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_AUTO = 1;
  89
  90         /** Disable writing for the given connection. Used internally. Do not use with DB_MASTER! */
  91         const CONN_NO_WRITE = 2;
  92
  93         /**
  94          * Construct a manager of IDatabase connection objects
  95          *
  96          * @param array $params Parameter map with keys:
  97          *  - servers : Required. Array of server info structures.
  98          *  - localDomain: A DatabaseDomain or domain ID string.
  99          *  - loadMonitor : Name of a class used to fetch server lag and load.
 100          *  - readOnlyReason : Reason the master DB is read-only if so [optional]
 101          *  - waitTimeout : Maximum time to wait for replicas for consistency [optional]
 102          *  - maxLag: Avoid replica DB servers with more lag than this [optional]
 103          *  - srvCache : BagOStuff object for server cache [optional]
 104          *  - wanCache : WANObjectCache object [optional]
 105          *  - chronologyProtector: ChronologyProtector object [optional]
 106          *  - hostname : The name of the current server [optional]
 107          *  - cliMode: Whether the execution context is a CLI script. [optional]
 108          *  - profiler : Class name or instance with profileIn()/profileOut() methods. [optional]
 109          *  - trxProfiler: TransactionProfiler instance. [optional]
 110          *  - replLogger: PSR-3 logger instance. [optional]
 111          *  - connLogger: PSR-3 logger instance. [optional]
 112          *  - queryLogger: PSR-3 logger instance. [optional]
 113          *  - perfLogger: PSR-3 logger instance. [optional]
 114          *  - errorLogger : Callback that takes an Exception and logs it. [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 by index
 177          *
 178          * Avoid using CONN_TRX_AUTO with sqlite (e.g. check getServerType() first)
 179          *
 180          * @param int $i Server index or DB_MASTER/DB_REPLICA
 181          * @param array|string|bool $groups Query group(s), or false for the generic reader
 182          * @param string|bool $domain Domain ID, or false for the current domain
 183          * @param int $flags Bitfield of CONN_* class constants
 184          *
 185          * @throws DBError
 186          * @return Database
 187          */
 188         public function getConnection( $i, $groups = [], $domain = false, $flags = 0 );
 189
 190         /**
 191          * Mark a foreign connection as being available for reuse under a different DB domain
 192          *
 193          * This mechanism is reference-counted, and must be called the same number of times
 194          * as getConnection() to work.
 195          *
 196          * @param IDatabase $conn
 197          * @throws InvalidArgumentException
 198          */
 199         public function reuseConnection( $conn );
 200
 201         /**
 202          * Get a database connection handle reference
 203          *
 204          * The handle's methods simply wrap those of a Database handle
 205          *
 206          * Avoid using CONN_TRX_AUTO with sqlite (e.g. check getServerType() first)
 207          *
 208          * @see ILoadBalancer::getConnection() for parameter information
 209          *
 210          * @param int $i Server index or DB_MASTER/DB_REPLICA
 211          * @param array|string|bool $groups Query group(s), or false for the generic reader
 212          * @param string|bool $domain Domain ID, or false for the current domain
 213          * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTO)
 214          * @return DBConnRef
 215          */
 216         public function getConnectionRef( $i, $groups = [], $domain = false, $flags = 0 );
 217
 218         /**
 219          * Get a database connection handle reference without connecting yet
 220          *
 221          * The handle's methods simply wrap those of a Database handle
 222          *
 223          * Avoid using CONN_TRX_AUTO with sqlite (e.g. check getServerType() first)
 224          *
 225          * @see ILoadBalancer::getConnection() for parameter information
 226          *
 227          * @param int $i Server index or DB_MASTER/DB_REPLICA
 228          * @param array|string|bool $groups Query group(s), or false for the generic reader
 229          * @param string|bool $domain Domain ID, or false for the current domain
 230          * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTO)
 231          * @return DBConnRef
 232          */
 233         public function getLazyConnectionRef( $i, $groups = [], $domain = false, $flags = 0 );
 234
 235         /**
 236          * Get a maintenance database connection handle reference for migrations and schema changes
 237          *
 238          * The handle's methods simply wrap those of a Database handle
 239          *
 240          * Avoid using CONN_TRX_AUTO with sqlite (e.g. check getServerType() first)
 241          *
 242          * @see ILoadBalancer::getConnection() for parameter information
 243          *
 244          * @param int $db Server index or DB_MASTER/DB_REPLICA
 245          * @param array|string|bool $groups Query group(s), or false for the generic reader
 246          * @param string|bool $domain Domain ID, or false for the current domain
 247          * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTO)
 248          * @return MaintainableDBConnRef
 249          */
 250         public function getMaintenanceConnectionRef( $db, $groups = [], $domain = false, $flags = 0 );
 251
 252         /**
 253          * Open a connection to the server given by the specified index
 254          *
 255          * The index must be an actual index into the array. If a connection to the server is
 256          * already open and not considered an "in use" foreign connection, this simply returns it.
 257          *
 258          * Avoid using CONN_TRX_AUTO with sqlite (e.g. check getServerType() first)
 259          *
 260          * @note If disable() was called on this LoadBalancer, this method will throw a DBAccessError.
 261          *
 262          * @param int $i Server index (does not support DB_MASTER/DB_REPLICA)
 263          * @param string|bool $domain Domain ID, or false for the current domain
 264          * @param int $flags Bitfield of CONN_* class constants (e.g. CONN_TRX_AUTO)
 265          * @return Database|bool Returns false on errors
 266          * @throws DBAccessError
 267          */
 268         public function openConnection( $i, $domain = false, $flags = 0 );
 269
 270         /**
 271          * @return int
 272          */
 273         public function getWriterIndex();
 274
 275         /**
 276          * Returns true if the specified index is a valid server index
 277          *
 278          * @param int $i
 279          * @return bool
 280          */
 281         public function haveIndex( $i );
 282
 283         /**
 284          * Returns true if the specified index is valid and has non-zero load
 285          *
 286          * @param int $i
 287          * @return bool
 288          */
 289         public function isNonZeroLoad( $i );
 290
 291         /**
 292          * Get the number of defined servers (not the number of open connections)
 293          *
 294          * @return int
 295          */
 296         public function getServerCount();
 297
 298         /**
 299          * Get the host name or IP address of the server with the specified index
 300          *
 301          * @param int $i
 302          * @return string Readable name if available or IP/host otherwise
 303          */
 304         public function getServerName( $i );
 305
 306         /**
 307          * Get DB type of the server with the specified index
 308          *
 309          * @param int $i
 310          * @return string One of (mysql,postgres,sqlite,...) or "unknown" for bad indexes
 311          * @since 1.30
 312          */
 313         public function getServerType( $i );
 314
 315         /**
 316          * Return the server info structure for a given index, or false if the index is invalid.
 317          * @param int $i
 318          * @return array|bool
 319          *
 320          * @deprecated Since 1.30, no alternative
 321          */
 322         public function getServerInfo( $i );
 323
 324         /**
 325          * Sets the server info structure for the given index. Entry at index $i
 326          * is created if it doesn't exist
 327          * @param int $i
 328          * @param array $serverInfo
 329          *
 330          * @deprecated Since 1.30, construct new object
 331          */
 332         public function setServerInfo( $i, array $serverInfo );
 333
 334         /**
 335          * Get the current master position for chronology control purposes
 336          * @return DBMasterPos|bool Returns false if not applicable
 337          */
 338         public function getMasterPos();
 339
 340         /**
 341          * Disable this load balancer. All connections are closed, and any attempt to
 342          * open a new connection will result in a DBAccessError.
 343          */
 344         public function disable();
 345
 346         /**
 347          * Close all open connections
 348          */
 349         public function closeAll();
 350
 351         /**
 352          * Close a connection
 353          *
 354          * Using this function makes sure the LoadBalancer knows the connection is closed.
 355          * If you use $conn->close() directly, the load balancer won't update its state.
 356          *
 357          * @param IDatabase $conn
 358          */
 359         public function closeConnection( IDatabase $conn );
 360
 361         /**
 362          * Commit transactions on all open connections
 363          * @param string $fname Caller name
 364          * @throws DBExpectedError
 365          */
 366         public function commitAll( $fname = __METHOD__ );
 367
 368         /**
 369          * Perform all pre-commit callbacks that remain part of the atomic transactions
 370          * and disable any post-commit callbacks until runMasterPostTrxCallbacks()
 371          *
 372          * Use this only for mutli-database commits
 373          */
 374         public function finalizeMasterChanges();
 375
 376         /**
 377          * Perform all pre-commit checks for things like replication safety
 378          *
 379          * Use this only for mutli-database commits
 380          *
 381          * @param array $options Includes:
 382          *   - maxWriteDuration : max write query duration time in seconds
 383          * @throws DBTransactionError
 384          */
 385         public function approveMasterChanges( array $options );
 386
 387         /**
 388          * Flush any master transaction snapshots and set DBO_TRX (if DBO_DEFAULT is set)
 389          *
 390          * The DBO_TRX setting will be reverted to the default in each of these methods:
 391          *   - commitMasterChanges()
 392          *   - rollbackMasterChanges()
 393          *   - commitAll()
 394          * This allows for custom transaction rounds from any outer transaction scope.
 395          *
 396          * @param string $fname
 397          * @throws DBExpectedError
 398          */
 399         public function beginMasterChanges( $fname = __METHOD__ );
 400
 401         /**
 402          * Issue COMMIT on all master connections where writes where done
 403          * @param string $fname Caller name
 404          * @throws DBExpectedError
 405          */
 406         public function commitMasterChanges( $fname = __METHOD__ );
 407
 408         /**
 409          * Issue all pending post-COMMIT/ROLLBACK callbacks
 410          *
 411          * Use this only for mutli-database commits
 412          *
 413          * @param int $type IDatabase::TRIGGER_* constant
 414          * @return Exception|null The first exception or null if there were none
 415          */
 416         public function runMasterPostTrxCallbacks( $type );
 417
 418         /**
 419          * Issue ROLLBACK only on master, only if queries were done on connection
 420          * @param string $fname Caller name
 421          * @throws DBExpectedError
 422          */
 423         public function rollbackMasterChanges( $fname = __METHOD__ );
 424
 425         /**
 426          * Suppress all pending post-COMMIT/ROLLBACK callbacks
 427          *
 428          * Use this only for mutli-database commits
 429          *
 430          * @return Exception|null The first exception or null if there were none
 431          */
 432         public function suppressTransactionEndCallbacks();
 433
 434         /**
 435          * Commit all replica DB transactions so as to flush any REPEATABLE-READ or SSI snapshot
 436          *
 437          * @param string $fname Caller name
 438          */
 439         public function flushReplicaSnapshots( $fname = __METHOD__ );
 440
 441         /**
 442          * @return bool Whether a master connection is already open
 443          */
 444         public function hasMasterConnection();
 445
 446         /**
 447          * Determine if there are pending changes in a transaction by this thread
 448          * @return bool
 449          */
 450         public function hasMasterChanges();
 451
 452         /**
 453          * Get the timestamp of the latest write query done by this thread
 454          * @return float|bool UNIX timestamp or false
 455          */
 456         public function lastMasterChangeTimestamp();
 457
 458         /**
 459          * Check if this load balancer object had any recent or still
 460          * pending writes issued against it by this PHP thread
 461          *
 462          * @param float $age How many seconds ago is "recent" [defaults to mWaitTimeout]
 463          * @return bool
 464          */
 465         public function hasOrMadeRecentMasterChanges( $age = null );
 466
 467         /**
 468          * Get the list of callers that have pending master changes
 469          *
 470          * @return string[] List of method names
 471          */
 472         public function pendingMasterChangeCallers();
 473
 474         /**
 475          * @note This method will trigger a DB connection if not yet done
 476          * @param string|bool $domain Domain ID, or false for the current domain
 477          * @return bool Whether the database for generic connections this request is highly "lagged"
 478          */
 479         public function getLaggedReplicaMode( $domain = false );
 480
 481         /**
 482          * Checks whether the database for generic connections this request was both:
 483          *   - a) Already choosen due to a prior connection attempt
 484          *   - b) Considered highly "lagged"
 485          *
 486          * @note This method will never cause a new DB connection
 487          * @return bool
 488          */
 489         public function laggedReplicaUsed();
 490
 491         /**
 492          * @note This method may trigger a DB connection if not yet done
 493          * @param string|bool $domain Domain ID, or false for the current domain
 494          * @param IDatabase|null $conn DB master connection; used to avoid loops [optional]
 495          * @return string|bool Reason the master is read-only or false if it is not
 496          */
 497         public function getReadOnlyReason( $domain = false, IDatabase $conn = null );
 498
 499         /**
 500          * Disables/enables lag checks
 501          * @param null|bool $mode
 502          * @return bool
 503          */
 504         public function allowLagged( $mode = null );
 505
 506         /**
 507          * @return bool
 508          */
 509         public function pingAll();
 510
 511         /**
 512          * Call a function with each open connection object
 513          * @param callable $callback
 514          * @param array $params
 515          */
 516         public function forEachOpenConnection( $callback, array $params = [] );
 517
 518         /**
 519          * Call a function with each open connection object to a master
 520          * @param callable $callback
 521          * @param array $params
 522          */
 523         public function forEachOpenMasterConnection( $callback, array $params = [] );
 524
 525         /**
 526          * Call a function with each open replica DB connection object
 527          * @param callable $callback
 528          * @param array $params
 529          */
 530         public function forEachOpenReplicaConnection( $callback, array $params = [] );
 531
 532         /**
 533          * Get the hostname and lag time of the most-lagged replica DB
 534          *
 535          * This is useful for maintenance scripts that need to throttle their updates.
 536          * May attempt to open connections to replica DBs on the default DB. If there is
 537          * no lag, the maximum lag will be reported as -1.
 538          *
 539          * @param bool|string $domain Domain ID, or false for the default database
 540          * @return array ( host, max lag, index of max lagged host )
 541          */
 542         public function getMaxLag( $domain = false );
 543
 544         /**
 545          * Get an estimate of replication lag (in seconds) for each server
 546          *
 547          * Results are cached for a short time in memcached/process cache
 548          *
 549          * Values may be "false" if replication is too broken to estimate
 550          *
 551          * @param string|bool $domain
 552          * @return int[] Map of (server index => float|int|bool)
 553          */
 554         public function getLagTimes( $domain = false );
 555
 556         /**
 557          * Get the lag in seconds for a given connection, or zero if this load
 558          * balancer does not have replication enabled.
 559          *
 560          * This should be used in preference to Database::getLag() in cases where
 561          * replication may not be in use, since there is no way to determine if
 562          * replication is in use at the connection level without running
 563          * potentially restricted queries such as SHOW SLAVE STATUS. Using this
 564          * function instead of Database::getLag() avoids a fatal error in this
 565          * case on many installations.
 566          *
 567          * @param IDatabase $conn
 568          * @return int|bool Returns false on error
 569          */
 570         public function safeGetLag( IDatabase $conn );
 571
 572         /**
 573          * Wait for a replica DB to reach a specified master position
 574          *
 575          * This will connect to the master to get an accurate position if $pos is not given
 576          *
 577          * @param IDatabase $conn Replica DB
 578          * @param DBMasterPos|bool $pos Master position; default: current position
 579          * @param int $timeout Timeout in seconds [optional]
 580          * @return bool Success
 581          */
 582         public function safeWaitForMasterPos( IDatabase $conn, $pos = false, $timeout = 10 );
 583
 584         /**
 585          * Set a callback via IDatabase::setTransactionListener() on
 586          * all current and future master connections of this load balancer
 587          *
 588          * @param string $name Callback name
 589          * @param callable|null $callback
 590          */
 591         public function setTransactionListener( $name, callable $callback = null );
 592
 593         /**
 594          * Set a new table prefix for the existing local domain ID for testing
 595          *
 596          * @param string $prefix
 597          */
 598         public function setDomainPrefix( $prefix );
 599
 600         /**
 601          * Make certain table names use their own database, schema, and table prefix
 602          * when passed into SQL queries pre-escaped and without a qualified database name
 603          *
 604          * For example, "user" can be converted to "myschema.mydbname.user" for convenience.
 605          * Appearances like `user`, somedb.user, somedb.someschema.user will used literally.
 606          *
 607          * Calling this twice will completely clear any old table aliases. Also, note that
 608          * callers are responsible for making sure the schemas and databases actually exist.
 609          *
 610          * @param array[] $aliases Map of (table => (dbname, schema, prefix) map)
 611          */
 612         public function setTableAliases( array $aliases );
 613 }