View Source pg (kernel v9.3)
Distributed named process groups.
This module implements process groups. A message can be sent to one, some, or all group members.
Up until OTP 17 there used to exist an experimental pg module in stdlib.
This pg module is not the same module as that experimental pg module, and
only share the same module name.
A group of processes can be accessed by a common name. For example, if there is
a group named foobar, there can be a set of processes (which can be located on
different nodes) that are all members of the group foobar. There are no
special functions for sending a message to the group. Instead, client functions
are to be written with the functions get_members/1 and get_local_members/1
to determine which processes are members of the group. Then the message can be
sent to one or more group members.
If a member terminates, it is automatically removed from the group.
A process may join multiple groups. It may join the same group multiple times. It is only allowed to join processes running on local node.
Process Groups implement strong eventual consistency. Process Groups membership
view may temporarily diverge. For example, when processes on node1 and node2
join concurrently, node3 and node4 may receive updates in a different order.
Membership view is not transitive. If node1 is not directly connected to
node2, they will not see each other groups. But if both are connected to
node3, node3 will have the full view.
Groups are automatically created when any process joins, and are removed when all processes leave the group. Non-existing group is considered empty (containing no processes).
Process groups can be organised into multiple scopes. Scopes are completely
independent of each other. A process may join any number of groups in any number
of scopes. Scopes are designed to decouple single mesh into a set of overlay
networks, reducing amount of traffic required to propagate group membership
information. Default scope pg is started automatically when
kernel(6) is configured to do so.
Note
Scope name is used to register process locally, and to name an ETS table. If there is another process registered under this name, or another ETS table exists, scope fails to start.
Local membership is not preserved if scope process exits and restarts.
A scope can be kept local-only by using a scope name that is unique cluster-wide, e.g. the node name:
- ____ -
pg:start_link(node()).
See Also
Summary
Functions
Equivalent to demonitor/2.
Unsubscribes the caller from updates (scope or group). Flushes all outstanding updates that were already in the message queue of the calling process.
Equivalent to get_local_members/2.
Returns all processes running on the local node in the group Group. Processes
are returned in no specific order. This function is optimised for speed.
Equivalent to get_members/2.
Returns all processes in the group Group. Processes are returned in no
specific order. This function is optimised for speed.
Equivalent to join/3.
Joins single process or multiple processes to the group Group. A process can
join a group many times and must then leave the group the same number of times.
Equivalent to leave/3.
Makes the process PidOrPids leave the group Group. If the process is not a
member of the group, not_joined is returned.
Equivalent to monitor/2.
Subscribes the caller to updates for the specified group. Returns list of processes currently in the group, and a reference to match the upcoming notifications.
Equivalent to monitor_scope/1.
Subscribes the caller to updates from the specified scope. Returns content of the entire scope and a reference to match the upcoming notifications.
Equivalent to start_link/1.
Starts the default pg scope within supervision tree. Kernel may be configured
to do it automatically, see kernel(6) configuration
manual.
Starts additional scope.
Equivalent to which_groups/1.
Returns a list of all known groups.
Types
-type group() :: any().
The identifier of a process group.
Functions
-spec demonitor(Ref :: reference()) -> ok | false.
Equivalent to demonitor/2.
Unsubscribes the caller from updates (scope or group). Flushes all outstanding updates that were already in the message queue of the calling process.
Equivalent to get_local_members/2.
Returns all processes running on the local node in the group Group. Processes
are returned in no specific order. This function is optimised for speed.
Equivalent to get_members/2.
Returns all processes in the group Group. Processes are returned in no
specific order. This function is optimised for speed.
Equivalent to join/3.
Joins single process or multiple processes to the group Group. A process can
join a group many times and must then leave the group the same number of times.
PidOrPids may contain the same process multiple times.
Equivalent to leave/3.
Makes the process PidOrPids leave the group Group. If the process is not a
member of the group, not_joined is returned.
When list of processes is passed as PidOrPids, function returns not_joined
only when all processes of the list are not joined.
Equivalent to monitor/2.
Subscribes the caller to updates for the specified group. Returns list of processes currently in the group, and a reference to match the upcoming notifications.
See monitor_scope/0 for the update message structure.
Equivalent to monitor_scope/1.
Subscribes the caller to updates from the specified scope. Returns content of the entire scope and a reference to match the upcoming notifications.
Whenever any group membership changes, an update message is sent to the subscriber:
{Ref, join, Group, [JoinPid1, JoinPid2]}{Ref, leave, Group, [LeavePid1]}Equivalent to start_link/1.
Starts the default pg scope within supervision tree. Kernel may be configured
to do it automatically, see kernel(6) configuration
manual.
Starts additional scope.
-spec which_groups() -> [Group :: group()].
Equivalent to which_groups/1.
Returns a list of all known groups.