenameAuthentication File Format
auth_file setting. It is a text file in the following format:"username1" "password" ..."username2" "md5abcdef012342345" ..."username2" "SCRAM-SHA-256$
iterations:salt$storedkey:serverkey""md5" + md5(password + username)
admin with password1234 will have MD5-hashed passwordmd545f2603610af569b6155c45067268c6b.SCRAM-SHA-256$
iterations:salt$storedkey:serverkey./etc/mkauth.py for a sample script to generate the authentication file from thepg_shadow system table.auth_query instead ofauth_file to avoid having to maintain a separate authentication file.HBA File Format
auth_hba_file. It is only used ifauth_type is set tohba.pg_hba.conf file described inSection 19.1.local,host,hostssl,hostnossl.all,sameuser, @file, multiple names. Not supported:replication,samerole,samegroup.all, @file, multiple names. Not supported:+groupname.IPv4,IPv6. Not supported: DNS names, domain prefixes.auth_type are supported, pluspeer andreject, but exceptany andpam, which only work globally. User name map (map=) parameter is not supported.Examples
[databases]template1 = host=localhost dbname=template1 auth_user=someuser[pgbouncer]pool_mode = sessionlisten_port = 6432listen_addr = localhostauth_type = md5auth_file = users.txtlogfile = pgbouncer.logpidfile = pgbouncer.pidadmin_users = someuserstats_users = stat_collector
[databases]; foodb over Unix socketfoodb =; redirect bardb to bazdb on localhostbardb = host=localhost dbname=bazdb; access to destination database will go with single userforcedb = host=localhost port=300 user=baz password=foo client_encoding=UNICODE datestyle=ISO
auth_query:CREATE OR REPLACE FUNCTION pgbouncer.user_lookup(in i_username text, out uname text, out phash text)RETURNS record AS $$BEGIN SELECT usename, passwd FROM pg_catalog.pg_shadow WHERE usename = i_username INTO uname, phash; REURN;END;$$ LANUAGE plpgsql SECURITY DEFINER;REVOKEALL ON FUNCTION pgbouncer.user_lookup(text) FROM public, pgbouncer;GRANT XECUTE ON FUNCTION pgbouncer.user_lookup(text) TO pgbouncer;
so_reuseport.[databases]postgres = host=localhost dbname=postgres[peers]1 = host=/tmp/pgbouncer12 = host=/tmp/pgbouncer2[pgbouncer]listen_addr=127.0.0.1auth_file=auth_file.confso_reuseport=1unix_socket_dir=/tmp/pgbouncer1peer_id=1
[databases]postgres = host=localhost dbname=postgres[peers]1 = host=/tmp/pgbouncer12 = host=/tmp/pgbouncer2[pgbouncer]listen_addr=127.0.0.1auth_file=auth_file.confso_reuseport=1; only unix_socket_dir and peer_id are differentunix_socket_dir=/tmp/pgbouncer2peer_id=2