3 * Database load balancing interface
5 * This program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 2 of the License, or
8 * (at your option) any later version.
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
15 * You should have received a copy of the GNU General Public License along
16 * with this program; if not, write to the Free Software Foundation, Inc.,
17 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 * http://www.gnu.org/copyleft/gpl.html
23 namespace Wikimedia\Rdbms
;
26 use InvalidArgumentException
;
29 * Database cluster connection, tracking, load balancing, and transaction manager interface
31 * A "cluster" is considered to be one master database and zero or more replica databases.
32 * Typically, the replica DBs replicate from the master asynchronously. The first node in the
33 * "servers" configuration array is always considered the "master". However, this class can still
34 * be used when all or some of the "replica" DBs are multi-master peers of the master or even
35 * when all the DBs are non-replicating clones of each other holding read-only data. Thus, the
36 * role of "master" is in some cases merely nominal.
38 * By default, each DB server uses DBO_DEFAULT for its 'flags' setting, unless explicitly set
39 * otherwise in configuration. DBO_DEFAULT behavior depends on whether 'cliMode' is set:
40 * - In CLI mode, the flag has no effect with regards to LoadBalancer.
41 * - In non-CLI mode, the flag causes implicit transactions to be used; the first query on
42 * a database starts a transaction on that database. The transactions are meant to remain
43 * pending until either commitMasterChanges() or rollbackMasterChanges() is called. The
44 * application must have some point where it calls commitMasterChanges() near the end of
46 * Every iteration of beginMasterChanges()/commitMasterChanges() is called a "transaction round".
47 * Rounds are useful on the master DB connections because they make single-DB (and by and large
48 * multi-DB) updates in web requests all-or-nothing. Also, transactions on replica DBs are useful
49 * when REPEATABLE-READ or SERIALIZABLE isolation is used because all foriegn keys and constraints
50 * hold across separate queries in the DB transaction since the data appears within a consistent
51 * point-in-time snapshot.
53 * The typical caller will use LoadBalancer::getConnection( DB_* ) to yield a live database
54 * connection handle. The choice of which DB server to use is based on pre-defined loads for
55 * weighted random selection, adjustments thereof by LoadMonitor, and the amount of replication
56 * lag on each DB server. Lag checks might cause problems in certain setups, so they should be
57 * tuned in the server configuration maps as follows:
58 * - Master + N Replica(s): set 'max lag' to an appropriate threshold for avoiding any database
59 * lagged by this much or more. If all DBs are this lagged, then the load balancer considers
60 * the cluster to be read-only.
61 * - Galera Cluster: Seconds_Behind_Master will be 0, so there probably is nothing to tune.
62 * Note that lag is still possible depending on how wsrep-sync-wait is set server-side.
63 * - Read-only archive clones: set 'is static' in the server configuration maps. This will
64 * treat all such DBs as having 0 lag.
65 * - SQL load balancing proxy: any proxy should handle lag checks on its own, so the 'max lag'
66 * parameter should probably be set to INF in the server configuration maps. This will make
67 * the load balancer ignore whatever it detects as the lag of the logical replica is (which
68 * would probably just randomly bounce around).
70 * If using a SQL proxy service, it would probably be best to have two proxy hosts for the
71 * load balancer to talk to. One would be the 'host' of the master server entry and another for
72 * the (logical) replica server entry. The proxy could map the load balancer's "replica" DB to
73 * any number of physical replica DBs.
78 interface ILoadBalancer
{
79 /** @var int Request a replica DB connection */
80 const DB_REPLICA
= -1;
81 /** @var int Request a master DB connection */
84 /** @var string Domain specifier when no specific database needs to be selected */
85 const DOMAIN_ANY
= '';
87 /** @var int DB handle should have DBO_TRX disabled and the caller will leave it as such */
88 const CONN_TRX_AUTOCOMMIT
= 1;
89 /** @var int Alias for CONN_TRX_AUTOCOMMIT for b/c; deprecated since 1.31 */
90 const CONN_TRX_AUTO
= 1;
93 * Construct a manager of IDatabase connection objects
95 * @param array $params Parameter map with keys:
96 * - servers : Required. Array of server info structures.
97 * - localDomain: A DatabaseDomain or domain ID string.
98 * - loadMonitor : Name of a class used to fetch server lag and load.
99 * - readOnlyReason : Reason the master DB is read-only if so [optional]
100 * - waitTimeout : Maximum time to wait for replicas for consistency [optional]
101 * - maxLag: Avoid replica DB servers with more lag than this [optional]
102 * - srvCache : BagOStuff object for server cache [optional]
103 * - wanCache : WANObjectCache object [optional]
104 * - chronologyProtector: ChronologyProtector object [optional]
105 * - hostname : The name of the current server [optional]
106 * - cliMode: Whether the execution context is a CLI script. [optional]
107 * - profiler : Class name or instance with profileIn()/profileOut() methods. [optional]
108 * - trxProfiler: TransactionProfiler instance. [optional]
109 * - replLogger: PSR-3 logger instance. [optional]
110 * - connLogger: PSR-3 logger instance. [optional]
111 * - queryLogger: PSR-3 logger instance. [optional]
112 * - perfLogger: PSR-3 logger instance. [optional]
113 * - errorLogger : Callback that takes an Exception and logs it. [optional]
114 * @throws InvalidArgumentException
116 public function __construct( array $params );
119 * Get the index of the reader connection, which may be a replica DB
121 * This takes into account load ratios and lag times. It should
122 * always return a consistent index during a given invocation.
124 * Side effect: opens connections to databases
125 * @param string|bool $group Query group, or false for the generic reader
126 * @param string|bool $domain Domain ID, or false for the current domain
128 * @return bool|int|string
130 public function getReaderIndex( $group = false, $domain = false );
133 * Set the master wait position
135 * If a DB_REPLICA connection has been opened already, then wait immediately.
136 * Otherwise sets a variable telling it to wait if such a connection is opened.
138 * This only applies to connections to the generic replica DB for this request.
139 * If a timeout happens when waiting, then getLaggedReplicaMode()/laggedReplicaUsed()
142 * @param DBMasterPos|bool $pos Master position or false
144 public function waitFor( $pos );
147 * Set the master wait position and wait for a "generic" replica DB to catch up to it
149 * This can be used a faster proxy for waitForAll()
151 * @param DBMasterPos|bool $pos Master position or false
152 * @param int $timeout Max seconds to wait; default is mWaitTimeout
153 * @return bool Success (able to connect and no timeouts reached)
155 public function waitForOne( $pos, $timeout = null );
158 * Set the master wait position and wait for ALL replica DBs to catch up to it
160 * @param DBMasterPos|bool $pos Master position or false
161 * @param int $timeout Max seconds to wait; default is mWaitTimeout
162 * @return bool Success (able to connect and no timeouts reached)
164 public function waitForAll( $pos, $timeout = null );
167 * Get any open connection to a given server index, local or foreign
169 * @param int $i Server index or DB_MASTER/DB_REPLICA
170 * @return Database|bool False if no such connection is open
172 public function getAnyOpenConnection( $i );
175 * Get a connection handle by server index
177 * The CONN_TRX_AUTOCOMMIT flag is ignored for databases with ATTR_DB_LEVEL_LOCKING
178 * (e.g. sqlite) in order to avoid deadlocks. ILoadBalancer::getServerAttributes()
179 * can be used to check such flags beforehand.
181 * If the caller uses $domain or sets CONN_TRX_AUTOCOMMIT in $flags, then it must also
182 * call ILoadBalancer::reuseConnection() on the handle when finished using it.
183 * In all other cases, this is not necessary, though not harmful either.
185 * @param int $i Server index or DB_MASTER/DB_REPLICA
186 * @param array|string|bool $groups Query group(s), or false for the generic reader
187 * @param string|bool $domain Domain ID, or false for the current domain
188 * @param int $flags Bitfield of CONN_* class constants
190 * @note This method throws DBAccessError if ILoadBalancer::disable() was called
195 public function getConnection( $i, $groups = [], $domain = false, $flags = 0 );
198 * Mark a foreign connection as being available for reuse under a different DB domain
200 * This mechanism is reference-counted, and must be called the same number of times
201 * as getConnection() to work.
203 * @param IDatabase $conn
204 * @throws InvalidArgumentException
206 public function reuseConnection( IDatabase
$conn );
209 * Get a database connection handle reference
211 * The handle's methods simply wrap those of a Database handle
213 * The CONN_TRX_AUTOCOMMIT flag is ignored for databases with ATTR_DB_LEVEL_LOCKING
214 * (e.g. sqlite) in order to avoid deadlocks. ILoadBalancer::getServerAttributes()
215 * can be used to check such flags beforehand.
217 * @see ILoadBalancer::getConnection() for parameter information
219 * @param int $i Server index or DB_MASTER/DB_REPLICA
220 * @param array|string|bool $groups Query group(s), or false for the generic reader
221 * @param string|bool $domain Domain ID, or false for the current domain
222 * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTOCOMMIT)
225 public function getConnectionRef( $i, $groups = [], $domain = false, $flags = 0 );
228 * Get a database connection handle reference without connecting yet
230 * The handle's methods simply wrap those of a Database handle
232 * The CONN_TRX_AUTOCOMMIT flag is ignored for databases with ATTR_DB_LEVEL_LOCKING
233 * (e.g. sqlite) in order to avoid deadlocks. ILoadBalancer::getServerAttributes()
234 * can be used to check such flags beforehand.
236 * @see ILoadBalancer::getConnection() for parameter information
238 * @param int $i Server index or DB_MASTER/DB_REPLICA
239 * @param array|string|bool $groups Query group(s), or false for the generic reader
240 * @param string|bool $domain Domain ID, or false for the current domain
241 * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTOCOMMIT)
244 public function getLazyConnectionRef( $i, $groups = [], $domain = false, $flags = 0 );
247 * Get a maintenance database connection handle reference for migrations and schema changes
249 * The handle's methods simply wrap those of a Database handle
251 * The CONN_TRX_AUTOCOMMIT flag is ignored for databases with ATTR_DB_LEVEL_LOCKING
252 * (e.g. sqlite) in order to avoid deadlocks. ILoadBalancer::getServerAttributes()
253 * can be used to check such flags beforehand.
255 * @see ILoadBalancer::getConnection() for parameter information
257 * @param int $db Server index or DB_MASTER/DB_REPLICA
258 * @param array|string|bool $groups Query group(s), or false for the generic reader
259 * @param string|bool $domain Domain ID, or false for the current domain
260 * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTOCOMMIT)
261 * @return MaintainableDBConnRef
263 public function getMaintenanceConnectionRef( $db, $groups = [], $domain = false, $flags = 0 );
266 * Open a connection to the server given by the specified index
268 * The index must be an actual index into the array. If a connection to the server is
269 * already open and not considered an "in use" foreign connection, this simply returns it.
271 * Avoid using CONN_TRX_AUTOCOMMIT for databases with ATTR_DB_LEVEL_LOCKING (e.g. sqlite) in
272 * order to avoid deadlocks. ILoadBalancer::getServerAttributes() can be used to check
273 * such flags beforehand.
275 * If the caller uses $domain or sets CONN_TRX_AUTOCOMMIT in $flags, then it must also
276 * call ILoadBalancer::reuseConnection() on the handle when finished using it.
277 * In all other cases, this is not necessary, though not harmful either.
279 * @note This method throws DBAccessError if ILoadBalancer::disable() was called
281 * @param int $i Server index (does not support DB_MASTER/DB_REPLICA)
282 * @param string|bool $domain Domain ID, or false for the current domain
283 * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTOCOMMIT)
284 * @return Database|bool Returns false on errors
285 * @throws DBAccessError
287 public function openConnection( $i, $domain = false, $flags = 0 );
292 public function getWriterIndex();
295 * Returns true if the specified index is a valid server index
300 public function haveIndex( $i );
303 * Returns true if the specified index is valid and has non-zero load
308 public function isNonZeroLoad( $i );
311 * Get the number of defined servers (not the number of open connections)
315 public function getServerCount();
318 * Get the host name or IP address of the server with the specified index
321 * @return string Readable name if available or IP/host otherwise
323 public function getServerName( $i );
326 * Get DB type of the server with the specified index
329 * @return string One of (mysql,postgres,sqlite,...) or "unknown" for bad indexes
332 public function getServerType( $i );
335 * @param int $i Server index
336 * @return array (Database::ATTRIBUTE_* constant => value) for all such constants
339 public function getServerAttributes( $i );
342 * Get the current master position for chronology control purposes
343 * @return DBMasterPos|bool Returns false if not applicable
345 public function getMasterPos();
348 * Disable this load balancer. All connections are closed, and any attempt to
349 * open a new connection will result in a DBAccessError.
351 public function disable();
354 * Close all open connections
356 public function closeAll();
361 * Using this function makes sure the LoadBalancer knows the connection is closed.
362 * If you use $conn->close() directly, the load balancer won't update its state.
364 * @param IDatabase $conn
366 public function closeConnection( IDatabase
$conn );
369 * Commit transactions on all open connections
370 * @param string $fname Caller name
371 * @throws DBExpectedError
373 public function commitAll( $fname = __METHOD__
);
376 * Perform all pre-commit callbacks that remain part of the atomic transactions
377 * and disable any post-commit callbacks until runMasterPostTrxCallbacks()
379 * Use this only for mutli-database commits
381 public function finalizeMasterChanges();
384 * Perform all pre-commit checks for things like replication safety
386 * Use this only for mutli-database commits
388 * @param array $options Includes:
389 * - maxWriteDuration : max write query duration time in seconds
390 * @throws DBTransactionError
392 public function approveMasterChanges( array $options );
395 * Flush any master transaction snapshots and set DBO_TRX (if DBO_DEFAULT is set)
397 * The DBO_TRX setting will be reverted to the default in each of these methods:
398 * - commitMasterChanges()
399 * - rollbackMasterChanges()
401 * This allows for custom transaction rounds from any outer transaction scope.
403 * @param string $fname
404 * @throws DBExpectedError
406 public function beginMasterChanges( $fname = __METHOD__
);
409 * Issue COMMIT on all master connections where writes where done
410 * @param string $fname Caller name
411 * @throws DBExpectedError
413 public function commitMasterChanges( $fname = __METHOD__
);
416 * Issue all pending post-COMMIT/ROLLBACK callbacks
418 * Use this only for mutli-database commits
420 * @param int $type IDatabase::TRIGGER_* constant
421 * @return Exception|null The first exception or null if there were none
423 public function runMasterPostTrxCallbacks( $type );
426 * Issue ROLLBACK only on master, only if queries were done on connection
427 * @param string $fname Caller name
428 * @throws DBExpectedError
430 public function rollbackMasterChanges( $fname = __METHOD__
);
433 * Suppress all pending post-COMMIT/ROLLBACK callbacks
435 * Use this only for mutli-database commits
437 * @return Exception|null The first exception or null if there were none
439 public function suppressTransactionEndCallbacks();
442 * Commit all replica DB transactions so as to flush any REPEATABLE-READ or SSI snapshot
444 * @param string $fname Caller name
446 public function flushReplicaSnapshots( $fname = __METHOD__
);
449 * @return bool Whether a master connection is already open
451 public function hasMasterConnection();
454 * Determine if there are pending changes in a transaction by this thread
457 public function hasMasterChanges();
460 * Get the timestamp of the latest write query done by this thread
461 * @return float|bool UNIX timestamp or false
463 public function lastMasterChangeTimestamp();
466 * Check if this load balancer object had any recent or still
467 * pending writes issued against it by this PHP thread
469 * @param float $age How many seconds ago is "recent" [defaults to mWaitTimeout]
472 public function hasOrMadeRecentMasterChanges( $age = null );
475 * Get the list of callers that have pending master changes
477 * @return string[] List of method names
479 public function pendingMasterChangeCallers();
482 * @note This method will trigger a DB connection if not yet done
483 * @param string|bool $domain Domain ID, or false for the current domain
484 * @return bool Whether the database for generic connections this request is highly "lagged"
486 public function getLaggedReplicaMode( $domain = false );
489 * Checks whether the database for generic connections this request was both:
490 * - a) Already choosen due to a prior connection attempt
491 * - b) Considered highly "lagged"
493 * @note This method will never cause a new DB connection
496 public function laggedReplicaUsed();
499 * @note This method may trigger a DB connection if not yet done
500 * @param string|bool $domain Domain ID, or false for the current domain
501 * @param IDatabase|null $conn DB master connection; used to avoid loops [optional]
502 * @return string|bool Reason the master is read-only or false if it is not
504 public function getReadOnlyReason( $domain = false, IDatabase
$conn = null );
507 * Disables/enables lag checks
508 * @param null|bool $mode
511 public function allowLagged( $mode = null );
516 public function pingAll();
519 * Call a function with each open connection object
520 * @param callable $callback
521 * @param array $params
523 public function forEachOpenConnection( $callback, array $params = [] );
526 * Call a function with each open connection object to a master
527 * @param callable $callback
528 * @param array $params
530 public function forEachOpenMasterConnection( $callback, array $params = [] );
533 * Call a function with each open replica DB connection object
534 * @param callable $callback
535 * @param array $params
537 public function forEachOpenReplicaConnection( $callback, array $params = [] );
540 * Get the hostname and lag time of the most-lagged replica DB
542 * This is useful for maintenance scripts that need to throttle their updates.
543 * May attempt to open connections to replica DBs on the default DB. If there is
544 * no lag, the maximum lag will be reported as -1.
546 * @param bool|string $domain Domain ID, or false for the default database
547 * @return array ( host, max lag, index of max lagged host )
549 public function getMaxLag( $domain = false );
552 * Get an estimate of replication lag (in seconds) for each server
554 * Results are cached for a short time in memcached/process cache
556 * Values may be "false" if replication is too broken to estimate
558 * @param string|bool $domain
559 * @return int[] Map of (server index => float|int|bool)
561 public function getLagTimes( $domain = false );
564 * Get the lag in seconds for a given connection, or zero if this load
565 * balancer does not have replication enabled.
567 * This should be used in preference to Database::getLag() in cases where
568 * replication may not be in use, since there is no way to determine if
569 * replication is in use at the connection level without running
570 * potentially restricted queries such as SHOW SLAVE STATUS. Using this
571 * function instead of Database::getLag() avoids a fatal error in this
572 * case on many installations.
574 * @param IDatabase $conn
575 * @return int|bool Returns false on error
577 public function safeGetLag( IDatabase
$conn );
580 * Wait for a replica DB to reach a specified master position
582 * This will connect to the master to get an accurate position if $pos is not given
584 * @param IDatabase $conn Replica DB
585 * @param DBMasterPos|bool $pos Master position; default: current position
586 * @param int $timeout Timeout in seconds [optional]
587 * @return bool Success
589 public function safeWaitForMasterPos( IDatabase
$conn, $pos = false, $timeout = 10 );
592 * Set a callback via IDatabase::setTransactionListener() on
593 * all current and future master connections of this load balancer
595 * @param string $name Callback name
596 * @param callable|null $callback
598 public function setTransactionListener( $name, callable
$callback = null );
601 * Set a new table prefix for the existing local domain ID for testing
603 * @param string $prefix
605 public function setDomainPrefix( $prefix );
608 * Make certain table names use their own database, schema, and table prefix
609 * when passed into SQL queries pre-escaped and without a qualified database name
611 * For example, "user" can be converted to "myschema.mydbname.user" for convenience.
612 * Appearances like `user`, somedb.user, somedb.someschema.user will used literally.
614 * Calling this twice will completely clear any old table aliases. Also, note that
615 * callers are responsible for making sure the schemas and databases actually exist.
617 * @param array[] $aliases Map of (table => (dbname, schema, prefix) map)
619 public function setTableAliases( array $aliases );
622 * Convert certain index names to alternative names before querying the DB
624 * Note that this applies to indexes regardless of the table they belong to.
626 * This can be employed when an index was renamed X => Y in code, but the new Y-named
627 * indexes were not yet built on all DBs. After all the Y-named ones are added by the DBA,
628 * the aliases can be removed, and then the old X-named indexes dropped.
630 * @param string[] $aliases
634 public function setIndexAliases( array $aliases );