Merge "rdbms: rename CONN_TRX_AUTO constant to CONN_TRX_AUTOCOMMIT"
[lhc/web/wiklou.git] / includes / libs / rdbms / loadbalancer / ILoadBalancer.php
1 <?php
2 /**
3 * Database load balancing interface
4 *
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.
9 *
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.
14 *
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
19 *
20 * @file
21 * @ingroup Database
22 */
23 namespace Wikimedia\Rdbms;
24
25 use Exception;
26 use InvalidArgumentException;
27
28 /**
29 * Database cluster connection, tracking, load balancing, and transaction manager interface
30 *
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.
37 *
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
45 * the PHP request.
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.
52 *
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).
69 *
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.
74 *
75 * @since 1.28
76 * @ingroup Database
77 */
78 interface ILoadBalancer {
79 /** @var int Request a replica DB connection */
80 const DB_REPLICA = -1;
81 /** @var int Request a master DB connection */
82 const DB_MASTER = -2;
83
84 /** @var string Domain specifier when no specific database needs to be selected */
85 const DOMAIN_ANY = '';
86
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;
91
92 /**
93 * Construct a manager of IDatabase connection objects
94 *
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 * - deprecationLogger: Callback to log a deprecation warning. [optional]
115 * @throws InvalidArgumentException
116 */
117 public function __construct( array $params );
118
119 /**
120 * Get the index of the reader connection, which may be a replica DB
121 *
122 * This takes into account load ratios and lag times. It should
123 * always return a consistent index during a given invocation.
124 *
125 * Side effect: opens connections to databases
126 * @param string|bool $group Query group, or false for the generic reader
127 * @param string|bool $domain Domain ID, or false for the current domain
128 * @throws DBError
129 * @return bool|int|string
130 */
131 public function getReaderIndex( $group = false, $domain = false );
132
133 /**
134 * Set the master wait position
135 *
136 * If a DB_REPLICA connection has been opened already, then wait immediately.
137 * Otherwise sets a variable telling it to wait if such a connection is opened.
138 *
139 * This only applies to connections to the generic replica DB for this request.
140 * If a timeout happens when waiting, then getLaggedReplicaMode()/laggedReplicaUsed()
141 * will return true.
142 *
143 * @param DBMasterPos|bool $pos Master position or false
144 */
145 public function waitFor( $pos );
146
147 /**
148 * Set the master wait position and wait for a "generic" replica DB to catch up to it
149 *
150 * This can be used a faster proxy for waitForAll()
151 *
152 * @param DBMasterPos|bool $pos Master position or false
153 * @param int $timeout Max seconds to wait; default is mWaitTimeout
154 * @return bool Success (able to connect and no timeouts reached)
155 */
156 public function waitForOne( $pos, $timeout = null );
157
158 /**
159 * Set the master wait position and wait for ALL replica DBs to catch up to it
160 *
161 * @param DBMasterPos|bool $pos Master position or false
162 * @param int $timeout Max seconds to wait; default is mWaitTimeout
163 * @return bool Success (able to connect and no timeouts reached)
164 */
165 public function waitForAll( $pos, $timeout = null );
166
167 /**
168 * Get any open connection to a given server index, local or foreign
169 *
170 * @param int $i Server index or DB_MASTER/DB_REPLICA
171 * @return Database|bool False if no such connection is open
172 */
173 public function getAnyOpenConnection( $i );
174
175 /**
176 * Get a connection handle by server index
177 *
178 * The CONN_TRX_AUTOCOMMIT flag is ignored for databases with ATTR_DB_LEVEL_LOCKING
179 * (e.g. sqlite) in order to avoid deadlocks. ILoadBalancer::getServerAttributes()
180 * can be used to check such flags beforehand.
181 *
182 * If the caller uses $domain or sets CONN_TRX_AUTOCOMMIT in $flags, then it must also
183 * call ILoadBalancer::reuseConnection() on the handle when finished using it.
184 * In all other cases, this is not necessary, though not harmful either.
185 *
186 * @param int $i Server index or DB_MASTER/DB_REPLICA
187 * @param array|string|bool $groups Query group(s), or false for the generic reader
188 * @param string|bool $domain Domain ID, or false for the current domain
189 * @param int $flags Bitfield of CONN_* class constants
190 *
191 * @note This method throws DBAccessError if ILoadBalancer::disable() was called
192 *
193 * @throws DBError
194 * @return Database
195 */
196 public function getConnection( $i, $groups = [], $domain = false, $flags = 0 );
197
198 /**
199 * Mark a foreign connection as being available for reuse under a different DB domain
200 *
201 * This mechanism is reference-counted, and must be called the same number of times
202 * as getConnection() to work.
203 *
204 * @param IDatabase $conn
205 * @throws InvalidArgumentException
206 */
207 public function reuseConnection( IDatabase $conn );
208
209 /**
210 * Get a database connection handle reference
211 *
212 * The handle's methods simply wrap those of a Database handle
213 *
214 * The CONN_TRX_AUTOCOMMIT flag is ignored for databases with ATTR_DB_LEVEL_LOCKING
215 * (e.g. sqlite) in order to avoid deadlocks. ILoadBalancer::getServerAttributes()
216 * can be used to check such flags beforehand.
217 *
218 * @see ILoadBalancer::getConnection() for parameter information
219 *
220 * @param int $i Server index or DB_MASTER/DB_REPLICA
221 * @param array|string|bool $groups Query group(s), or false for the generic reader
222 * @param string|bool $domain Domain ID, or false for the current domain
223 * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTOCOMMIT)
224 * @return DBConnRef
225 */
226 public function getConnectionRef( $i, $groups = [], $domain = false, $flags = 0 );
227
228 /**
229 * Get a database connection handle reference without connecting yet
230 *
231 * The handle's methods simply wrap those of a Database handle
232 *
233 * The CONN_TRX_AUTOCOMMIT flag is ignored for databases with ATTR_DB_LEVEL_LOCKING
234 * (e.g. sqlite) in order to avoid deadlocks. ILoadBalancer::getServerAttributes()
235 * can be used to check such flags beforehand.
236 *
237 * @see ILoadBalancer::getConnection() for parameter information
238 *
239 * @param int $i Server index or DB_MASTER/DB_REPLICA
240 * @param array|string|bool $groups Query group(s), or false for the generic reader
241 * @param string|bool $domain Domain ID, or false for the current domain
242 * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTOCOMMIT)
243 * @return DBConnRef
244 */
245 public function getLazyConnectionRef( $i, $groups = [], $domain = false, $flags = 0 );
246
247 /**
248 * Get a maintenance database connection handle reference for migrations and schema changes
249 *
250 * The handle's methods simply wrap those of a Database handle
251 *
252 * The CONN_TRX_AUTOCOMMIT flag is ignored for databases with ATTR_DB_LEVEL_LOCKING
253 * (e.g. sqlite) in order to avoid deadlocks. ILoadBalancer::getServerAttributes()
254 * can be used to check such flags beforehand.
255 *
256 * @see ILoadBalancer::getConnection() for parameter information
257 *
258 * @param int $db Server index or DB_MASTER/DB_REPLICA
259 * @param array|string|bool $groups Query group(s), or false for the generic reader
260 * @param string|bool $domain Domain ID, or false for the current domain
261 * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTOCOMMIT)
262 * @return MaintainableDBConnRef
263 */
264 public function getMaintenanceConnectionRef( $db, $groups = [], $domain = false, $flags = 0 );
265
266 /**
267 * Open a connection to the server given by the specified index
268 *
269 * The index must be an actual index into the array. If a connection to the server is
270 * already open and not considered an "in use" foreign connection, this simply returns it.
271 *
272 * Avoid using CONN_TRX_AUTOCOMMIT for databases with ATTR_DB_LEVEL_LOCKING (e.g. sqlite) in
273 * order to avoid deadlocks. ILoadBalancer::getServerAttributes() can be used to check
274 * such flags beforehand.
275 *
276 * If the caller uses $domain or sets CONN_TRX_AUTOCOMMIT in $flags, then it must also
277 * call ILoadBalancer::reuseConnection() on the handle when finished using it.
278 * In all other cases, this is not necessary, though not harmful either.
279 *
280 * @note This method throws DBAccessError if ILoadBalancer::disable() was called
281 *
282 * @param int $i Server index (does not support DB_MASTER/DB_REPLICA)
283 * @param string|bool $domain Domain ID, or false for the current domain
284 * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTOCOMMIT)
285 * @return Database|bool Returns false on errors
286 * @throws DBAccessError
287 */
288 public function openConnection( $i, $domain = false, $flags = 0 );
289
290 /**
291 * @return int
292 */
293 public function getWriterIndex();
294
295 /**
296 * Returns true if the specified index is a valid server index
297 *
298 * @param int $i
299 * @return bool
300 */
301 public function haveIndex( $i );
302
303 /**
304 * Returns true if the specified index is valid and has non-zero load
305 *
306 * @param int $i
307 * @return bool
308 */
309 public function isNonZeroLoad( $i );
310
311 /**
312 * Get the number of defined servers (not the number of open connections)
313 *
314 * @return int
315 */
316 public function getServerCount();
317
318 /**
319 * Get the host name or IP address of the server with the specified index
320 *
321 * @param int $i
322 * @return string Readable name if available or IP/host otherwise
323 */
324 public function getServerName( $i );
325
326 /**
327 * Get DB type of the server with the specified index
328 *
329 * @param int $i
330 * @return string One of (mysql,postgres,sqlite,...) or "unknown" for bad indexes
331 * @since 1.30
332 */
333 public function getServerType( $i );
334
335 /**
336 * @param int $i Server index
337 * @return array (Database::ATTRIBUTE_* constant => value) for all such constants
338 * @since 1.31
339 */
340 public function getServerAttributes( $i );
341
342 /**
343 * Get the current master position for chronology control purposes
344 * @return DBMasterPos|bool Returns false if not applicable
345 */
346 public function getMasterPos();
347
348 /**
349 * Disable this load balancer. All connections are closed, and any attempt to
350 * open a new connection will result in a DBAccessError.
351 */
352 public function disable();
353
354 /**
355 * Close all open connections
356 */
357 public function closeAll();
358
359 /**
360 * Close a connection
361 *
362 * Using this function makes sure the LoadBalancer knows the connection is closed.
363 * If you use $conn->close() directly, the load balancer won't update its state.
364 *
365 * @param IDatabase $conn
366 */
367 public function closeConnection( IDatabase $conn );
368
369 /**
370 * Commit transactions on all open connections
371 * @param string $fname Caller name
372 * @throws DBExpectedError
373 */
374 public function commitAll( $fname = __METHOD__ );
375
376 /**
377 * Perform all pre-commit callbacks that remain part of the atomic transactions
378 * and disable any post-commit callbacks until runMasterPostTrxCallbacks()
379 *
380 * Use this only for mutli-database commits
381 */
382 public function finalizeMasterChanges();
383
384 /**
385 * Perform all pre-commit checks for things like replication safety
386 *
387 * Use this only for mutli-database commits
388 *
389 * @param array $options Includes:
390 * - maxWriteDuration : max write query duration time in seconds
391 * @throws DBTransactionError
392 */
393 public function approveMasterChanges( array $options );
394
395 /**
396 * Flush any master transaction snapshots and set DBO_TRX (if DBO_DEFAULT is set)
397 *
398 * The DBO_TRX setting will be reverted to the default in each of these methods:
399 * - commitMasterChanges()
400 * - rollbackMasterChanges()
401 * - commitAll()
402 * This allows for custom transaction rounds from any outer transaction scope.
403 *
404 * @param string $fname
405 * @throws DBExpectedError
406 */
407 public function beginMasterChanges( $fname = __METHOD__ );
408
409 /**
410 * Issue COMMIT on all master connections where writes where done
411 * @param string $fname Caller name
412 * @throws DBExpectedError
413 */
414 public function commitMasterChanges( $fname = __METHOD__ );
415
416 /**
417 * Issue all pending post-COMMIT/ROLLBACK callbacks
418 *
419 * Use this only for mutli-database commits
420 *
421 * @param int $type IDatabase::TRIGGER_* constant
422 * @return Exception|null The first exception or null if there were none
423 */
424 public function runMasterPostTrxCallbacks( $type );
425
426 /**
427 * Issue ROLLBACK only on master, only if queries were done on connection
428 * @param string $fname Caller name
429 * @throws DBExpectedError
430 */
431 public function rollbackMasterChanges( $fname = __METHOD__ );
432
433 /**
434 * Suppress all pending post-COMMIT/ROLLBACK callbacks
435 *
436 * Use this only for mutli-database commits
437 *
438 * @return Exception|null The first exception or null if there were none
439 */
440 public function suppressTransactionEndCallbacks();
441
442 /**
443 * Commit all replica DB transactions so as to flush any REPEATABLE-READ or SSI snapshot
444 *
445 * @param string $fname Caller name
446 */
447 public function flushReplicaSnapshots( $fname = __METHOD__ );
448
449 /**
450 * @return bool Whether a master connection is already open
451 */
452 public function hasMasterConnection();
453
454 /**
455 * Determine if there are pending changes in a transaction by this thread
456 * @return bool
457 */
458 public function hasMasterChanges();
459
460 /**
461 * Get the timestamp of the latest write query done by this thread
462 * @return float|bool UNIX timestamp or false
463 */
464 public function lastMasterChangeTimestamp();
465
466 /**
467 * Check if this load balancer object had any recent or still
468 * pending writes issued against it by this PHP thread
469 *
470 * @param float $age How many seconds ago is "recent" [defaults to mWaitTimeout]
471 * @return bool
472 */
473 public function hasOrMadeRecentMasterChanges( $age = null );
474
475 /**
476 * Get the list of callers that have pending master changes
477 *
478 * @return string[] List of method names
479 */
480 public function pendingMasterChangeCallers();
481
482 /**
483 * @note This method will trigger a DB connection if not yet done
484 * @param string|bool $domain Domain ID, or false for the current domain
485 * @return bool Whether the database for generic connections this request is highly "lagged"
486 */
487 public function getLaggedReplicaMode( $domain = false );
488
489 /**
490 * Checks whether the database for generic connections this request was both:
491 * - a) Already choosen due to a prior connection attempt
492 * - b) Considered highly "lagged"
493 *
494 * @note This method will never cause a new DB connection
495 * @return bool
496 */
497 public function laggedReplicaUsed();
498
499 /**
500 * @note This method may trigger a DB connection if not yet done
501 * @param string|bool $domain Domain ID, or false for the current domain
502 * @param IDatabase|null $conn DB master connection; used to avoid loops [optional]
503 * @return string|bool Reason the master is read-only or false if it is not
504 */
505 public function getReadOnlyReason( $domain = false, IDatabase $conn = null );
506
507 /**
508 * Disables/enables lag checks
509 * @param null|bool $mode
510 * @return bool
511 */
512 public function allowLagged( $mode = null );
513
514 /**
515 * @return bool
516 */
517 public function pingAll();
518
519 /**
520 * Call a function with each open connection object
521 * @param callable $callback
522 * @param array $params
523 */
524 public function forEachOpenConnection( $callback, array $params = [] );
525
526 /**
527 * Call a function with each open connection object to a master
528 * @param callable $callback
529 * @param array $params
530 */
531 public function forEachOpenMasterConnection( $callback, array $params = [] );
532
533 /**
534 * Call a function with each open replica DB connection object
535 * @param callable $callback
536 * @param array $params
537 */
538 public function forEachOpenReplicaConnection( $callback, array $params = [] );
539
540 /**
541 * Get the hostname and lag time of the most-lagged replica DB
542 *
543 * This is useful for maintenance scripts that need to throttle their updates.
544 * May attempt to open connections to replica DBs on the default DB. If there is
545 * no lag, the maximum lag will be reported as -1.
546 *
547 * @param bool|string $domain Domain ID, or false for the default database
548 * @return array ( host, max lag, index of max lagged host )
549 */
550 public function getMaxLag( $domain = false );
551
552 /**
553 * Get an estimate of replication lag (in seconds) for each server
554 *
555 * Results are cached for a short time in memcached/process cache
556 *
557 * Values may be "false" if replication is too broken to estimate
558 *
559 * @param string|bool $domain
560 * @return int[] Map of (server index => float|int|bool)
561 */
562 public function getLagTimes( $domain = false );
563
564 /**
565 * Get the lag in seconds for a given connection, or zero if this load
566 * balancer does not have replication enabled.
567 *
568 * This should be used in preference to Database::getLag() in cases where
569 * replication may not be in use, since there is no way to determine if
570 * replication is in use at the connection level without running
571 * potentially restricted queries such as SHOW SLAVE STATUS. Using this
572 * function instead of Database::getLag() avoids a fatal error in this
573 * case on many installations.
574 *
575 * @param IDatabase $conn
576 * @return int|bool Returns false on error
577 */
578 public function safeGetLag( IDatabase $conn );
579
580 /**
581 * Wait for a replica DB to reach a specified master position
582 *
583 * This will connect to the master to get an accurate position if $pos is not given
584 *
585 * @param IDatabase $conn Replica DB
586 * @param DBMasterPos|bool $pos Master position; default: current position
587 * @param int $timeout Timeout in seconds [optional]
588 * @return bool Success
589 */
590 public function safeWaitForMasterPos( IDatabase $conn, $pos = false, $timeout = 10 );
591
592 /**
593 * Set a callback via IDatabase::setTransactionListener() on
594 * all current and future master connections of this load balancer
595 *
596 * @param string $name Callback name
597 * @param callable|null $callback
598 */
599 public function setTransactionListener( $name, callable $callback = null );
600
601 /**
602 * Set a new table prefix for the existing local domain ID for testing
603 *
604 * @param string $prefix
605 */
606 public function setDomainPrefix( $prefix );
607
608 /**
609 * Make certain table names use their own database, schema, and table prefix
610 * when passed into SQL queries pre-escaped and without a qualified database name
611 *
612 * For example, "user" can be converted to "myschema.mydbname.user" for convenience.
613 * Appearances like `user`, somedb.user, somedb.someschema.user will used literally.
614 *
615 * Calling this twice will completely clear any old table aliases. Also, note that
616 * callers are responsible for making sure the schemas and databases actually exist.
617 *
618 * @param array[] $aliases Map of (table => (dbname, schema, prefix) map)
619 */
620 public function setTableAliases( array $aliases );
621
622 /**
623 * Convert certain index names to alternative names before querying the DB
624 *
625 * Note that this applies to indexes regardless of the table they belong to.
626 *
627 * This can be employed when an index was renamed X => Y in code, but the new Y-named
628 * indexes were not yet built on all DBs. After all the Y-named ones are added by the DBA,
629 * the aliases can be removed, and then the old X-named indexes dropped.
630 *
631 * @param string[] $aliases
632 * @return mixed
633 * @since 1.31
634 */
635 public function setIndexAliases( array $aliases );
636 }