patroni.postgresql.mpp.citus module

Bases: AbstractMPP

_abc_impl = <_abc._abc_data object>

property coordinator_group_id : int View on GitHub

The group id of the Citus coordinator PostgreSQL cluster.

property group : int View on GitHub

The group of this Citus node.

group_re : Any = re.compile('^(0|[1-9][0-9]*)$')

static validate_config ( config : Any ) → bool View on GitHub

Check whether provided config is good for a given MPP.

Parameters :

config – configuration of citus MPP section.

Returns :

True is config passes validation, otherwise False .

Bases: Citus , AbstractMPPHandler , Thread

Define the interfaces for handling an underlying Citus cluster.

__init__ ( postgresql : Postgresql , config : Dict [ str , str | int ] ) → None View on GitHub

"Initialize a new instance of CitusHandler .

Parameters :

• postgresql – the Postgres node.

• config – the citus MPP config section.

_abc_impl = <_abc._abc_data object>

_add_task ( task : PgDistTask ) → bool View on GitHub

static _pg_dist_node ( role : str , conn_url : str ) → PgDistNode | None View on GitHub

add_task ( event : str , groupid : int , cluster : Cluster , leader_name : str , leader_url : str , timeout : float | None = None , cooldown : float | None = None ) → PgDistTask | None View on GitHub

adjust_postgres_gucs ( parameters : Dict [ str , Any ] ) → None View on GitHub

Adjust GUCs in the current PostgreSQL configuration.

Parameters :

parameters – dictionary of GUCs, with key as GUC name and the corresponding value as current GUC value.

bootstrap ( ) → None View on GitHub

Bootstrap handler.

Is called when the new cluster is initialized (through initdb or a custom bootstrap method).

find_task_by_groupid ( groupid : int ) → int | None View on GitHub

handle_event ( cluster : Cluster , event : Dict [ str , Any ] ) → None View on GitHub

Handle an event sent from a worker node.

Parameters :

• cluster – the currently known cluster state from DCS.

• event – the event to be handled.

ignore_replication_slot ( slot : Dict [ str , str ] ) → bool View on GitHub

Check whether provided replication slot existing in the database should not be removed.

Note

MPP database may create replication slots for its own use, for example to migrate data between workers using logical replication, and we don’t want to suddenly drop them.

Parameters :

slot – dictionary containing the replication slot settings, like name , database , type , and plugin .

Returns :

True if the replication slots should not be removed, otherwise False .

load_pg_dist_group ( ) → bool View on GitHub

Read from the pg_dist_node table and put it into the local cache

on_demote ( ) → None View on GitHub

On demote handler.

Is called when the primary was demoted.

pick_task ( ) → Tuple [ int | None , PgDistTask | None ] View on GitHub

Returns the tuple(i, task), where i - is the task index in the self._tasks list

Tasks are picked by following priorities:

1. If there is already a transaction in progress, pick a task that that will change already affected worker primary.

2. If the coordinator address should be changed - pick a task with groupid=0 (coordinators are always in groupid 0).

3. Pick a task that is the oldest (first from the self._tasks)

process_task ( task : PgDistTask ) → bool View on GitHub

Updates a single row in pg_dist_group table, optionally in a transaction.

The transaction is started if we do a demote of the worker node or before promoting the other worker if there is no transaction in progress. And, the transaction is committed when the switchover/failover completed.

Parameters :

task – reference to a PgDistTask object that represents a row to be updated/created.

Returns :

True if the row was successfully created/updated or transaction in progress was committed as an indicator that the self._pg_dist_group cache should be updated, or, if the new transaction was opened, this method returns False .

process_tasks ( ) → None View on GitHub

query ( sql : str , * params : Any ) → List [ Tuple [ Any , ... ] ] View on GitHub

run ( ) → None View on GitHub

Method representing the thread’s activity.

You may override this method in a subclass. The standard run() method invokes the callable object passed to the object’s constructor as the target argument, if any, with sequential and keyword arguments taken from the args and kwargs arguments, respectively.

schedule_cache_rebuild ( ) → None View on GitHub

Cache rebuild handler.

Is called to notify handler that it has to refresh its metadata cache from the database.

sync_meta_data ( cluster : Cluster ) → None View on GitHub

Maintain the pg_dist_node from the coordinator leader every heartbeat loop.

We can’t always rely on REST API calls from worker nodes in order to maintain pg_dist_node , therefore at least once per heartbeat loop we make sure that works registered in self._pg_dist_group cache are matching the cluster view from DCS by creating tasks the same way as it is done from the REST API.

update_group ( task : PgDistTask , transaction : bool ) → None View on GitHub

update_node ( groupid : int , node : PgDistNode , cooldown : float = 10000 ) → None View on GitHub

Bases: Set [ PgDistNode ]

A set -like object that represents a Citus group in "pg_dist_node" table.

This class implements a set of methods to compare topology and if it is necessary to transition from the old to the new topology in a "safe" manner:

• register new primary/secondaries

• replace gone secondaries with added secondaries

• failover and switchover

Typically there will be at least one PgDistNode object registered ( primary ). In addition to that there could be one or more secondary nodes.

Variables :

• failover – whether the primary row should be updated as a result of transition() method call.

• groupid – the "groupid" from "pg_dist_node".

__init__ ( groupid : int , nodes : Collection [ PgDistNode ] | None = None ) → None View on GitHub

Creates a PgDistGroup object based on given arguments.

Parameters :

• groupid – the groupid from "pg_dist_node".

• nodes – a collection of PgDistNode objects that belong to a groupid .

equals ( other : PgDistGroup , check_nodeid : bool = False ) → bool View on GitHub

Compares two PgDistGroup objects.

Parameters :

• other – what we want to compare with.

• check_nodeid – whether PgDistNode.nodeid should be compared in addition to PgDistNode.host , PgDistNode.port , and PgDistNode.role .

Returns :

True if two PgDistGroup objects are fully identical.

get ( value : PgDistNode ) → PgDistNode | None View on GitHub

Performs a lookup of the actual value in a set.

Note

It is necessary because __hash__() and __eq__() methods in PgDistNode are redefined and effectively they check only PgDistNode.host and PgDistNode.port attributes.

Parameters :

value – the key we search for.

Returns :

the actual PgDistNode value from this PgDistGroup object or None if not found.

primary ( ) → PgDistNode | None View on GitHub

Finds and returns PgDistNode object that represents the "primary".

Returns :

PgDistNode object which represents the "primary" or None if not found.

transition ( old : PgDistGroup ) → Iterator [ PgDistNode ] View on GitHub

Compares this topology with the old one and yields transitions that transform the old to the new one.

Note

The actual yielded object is PgDistNode that will be passed to the CitusHandler.update_node() to execute all transitions in a transaction.

In addition to the yielding transactions this method fills up PgDistNode.nodeid attribute for nodes that are presented in the old and in the new topology.

There are a few simple rules/constraints that are imposed by Citus and must be followed: - adding/removing nodes is only possible when metadata is synced to all registered "priorities".

• the "primary" row in "pg_dist_node" always keeps the nodeid (unless it is removed, but it is not supported by Patroni).

• "nodename", "nodeport" must be unique across all rows in the "pg_dist_node". This means that every time we want to change the nodeid of an existing node (i.e. to change it from secondary to primary), we should first write some other "nodename"/"nodeport" to the row it’s currently in.

• updating "broken" nodes always works and metadata is synced asynchnonously after the commit.

Following these rules below is an example of the switchover between node1 (primary, nodeid=4) and node2 (secondary, nodeid=5).

BEGIN;
    SELECT citus_update_node(4, 'node1-demoted', 5432);
    SELECT citus_update_node(5, 'node1', 5432);
    SELECT citus_update_node(4, 'node2', 5432);
COMMIT;

Parameters :

old – the last known topology registered in "pg_dist_node" for a given groupid .

Yields :

PgDistNode objects that must be updated/added/removed in "pg_dist_node".

Bases: object

Represents a single row in "pg_dist_node" table.

Variables :

• host – "nodename" value

• port – "nodeport" value

• role – "noderole" value

• nodeid – "nodeid" value

__init__ ( host : str , port : int , role : str , nodeid : int | None = None ) → None View on GitHub

Create a PgDistNode object based on given arguments.

Parameters :

• host – "nodename" of the Citus coordinator or worker.

• port – "nodeport" of the Citus coordinator or worker.

• role – "noderole" value.

• nodeid – id of the row in the "pg_dist_node".

as_tuple ( include_nodeid : bool = False ) → Tuple [ str , int , str , int | None ] View on GitHub

Helper method to compare two PgDistGroup objects.

Note

include_nodeid is set to True only in unit-tests.

Parameters :

include_nodeid – whether nodeid should be taken into account when comparison is performed.

Returns :

tuple object with host , port , role , and optionally nodeid .

is_primary ( ) → bool View on GitHub

Checks whether this object represents "primary" in a corresponding group.

Returns :

True if this object represents the primary .

Bases: PgDistGroup

A "task" that represents the current or desired state of "pg_dist_node" for a provided groupid .

Variables :

• group – the "groupid" in "pg_dist_node".

• event – an "event" that resulted in creating this task. possible values: "before_demote", "before_promote", "after_promote".

• timeout – a transaction timeout if the task resulted in starting a transaction.

• cooldown – the cooldown value for citus_update_node() UDF call.

• deadline – the time in unix seconds when the transaction is allowed to be rolled back.

__init__ ( groupid : int , nodes : Collection [ PgDistNode ] | None , event : str , timeout : float | None = None , cooldown : float | None = None ) → None View on GitHub

Create a PgDistTask object based on given arguments.

Parameters :

• groupid – the groupid from "pg_dist_node".

• nodes – a collection of PgDistNode objects that belong to a groupid .

• event – an "event" that resulted in creating this task.

• timeout – a transaction timeout if the task resulted in starting a transaction.

• cooldown – the cooldown value for citus_update_node() UDF call.

wait ( ) → None View on GitHub

Wait until this task is processed by a dedicated thread.

wakeup ( ) → None View on GitHub

Notify a thread that created a task that it was processed.