Line data Source code
1 : /*-------------------------------------------------------------------------
2 : *
3 : * syncrep.c
4 : *
5 : * Synchronous replication is new as of PostgreSQL 9.1.
6 : *
7 : * If requested, transaction commits wait until their commit LSN are
8 : * acknowledged by the synchronous standbys.
9 : *
10 : * This module contains the code for waiting and release of backends.
11 : * All code in this module executes on the primary. The core streaming
12 : * replication transport remains within WALreceiver/WALsender modules.
13 : *
14 : * The essence of this design is that it isolates all logic about
15 : * waiting/releasing onto the primary. The primary defines which standbys
16 : * it wishes to wait for. The standbys are completely unaware of the
17 : * durability requirements of transactions on the primary, reducing the
18 : * complexity of the code and streamlining both standby operations and
19 : * network bandwidth because there is no requirement to ship
20 : * per-transaction state information.
21 : *
22 : * Replication is either synchronous or not synchronous (async). If it is
23 : * async, we just fastpath out of here. If it is sync, then we wait for
24 : * the write, flush or apply location on the standby before releasing
25 : * the waiting backend. Further complexity in that interaction is
26 : * expected in later releases.
27 : *
28 : * The best performing way to manage the waiting backends is to have a
29 : * single ordered queue of waiting backends, so that we can avoid
30 : * searching the through all waiters each time we receive a reply.
31 : *
32 : * In 9.5 or before only a single standby could be considered as
33 : * synchronous. In 9.6 we support a priority-based multiple synchronous
34 : * standbys. In 10.0 a quorum-based multiple synchronous standbys is also
35 : * supported. The number of synchronous standbys that transactions
36 : * must wait for replies from is specified in synchronous_standby_names.
37 : * This parameter also specifies a list of standby names and the method
38 : * (FIRST and ANY) to choose synchronous standbys from the listed ones.
39 : *
40 : * The method FIRST specifies a priority-based synchronous replication
41 : * and makes transaction commits wait until their WAL records are
42 : * replicated to the requested number of synchronous standbys chosen based
43 : * on their priorities. The standbys whose names appear earlier in the list
44 : * are given higher priority and will be considered as synchronous.
45 : * Other standby servers appearing later in this list represent potential
46 : * synchronous standbys. If any of the current synchronous standbys
47 : * disconnects for whatever reason, it will be replaced immediately with
48 : * the next-highest-priority standby.
49 : *
50 : * The method ANY specifies a quorum-based synchronous replication
51 : * and makes transaction commits wait until their WAL records are
52 : * replicated to at least the requested number of synchronous standbys
53 : * in the list. All the standbys appearing in the list are considered as
54 : * candidates for quorum synchronous standbys.
55 : *
56 : * If neither FIRST nor ANY is specified, FIRST is used as the method.
57 : * This is for backward compatibility with 9.6 or before where only a
58 : * priority-based sync replication was supported.
59 : *
60 : * Before the standbys chosen from synchronous_standby_names can
61 : * become the synchronous standbys they must have caught up with
62 : * the primary; that may take some time. Once caught up,
63 : * the standbys which are considered as synchronous at that moment
64 : * will release waiters from the queue.
65 : *
66 : * Portions Copyright (c) 2010-2026, PostgreSQL Global Development Group
67 : *
68 : * IDENTIFICATION
69 : * src/backend/replication/syncrep.c
70 : *
71 : *-------------------------------------------------------------------------
72 : */
73 : #include "postgres.h"
74 :
75 : #include <unistd.h>
76 :
77 : #include "access/xact.h"
78 : #include "common/int.h"
79 : #include "miscadmin.h"
80 : #include "pgstat.h"
81 : #include "replication/syncrep.h"
82 : #include "replication/walsender.h"
83 : #include "replication/walsender_private.h"
84 : #include "storage/proc.h"
85 : #include "tcop/tcopprot.h"
86 : #include "utils/guc_hooks.h"
87 : #include "utils/ps_status.h"
88 :
89 : /* User-settable parameters for sync rep */
90 : char *SyncRepStandbyNames;
91 :
92 : #define SyncStandbysDefined() \
93 : (SyncRepStandbyNames != NULL && SyncRepStandbyNames[0] != '\0')
94 :
95 : static bool announce_next_takeover = true;
96 :
97 : SyncRepConfigData *SyncRepConfig = NULL;
98 : static int SyncRepWaitMode = SYNC_REP_NO_WAIT;
99 :
100 : static void SyncRepQueueInsert(int mode);
101 : static void SyncRepCancelWait(void);
102 : static int SyncRepWakeQueue(bool all, int mode);
103 :
104 : static bool SyncRepGetSyncRecPtr(XLogRecPtr *writePtr,
105 : XLogRecPtr *flushPtr,
106 : XLogRecPtr *applyPtr,
107 : bool *am_sync);
108 : static void SyncRepGetOldestSyncRecPtr(XLogRecPtr *writePtr,
109 : XLogRecPtr *flushPtr,
110 : XLogRecPtr *applyPtr,
111 : SyncRepStandbyData *sync_standbys,
112 : int num_standbys);
113 : static void SyncRepGetNthLatestSyncRecPtr(XLogRecPtr *writePtr,
114 : XLogRecPtr *flushPtr,
115 : XLogRecPtr *applyPtr,
116 : SyncRepStandbyData *sync_standbys,
117 : int num_standbys,
118 : uint8 nth);
119 : static int SyncRepGetStandbyPriority(void);
120 : static int standby_priority_comparator(const void *a, const void *b);
121 : static int cmp_lsn(const void *a, const void *b);
122 :
123 : #ifdef USE_ASSERT_CHECKING
124 : static bool SyncRepQueueIsOrderedByLSN(int mode);
125 : #endif
126 :
127 : /*
128 : * ===========================================================
129 : * Synchronous Replication functions for normal user backends
130 : * ===========================================================
131 : */
132 :
133 : /*
134 : * Wait for synchronous replication, if requested by user.
135 : *
136 : * Initially backends start in state SYNC_REP_NOT_WAITING and then
137 : * change that state to SYNC_REP_WAITING before adding ourselves
138 : * to the wait queue. During SyncRepWakeQueue() a WALSender changes
139 : * the state to SYNC_REP_WAIT_COMPLETE once replication is confirmed.
140 : * This backend then resets its state to SYNC_REP_NOT_WAITING.
141 : *
142 : * 'lsn' represents the LSN to wait for. 'commit' indicates whether this LSN
143 : * represents a commit record. If it doesn't, then we wait only for the WAL
144 : * to be flushed if synchronous_commit is set to the higher level of
145 : * remote_apply, because only commit records provide apply feedback.
146 : */
147 : void
148 127775 : SyncRepWaitForLSN(XLogRecPtr lsn, bool commit)
149 : {
150 : int mode;
151 :
152 : /*
153 : * This should be called while holding interrupts during a transaction
154 : * commit to prevent the follow-up shared memory queue cleanups to be
155 : * influenced by external interruptions.
156 : */
157 : Assert(InterruptHoldoffCount > 0);
158 :
159 : /*
160 : * Fast exit if user has not requested sync replication, or there are no
161 : * sync replication standby names defined.
162 : *
163 : * Since this routine gets called every commit time, it's important to
164 : * exit quickly if sync replication is not requested.
165 : *
166 : * We check WalSndCtl->sync_standbys_status flag without the lock and exit
167 : * immediately if SYNC_STANDBY_INIT is set (the checkpointer has
168 : * initialized this data) but SYNC_STANDBY_DEFINED is missing (no sync
169 : * replication requested).
170 : *
171 : * If SYNC_STANDBY_DEFINED is set, we need to check the status again later
172 : * while holding the lock, to check the flag and operate the sync rep
173 : * queue atomically. This is necessary to avoid the race condition
174 : * described in SyncRepUpdateSyncStandbysDefined(). On the other hand, if
175 : * SYNC_STANDBY_DEFINED is not set, the lock is not necessary because we
176 : * don't touch the queue.
177 : */
178 127775 : if (!SyncRepRequested() ||
179 95822 : ((((volatile WalSndCtlData *) WalSndCtl)->sync_standbys_status) &
180 : (SYNC_STANDBY_INIT | SYNC_STANDBY_DEFINED)) == SYNC_STANDBY_INIT)
181 92011 : return;
182 :
183 : /* Cap the level for anything other than commit to remote flush only. */
184 35764 : if (commit)
185 35746 : mode = SyncRepWaitMode;
186 : else
187 18 : mode = Min(SyncRepWaitMode, SYNC_REP_WAIT_FLUSH);
188 :
189 : Assert(dlist_node_is_detached(&MyProc->syncRepLinks));
190 : Assert(WalSndCtl != NULL);
191 :
192 35764 : LWLockAcquire(SyncRepLock, LW_EXCLUSIVE);
193 : Assert(MyProc->syncRepState == SYNC_REP_NOT_WAITING);
194 :
195 : /*
196 : * We don't wait for sync rep if SYNC_STANDBY_DEFINED is not set. See
197 : * SyncRepUpdateSyncStandbysDefined().
198 : *
199 : * Also check that the standby hasn't already replied. Unlikely race
200 : * condition but we'll be fetching that cache line anyway so it's likely
201 : * to be a low cost check.
202 : *
203 : * If the sync standby data has not been initialized yet
204 : * (SYNC_STANDBY_INIT is not set), fall back to a check based on the LSN,
205 : * then do a direct GUC check.
206 : */
207 35764 : if (WalSndCtl->sync_standbys_status & SYNC_STANDBY_INIT)
208 : {
209 40 : if ((WalSndCtl->sync_standbys_status & SYNC_STANDBY_DEFINED) == 0 ||
210 40 : lsn <= WalSndCtl->lsn[mode])
211 : {
212 6 : LWLockRelease(SyncRepLock);
213 6 : return;
214 : }
215 : }
216 35724 : else if (lsn <= WalSndCtl->lsn[mode])
217 : {
218 : /*
219 : * The LSN is older than what we need to wait for. The sync standby
220 : * data has not been initialized yet, but we are OK to not wait
221 : * because we know that there is no point in doing so based on the
222 : * LSN.
223 : */
224 0 : LWLockRelease(SyncRepLock);
225 0 : return;
226 : }
227 35724 : else if (!SyncStandbysDefined())
228 : {
229 : /*
230 : * If we are here, the sync standby data has not been initialized yet,
231 : * and the LSN is newer than what need to wait for, so we have fallen
232 : * back to the best thing we could do in this case: a check on
233 : * SyncStandbysDefined() to see if the GUC is set or not.
234 : *
235 : * When the GUC has a value, we wait until the checkpointer updates
236 : * the status data because we cannot be sure yet if we should wait or
237 : * not. Here, the GUC has *no* value, we are sure that there is no
238 : * point to wait; this matters for example when initializing a
239 : * cluster, where we should never wait, and no sync standbys is the
240 : * default behavior.
241 : */
242 35724 : LWLockRelease(SyncRepLock);
243 35724 : return;
244 : }
245 :
246 : /*
247 : * Set our waitLSN so WALSender will know when to wake us, and add
248 : * ourselves to the queue.
249 : */
250 34 : MyProc->waitLSN = lsn;
251 34 : MyProc->syncRepState = SYNC_REP_WAITING;
252 34 : SyncRepQueueInsert(mode);
253 : Assert(SyncRepQueueIsOrderedByLSN(mode));
254 34 : LWLockRelease(SyncRepLock);
255 :
256 : /* Alter ps display to show waiting for sync rep. */
257 34 : if (update_process_title)
258 : {
259 : char buffer[32];
260 :
261 34 : sprintf(buffer, "waiting for %X/%08X", LSN_FORMAT_ARGS(lsn));
262 34 : set_ps_display_suffix(buffer);
263 : }
264 :
265 : /*
266 : * Wait for specified LSN to be confirmed.
267 : *
268 : * Each proc has its own wait latch, so we perform a normal latch
269 : * check/wait loop here.
270 : */
271 : for (;;)
272 34 : {
273 : int rc;
274 :
275 : /* Must reset the latch before testing state. */
276 68 : ResetLatch(MyLatch);
277 :
278 : /*
279 : * Acquiring the lock is not needed, the latch ensures proper
280 : * barriers. If it looks like we're done, we must really be done,
281 : * because once walsender changes the state to SYNC_REP_WAIT_COMPLETE,
282 : * it will never update it again, so we can't be seeing a stale value
283 : * in that case.
284 : */
285 68 : if (MyProc->syncRepState == SYNC_REP_WAIT_COMPLETE)
286 34 : break;
287 :
288 : /*
289 : * If a wait for synchronous replication is pending, we can neither
290 : * acknowledge the commit nor raise ERROR or FATAL. The latter would
291 : * lead the client to believe that the transaction aborted, which is
292 : * not true: it's already committed locally. The former is no good
293 : * either: the client has requested synchronous replication, and is
294 : * entitled to assume that an acknowledged commit is also replicated,
295 : * which might not be true. So in this case we issue a WARNING (which
296 : * some clients may be able to interpret) and shut off further output.
297 : * We do NOT reset ProcDiePending, so that the process will die after
298 : * the commit is cleaned up.
299 : */
300 34 : if (ProcDiePending)
301 : {
302 0 : ereport(WARNING,
303 : (errcode(ERRCODE_ADMIN_SHUTDOWN),
304 : errmsg("canceling the wait for synchronous replication and terminating connection due to administrator command"),
305 : errdetail("The transaction has already committed locally, but might not have been replicated to the standby.")));
306 0 : whereToSendOutput = DestNone;
307 0 : SyncRepCancelWait();
308 0 : break;
309 : }
310 :
311 : /*
312 : * It's unclear what to do if a query cancel interrupt arrives. We
313 : * can't actually abort at this point, but ignoring the interrupt
314 : * altogether is not helpful, so we just terminate the wait with a
315 : * suitable warning.
316 : */
317 34 : if (QueryCancelPending)
318 : {
319 0 : QueryCancelPending = false;
320 0 : ereport(WARNING,
321 : (errmsg("canceling wait for synchronous replication due to user request"),
322 : errdetail("The transaction has already committed locally, but might not have been replicated to the standby.")));
323 0 : SyncRepCancelWait();
324 0 : break;
325 : }
326 :
327 : /*
328 : * Wait on latch. Any condition that should wake us up will set the
329 : * latch, so no need for timeout.
330 : */
331 34 : rc = WaitLatch(MyLatch, WL_LATCH_SET | WL_POSTMASTER_DEATH, -1,
332 : WAIT_EVENT_SYNC_REP);
333 :
334 : /*
335 : * If the postmaster dies, we'll probably never get an acknowledgment,
336 : * because all the wal sender processes will exit. So just bail out.
337 : */
338 34 : if (rc & WL_POSTMASTER_DEATH)
339 : {
340 0 : ProcDiePending = true;
341 0 : whereToSendOutput = DestNone;
342 0 : SyncRepCancelWait();
343 0 : break;
344 : }
345 : }
346 :
347 : /*
348 : * WalSender has checked our LSN and has removed us from queue. Clean up
349 : * state and leave. It's OK to reset these shared memory fields without
350 : * holding SyncRepLock, because any walsenders will ignore us anyway when
351 : * we're not on the queue. We need a read barrier to make sure we see the
352 : * changes to the queue link (this might be unnecessary without
353 : * assertions, but better safe than sorry).
354 : */
355 34 : pg_read_barrier();
356 : Assert(dlist_node_is_detached(&MyProc->syncRepLinks));
357 34 : MyProc->syncRepState = SYNC_REP_NOT_WAITING;
358 34 : MyProc->waitLSN = InvalidXLogRecPtr;
359 :
360 : /* reset ps display to remove the suffix */
361 34 : if (update_process_title)
362 34 : set_ps_display_remove_suffix();
363 : }
364 :
365 : /*
366 : * Insert MyProc into the specified SyncRepQueue, maintaining sorted invariant.
367 : *
368 : * Usually we will go at tail of queue, though it's possible that we arrive
369 : * here out of order, so start at tail and work back to insertion point.
370 : */
371 : static void
372 34 : SyncRepQueueInsert(int mode)
373 : {
374 : dlist_head *queue;
375 : dlist_iter iter;
376 :
377 : Assert(mode >= 0 && mode < NUM_SYNC_REP_WAIT_MODE);
378 34 : queue = &WalSndCtl->SyncRepQueue[mode];
379 :
380 34 : dlist_reverse_foreach(iter, queue)
381 : {
382 0 : PGPROC *proc = dlist_container(PGPROC, syncRepLinks, iter.cur);
383 :
384 : /*
385 : * Stop at the queue element that we should insert after to ensure the
386 : * queue is ordered by LSN.
387 : */
388 0 : if (proc->waitLSN < MyProc->waitLSN)
389 : {
390 0 : dlist_insert_after(&proc->syncRepLinks, &MyProc->syncRepLinks);
391 0 : return;
392 : }
393 : }
394 :
395 : /*
396 : * If we get here, the list was either empty, or this process needs to be
397 : * at the head.
398 : */
399 34 : dlist_push_head(queue, &MyProc->syncRepLinks);
400 : }
401 :
402 : /*
403 : * Acquire SyncRepLock and cancel any wait currently in progress.
404 : */
405 : static void
406 0 : SyncRepCancelWait(void)
407 : {
408 0 : LWLockAcquire(SyncRepLock, LW_EXCLUSIVE);
409 0 : if (!dlist_node_is_detached(&MyProc->syncRepLinks))
410 0 : dlist_delete_thoroughly(&MyProc->syncRepLinks);
411 0 : MyProc->syncRepState = SYNC_REP_NOT_WAITING;
412 0 : LWLockRelease(SyncRepLock);
413 0 : }
414 :
415 : void
416 18348 : SyncRepCleanupAtProcExit(void)
417 : {
418 : /*
419 : * First check if we are removed from the queue without the lock to not
420 : * slow down backend exit.
421 : */
422 18348 : if (!dlist_node_is_detached(&MyProc->syncRepLinks))
423 : {
424 0 : LWLockAcquire(SyncRepLock, LW_EXCLUSIVE);
425 :
426 : /* maybe we have just been removed, so recheck */
427 0 : if (!dlist_node_is_detached(&MyProc->syncRepLinks))
428 0 : dlist_delete_thoroughly(&MyProc->syncRepLinks);
429 :
430 0 : LWLockRelease(SyncRepLock);
431 : }
432 18348 : }
433 :
434 : /*
435 : * ===========================================================
436 : * Synchronous Replication functions for wal sender processes
437 : * ===========================================================
438 : */
439 :
440 : /*
441 : * Take any action required to initialise sync rep state from config
442 : * data. Called at WALSender startup and after each SIGHUP.
443 : */
444 : void
445 746 : SyncRepInitConfig(void)
446 : {
447 : int priority;
448 :
449 : /*
450 : * Determine if we are a potential sync standby and remember the result
451 : * for handling replies from standby.
452 : */
453 746 : priority = SyncRepGetStandbyPriority();
454 746 : if (MyWalSnd->sync_standby_priority != priority)
455 : {
456 17 : SpinLockAcquire(&MyWalSnd->mutex);
457 17 : MyWalSnd->sync_standby_priority = priority;
458 17 : SpinLockRelease(&MyWalSnd->mutex);
459 :
460 17 : ereport(DEBUG1,
461 : (errmsg_internal("standby \"%s\" now has synchronous standby priority %d",
462 : application_name, priority)));
463 : }
464 746 : }
465 :
466 : /*
467 : * Update the LSNs on each queue based upon our latest state. This
468 : * implements a simple policy of first-valid-sync-standby-releases-waiter.
469 : *
470 : * Other policies are possible, which would change what we do here and
471 : * perhaps also which information we store as well.
472 : */
473 : void
474 114766 : SyncRepReleaseWaiters(void)
475 : {
476 114766 : volatile WalSndCtlData *walsndctl = WalSndCtl;
477 : XLogRecPtr writePtr;
478 : XLogRecPtr flushPtr;
479 : XLogRecPtr applyPtr;
480 : bool got_recptr;
481 : bool am_sync;
482 114766 : int numwrite = 0;
483 114766 : int numflush = 0;
484 114766 : int numapply = 0;
485 :
486 : /*
487 : * If this WALSender is serving a standby that is not on the list of
488 : * potential sync standbys then we have nothing to do. If we are still
489 : * starting up, still running base backup or the current flush position is
490 : * still invalid, then leave quickly also. Streaming or stopping WAL
491 : * senders are allowed to release waiters.
492 : */
493 114766 : if (MyWalSnd->sync_standby_priority == 0 ||
494 151 : (MyWalSnd->state != WALSNDSTATE_STREAMING &&
495 33 : MyWalSnd->state != WALSNDSTATE_STOPPING) ||
496 145 : !XLogRecPtrIsValid(MyWalSnd->flush))
497 : {
498 114621 : announce_next_takeover = true;
499 114625 : return;
500 : }
501 :
502 : /*
503 : * We're a potential sync standby. Release waiters if there are enough
504 : * sync standbys and we are considered as sync.
505 : */
506 145 : LWLockAcquire(SyncRepLock, LW_EXCLUSIVE);
507 :
508 : /*
509 : * Check whether we are a sync standby or not, and calculate the synced
510 : * positions among all sync standbys. (Note: although this step does not
511 : * of itself require holding SyncRepLock, it seems like a good idea to do
512 : * it after acquiring the lock. This ensures that the WAL pointers we use
513 : * to release waiters are newer than any previous execution of this
514 : * routine used.)
515 : */
516 145 : got_recptr = SyncRepGetSyncRecPtr(&writePtr, &flushPtr, &applyPtr, &am_sync);
517 :
518 : /*
519 : * If we are managing a sync standby, though we weren't prior to this,
520 : * then announce we are now a sync standby.
521 : */
522 145 : if (announce_next_takeover && am_sync)
523 : {
524 12 : announce_next_takeover = false;
525 :
526 12 : if (SyncRepConfig->syncrep_method == SYNC_REP_PRIORITY)
527 12 : ereport(LOG,
528 : (errmsg("standby \"%s\" is now a synchronous standby with priority %d",
529 : application_name, MyWalSnd->sync_standby_priority)));
530 : else
531 0 : ereport(LOG,
532 : (errmsg("standby \"%s\" is now a candidate for quorum synchronous standby",
533 : application_name)));
534 : }
535 :
536 : /*
537 : * If the number of sync standbys is less than requested or we aren't
538 : * managing a sync standby then just leave.
539 : */
540 145 : if (!got_recptr || !am_sync)
541 : {
542 4 : LWLockRelease(SyncRepLock);
543 4 : announce_next_takeover = !am_sync;
544 4 : return;
545 : }
546 :
547 : /*
548 : * Set the lsn first so that when we wake backends they will release up to
549 : * this location.
550 : */
551 141 : if (walsndctl->lsn[SYNC_REP_WAIT_WRITE] < writePtr)
552 : {
553 48 : walsndctl->lsn[SYNC_REP_WAIT_WRITE] = writePtr;
554 48 : numwrite = SyncRepWakeQueue(false, SYNC_REP_WAIT_WRITE);
555 : }
556 141 : if (walsndctl->lsn[SYNC_REP_WAIT_FLUSH] < flushPtr)
557 : {
558 50 : walsndctl->lsn[SYNC_REP_WAIT_FLUSH] = flushPtr;
559 50 : numflush = SyncRepWakeQueue(false, SYNC_REP_WAIT_FLUSH);
560 : }
561 141 : if (walsndctl->lsn[SYNC_REP_WAIT_APPLY] < applyPtr)
562 : {
563 45 : walsndctl->lsn[SYNC_REP_WAIT_APPLY] = applyPtr;
564 45 : numapply = SyncRepWakeQueue(false, SYNC_REP_WAIT_APPLY);
565 : }
566 :
567 141 : LWLockRelease(SyncRepLock);
568 :
569 141 : elog(DEBUG3, "released %d procs up to write %X/%08X, %d procs up to flush %X/%08X, %d procs up to apply %X/%08X",
570 : numwrite, LSN_FORMAT_ARGS(writePtr),
571 : numflush, LSN_FORMAT_ARGS(flushPtr),
572 : numapply, LSN_FORMAT_ARGS(applyPtr));
573 : }
574 :
575 : /*
576 : * Calculate the synced Write, Flush and Apply positions among sync standbys.
577 : *
578 : * Return false if the number of sync standbys is less than
579 : * synchronous_standby_names specifies. Otherwise return true and
580 : * store the positions into *writePtr, *flushPtr and *applyPtr.
581 : *
582 : * On return, *am_sync is set to true if this walsender is connecting to
583 : * sync standby. Otherwise it's set to false.
584 : */
585 : static bool
586 145 : SyncRepGetSyncRecPtr(XLogRecPtr *writePtr, XLogRecPtr *flushPtr,
587 : XLogRecPtr *applyPtr, bool *am_sync)
588 : {
589 : SyncRepStandbyData *sync_standbys;
590 : int num_standbys;
591 : int i;
592 :
593 : /* Initialize default results */
594 145 : *writePtr = InvalidXLogRecPtr;
595 145 : *flushPtr = InvalidXLogRecPtr;
596 145 : *applyPtr = InvalidXLogRecPtr;
597 145 : *am_sync = false;
598 :
599 : /* Quick out if not even configured to be synchronous */
600 145 : if (SyncRepConfig == NULL)
601 0 : return false;
602 :
603 : /* Get standbys that are considered as synchronous at this moment */
604 145 : num_standbys = SyncRepGetCandidateStandbys(&sync_standbys);
605 :
606 : /* Am I among the candidate sync standbys? */
607 151 : for (i = 0; i < num_standbys; i++)
608 : {
609 148 : if (sync_standbys[i].is_me)
610 : {
611 142 : *am_sync = true;
612 142 : break;
613 : }
614 : }
615 :
616 : /*
617 : * Nothing more to do if we are not managing a sync standby or there are
618 : * not enough synchronous standbys.
619 : */
620 145 : if (!(*am_sync) ||
621 142 : num_standbys < SyncRepConfig->num_sync)
622 : {
623 4 : pfree(sync_standbys);
624 4 : return false;
625 : }
626 :
627 : /*
628 : * In a priority-based sync replication, the synced positions are the
629 : * oldest ones among sync standbys. In a quorum-based, they are the Nth
630 : * latest ones.
631 : *
632 : * SyncRepGetNthLatestSyncRecPtr() also can calculate the oldest
633 : * positions. But we use SyncRepGetOldestSyncRecPtr() for that calculation
634 : * because it's a bit more efficient.
635 : *
636 : * XXX If the numbers of current and requested sync standbys are the same,
637 : * we can use SyncRepGetOldestSyncRecPtr() to calculate the synced
638 : * positions even in a quorum-based sync replication.
639 : */
640 141 : if (SyncRepConfig->syncrep_method == SYNC_REP_PRIORITY)
641 : {
642 141 : SyncRepGetOldestSyncRecPtr(writePtr, flushPtr, applyPtr,
643 : sync_standbys, num_standbys);
644 : }
645 : else
646 : {
647 0 : SyncRepGetNthLatestSyncRecPtr(writePtr, flushPtr, applyPtr,
648 : sync_standbys, num_standbys,
649 0 : SyncRepConfig->num_sync);
650 : }
651 :
652 141 : pfree(sync_standbys);
653 141 : return true;
654 : }
655 :
656 : /*
657 : * Calculate the oldest Write, Flush and Apply positions among sync standbys.
658 : */
659 : static void
660 141 : SyncRepGetOldestSyncRecPtr(XLogRecPtr *writePtr,
661 : XLogRecPtr *flushPtr,
662 : XLogRecPtr *applyPtr,
663 : SyncRepStandbyData *sync_standbys,
664 : int num_standbys)
665 : {
666 : int i;
667 :
668 : /*
669 : * Scan through all sync standbys and calculate the oldest Write, Flush
670 : * and Apply positions. We assume *writePtr et al were initialized to
671 : * InvalidXLogRecPtr.
672 : */
673 285 : for (i = 0; i < num_standbys; i++)
674 : {
675 144 : XLogRecPtr write = sync_standbys[i].write;
676 144 : XLogRecPtr flush = sync_standbys[i].flush;
677 144 : XLogRecPtr apply = sync_standbys[i].apply;
678 :
679 144 : if (!XLogRecPtrIsValid(*writePtr) || *writePtr > write)
680 141 : *writePtr = write;
681 144 : if (!XLogRecPtrIsValid(*flushPtr) || *flushPtr > flush)
682 141 : *flushPtr = flush;
683 144 : if (!XLogRecPtrIsValid(*applyPtr) || *applyPtr > apply)
684 141 : *applyPtr = apply;
685 : }
686 141 : }
687 :
688 : /*
689 : * Calculate the Nth latest Write, Flush and Apply positions among sync
690 : * standbys.
691 : */
692 : static void
693 0 : SyncRepGetNthLatestSyncRecPtr(XLogRecPtr *writePtr,
694 : XLogRecPtr *flushPtr,
695 : XLogRecPtr *applyPtr,
696 : SyncRepStandbyData *sync_standbys,
697 : int num_standbys,
698 : uint8 nth)
699 : {
700 : XLogRecPtr *write_array;
701 : XLogRecPtr *flush_array;
702 : XLogRecPtr *apply_array;
703 : int i;
704 :
705 : /* Should have enough candidates, or somebody messed up */
706 : Assert(nth > 0 && nth <= num_standbys);
707 :
708 0 : write_array = palloc_array(XLogRecPtr, num_standbys);
709 0 : flush_array = palloc_array(XLogRecPtr, num_standbys);
710 0 : apply_array = palloc_array(XLogRecPtr, num_standbys);
711 :
712 0 : for (i = 0; i < num_standbys; i++)
713 : {
714 0 : write_array[i] = sync_standbys[i].write;
715 0 : flush_array[i] = sync_standbys[i].flush;
716 0 : apply_array[i] = sync_standbys[i].apply;
717 : }
718 :
719 : /* Sort each array in descending order */
720 0 : qsort(write_array, num_standbys, sizeof(XLogRecPtr), cmp_lsn);
721 0 : qsort(flush_array, num_standbys, sizeof(XLogRecPtr), cmp_lsn);
722 0 : qsort(apply_array, num_standbys, sizeof(XLogRecPtr), cmp_lsn);
723 :
724 : /* Get Nth latest Write, Flush, Apply positions */
725 0 : *writePtr = write_array[nth - 1];
726 0 : *flushPtr = flush_array[nth - 1];
727 0 : *applyPtr = apply_array[nth - 1];
728 :
729 0 : pfree(write_array);
730 0 : pfree(flush_array);
731 0 : pfree(apply_array);
732 0 : }
733 :
734 : /*
735 : * Compare lsn in order to sort array in descending order.
736 : */
737 : static int
738 0 : cmp_lsn(const void *a, const void *b)
739 : {
740 0 : XLogRecPtr lsn1 = *((const XLogRecPtr *) a);
741 0 : XLogRecPtr lsn2 = *((const XLogRecPtr *) b);
742 :
743 0 : return pg_cmp_u64(lsn2, lsn1);
744 : }
745 :
746 : /*
747 : * Return data about walsenders that are candidates to be sync standbys.
748 : *
749 : * *standbys is set to a palloc'd array of structs of per-walsender data,
750 : * and the number of valid entries (candidate sync senders) is returned.
751 : * (This might be more or fewer than num_sync; caller must check.)
752 : */
753 : int
754 800 : SyncRepGetCandidateStandbys(SyncRepStandbyData **standbys)
755 : {
756 : int i;
757 : int n;
758 :
759 : /* Create result array */
760 800 : *standbys = palloc_array(SyncRepStandbyData, max_wal_senders);
761 :
762 : /* Quick exit if sync replication is not requested */
763 800 : if (SyncRepConfig == NULL)
764 637 : return 0;
765 :
766 : /* Collect raw data from shared memory */
767 163 : n = 0;
768 1793 : for (i = 0; i < max_wal_senders; i++)
769 : {
770 : volatile WalSnd *walsnd; /* Use volatile pointer to prevent code
771 : * rearrangement */
772 : SyncRepStandbyData *stby;
773 : WalSndState state; /* not included in SyncRepStandbyData */
774 :
775 1630 : walsnd = &WalSndCtl->walsnds[i];
776 1630 : stby = *standbys + n;
777 :
778 1630 : SpinLockAcquire(&walsnd->mutex);
779 1630 : stby->pid = walsnd->pid;
780 1630 : state = walsnd->state;
781 1630 : stby->write = walsnd->write;
782 1630 : stby->flush = walsnd->flush;
783 1630 : stby->apply = walsnd->apply;
784 1630 : stby->sync_standby_priority = walsnd->sync_standby_priority;
785 1630 : SpinLockRelease(&walsnd->mutex);
786 :
787 : /* Must be active */
788 1630 : if (stby->pid == 0)
789 1415 : continue;
790 :
791 : /* Must be streaming or stopping */
792 215 : if (state != WALSNDSTATE_STREAMING &&
793 : state != WALSNDSTATE_STOPPING)
794 0 : continue;
795 :
796 : /* Must be synchronous */
797 215 : if (stby->sync_standby_priority == 0)
798 12 : continue;
799 :
800 : /* Must have a valid flush position */
801 203 : if (!XLogRecPtrIsValid(stby->flush))
802 0 : continue;
803 :
804 : /* OK, it's a candidate */
805 203 : stby->walsnd_index = i;
806 203 : stby->is_me = (walsnd == MyWalSnd);
807 203 : n++;
808 : }
809 :
810 : /*
811 : * In quorum mode, we return all the candidates. In priority mode, if we
812 : * have too many candidates then return only the num_sync ones of highest
813 : * priority.
814 : */
815 163 : if (SyncRepConfig->syncrep_method == SYNC_REP_PRIORITY &&
816 161 : n > SyncRepConfig->num_sync)
817 : {
818 : /* Sort by priority ... */
819 15 : qsort(*standbys, n, sizeof(SyncRepStandbyData),
820 : standby_priority_comparator);
821 : /* ... then report just the first num_sync ones */
822 15 : n = SyncRepConfig->num_sync;
823 : }
824 :
825 163 : return n;
826 : }
827 :
828 : /*
829 : * qsort comparator to sort SyncRepStandbyData entries by priority
830 : */
831 : static int
832 39 : standby_priority_comparator(const void *a, const void *b)
833 : {
834 39 : const SyncRepStandbyData *sa = (const SyncRepStandbyData *) a;
835 39 : const SyncRepStandbyData *sb = (const SyncRepStandbyData *) b;
836 :
837 : /* First, sort by increasing priority value */
838 39 : if (sa->sync_standby_priority != sb->sync_standby_priority)
839 21 : return sa->sync_standby_priority - sb->sync_standby_priority;
840 :
841 : /*
842 : * We might have equal priority values; arbitrarily break ties by position
843 : * in the WalSnd array. (This is utterly bogus, since that is arrival
844 : * order dependent, but there are regression tests that rely on it.)
845 : */
846 18 : return sa->walsnd_index - sb->walsnd_index;
847 : }
848 :
849 :
850 : /*
851 : * Check if we are in the list of sync standbys, and if so, determine
852 : * priority sequence. Return priority if set, or zero to indicate that
853 : * we are not a potential sync standby.
854 : *
855 : * Compare the parameter SyncRepStandbyNames against the application_name
856 : * for this WALSender, or allow any name if we find a wildcard "*".
857 : */
858 : static int
859 746 : SyncRepGetStandbyPriority(void)
860 : {
861 : const char *standby_name;
862 : int priority;
863 746 : bool found = false;
864 :
865 : /*
866 : * Since synchronous cascade replication is not allowed, we always set the
867 : * priority of cascading walsender to zero.
868 : */
869 746 : if (am_cascading_walsender)
870 26 : return 0;
871 :
872 720 : if (!SyncStandbysDefined() || SyncRepConfig == NULL)
873 696 : return 0;
874 :
875 24 : standby_name = SyncRepConfig->member_names;
876 32 : for (priority = 1; priority <= SyncRepConfig->nmembers; priority++)
877 : {
878 31 : if (pg_strcasecmp(standby_name, application_name) == 0 ||
879 18 : strcmp(standby_name, "*") == 0)
880 : {
881 23 : found = true;
882 23 : break;
883 : }
884 8 : standby_name += strlen(standby_name) + 1;
885 : }
886 :
887 24 : if (!found)
888 1 : return 0;
889 :
890 : /*
891 : * In quorum-based sync replication, all the standbys in the list have the
892 : * same priority, one.
893 : */
894 23 : return (SyncRepConfig->syncrep_method == SYNC_REP_PRIORITY) ? priority : 1;
895 : }
896 :
897 : /*
898 : * Walk the specified queue from head. Set the state of any backends that
899 : * need to be woken, remove them from the queue, and then wake them.
900 : * Pass all = true to wake whole queue; otherwise, just wake up to
901 : * the walsender's LSN.
902 : *
903 : * The caller must hold SyncRepLock in exclusive mode.
904 : */
905 : static int
906 146 : SyncRepWakeQueue(bool all, int mode)
907 : {
908 146 : volatile WalSndCtlData *walsndctl = WalSndCtl;
909 146 : int numprocs = 0;
910 : dlist_mutable_iter iter;
911 :
912 : Assert(mode >= 0 && mode < NUM_SYNC_REP_WAIT_MODE);
913 : Assert(LWLockHeldByMeInMode(SyncRepLock, LW_EXCLUSIVE));
914 : Assert(SyncRepQueueIsOrderedByLSN(mode));
915 :
916 165 : dlist_foreach_modify(iter, &WalSndCtl->SyncRepQueue[mode])
917 : {
918 21 : PGPROC *proc = dlist_container(PGPROC, syncRepLinks, iter.cur);
919 :
920 : /*
921 : * Assume the queue is ordered by LSN
922 : */
923 21 : if (!all && walsndctl->lsn[mode] < proc->waitLSN)
924 2 : return numprocs;
925 :
926 : /*
927 : * Remove from queue.
928 : */
929 19 : dlist_delete_thoroughly(&proc->syncRepLinks);
930 :
931 : /*
932 : * SyncRepWaitForLSN() reads syncRepState without holding the lock, so
933 : * make sure that it sees the queue link being removed before the
934 : * syncRepState change.
935 : */
936 19 : pg_write_barrier();
937 :
938 : /*
939 : * Set state to complete; see SyncRepWaitForLSN() for discussion of
940 : * the various states.
941 : */
942 19 : proc->syncRepState = SYNC_REP_WAIT_COMPLETE;
943 :
944 : /*
945 : * Wake only when we have set state and removed from queue.
946 : */
947 19 : SetLatch(&(proc->procLatch));
948 :
949 19 : numprocs++;
950 : }
951 :
952 144 : return numprocs;
953 : }
954 :
955 : /*
956 : * The checkpointer calls this as needed to update the shared
957 : * sync_standbys_status flag, so that backends don't remain permanently wedged
958 : * if synchronous_standby_names is unset. It's safe to check the current value
959 : * without the lock, because it's only ever updated by one process. But we
960 : * must take the lock to change it.
961 : */
962 : void
963 648 : SyncRepUpdateSyncStandbysDefined(void)
964 : {
965 648 : bool sync_standbys_defined = SyncStandbysDefined();
966 :
967 648 : if (sync_standbys_defined !=
968 648 : ((WalSndCtl->sync_standbys_status & SYNC_STANDBY_DEFINED) != 0))
969 : {
970 13 : LWLockAcquire(SyncRepLock, LW_EXCLUSIVE);
971 :
972 : /*
973 : * If synchronous_standby_names has been reset to empty, it's futile
974 : * for backends to continue waiting. Since the user no longer wants
975 : * synchronous replication, we'd better wake them up.
976 : */
977 13 : if (!sync_standbys_defined)
978 : {
979 : int i;
980 :
981 4 : for (i = 0; i < NUM_SYNC_REP_WAIT_MODE; i++)
982 3 : SyncRepWakeQueue(true, i);
983 : }
984 :
985 : /*
986 : * Only allow people to join the queue when there are synchronous
987 : * standbys defined. Without this interlock, there's a race
988 : * condition: we might wake up all the current waiters; then, some
989 : * backend that hasn't yet reloaded its config might go to sleep on
990 : * the queue (and never wake up). This prevents that.
991 : */
992 13 : WalSndCtl->sync_standbys_status = SYNC_STANDBY_INIT |
993 : (sync_standbys_defined ? SYNC_STANDBY_DEFINED : 0);
994 :
995 13 : LWLockRelease(SyncRepLock);
996 : }
997 635 : else if ((WalSndCtl->sync_standbys_status & SYNC_STANDBY_INIT) == 0)
998 : {
999 572 : LWLockAcquire(SyncRepLock, LW_EXCLUSIVE);
1000 :
1001 : /*
1002 : * Note that there is no need to wake up the queues here. We would
1003 : * reach this path only if SyncStandbysDefined() returns false, or it
1004 : * would mean that some backends are waiting with the GUC set. See
1005 : * SyncRepWaitForLSN().
1006 : */
1007 : Assert(!SyncStandbysDefined());
1008 :
1009 : /*
1010 : * Even if there is no sync standby defined, let the readers of this
1011 : * information know that the sync standby data has been initialized.
1012 : * This can just be done once, hence the previous check on
1013 : * SYNC_STANDBY_INIT to avoid useless work.
1014 : */
1015 572 : WalSndCtl->sync_standbys_status |= SYNC_STANDBY_INIT;
1016 :
1017 572 : LWLockRelease(SyncRepLock);
1018 : }
1019 648 : }
1020 :
1021 : #ifdef USE_ASSERT_CHECKING
1022 : static bool
1023 : SyncRepQueueIsOrderedByLSN(int mode)
1024 : {
1025 : XLogRecPtr lastLSN;
1026 : dlist_iter iter;
1027 :
1028 : Assert(mode >= 0 && mode < NUM_SYNC_REP_WAIT_MODE);
1029 :
1030 : lastLSN = InvalidXLogRecPtr;
1031 :
1032 : dlist_foreach(iter, &WalSndCtl->SyncRepQueue[mode])
1033 : {
1034 : PGPROC *proc = dlist_container(PGPROC, syncRepLinks, iter.cur);
1035 :
1036 : /*
1037 : * Check the queue is ordered by LSN and that multiple procs don't
1038 : * have matching LSNs
1039 : */
1040 : if (proc->waitLSN <= lastLSN)
1041 : return false;
1042 :
1043 : lastLSN = proc->waitLSN;
1044 : }
1045 :
1046 : return true;
1047 : }
1048 : #endif
1049 :
1050 : /*
1051 : * ===========================================================
1052 : * Synchronous Replication functions executed by any process
1053 : * ===========================================================
1054 : */
1055 :
1056 : bool
1057 1279 : check_synchronous_standby_names(char **newval, void **extra, GucSource source)
1058 : {
1059 1279 : if (*newval != NULL && (*newval)[0] != '\0')
1060 78 : {
1061 : yyscan_t scanner;
1062 : int parse_rc;
1063 : SyncRepConfigData *pconf;
1064 :
1065 : /* Result of parsing is returned in one of these two variables */
1066 78 : SyncRepConfigData *syncrep_parse_result = NULL;
1067 78 : char *syncrep_parse_error_msg = NULL;
1068 :
1069 : /* Parse the synchronous_standby_names string */
1070 78 : syncrep_scanner_init(*newval, &scanner);
1071 78 : parse_rc = syncrep_yyparse(&syncrep_parse_result, &syncrep_parse_error_msg, scanner);
1072 78 : syncrep_scanner_finish(scanner);
1073 :
1074 78 : if (parse_rc != 0 || syncrep_parse_result == NULL)
1075 : {
1076 0 : GUC_check_errcode(ERRCODE_SYNTAX_ERROR);
1077 0 : if (syncrep_parse_error_msg)
1078 0 : GUC_check_errdetail("%s", syncrep_parse_error_msg);
1079 : else
1080 : /* translator: %s is a GUC name */
1081 0 : GUC_check_errdetail("\"%s\" parser failed.",
1082 : "synchronous_standby_names");
1083 0 : return false;
1084 : }
1085 :
1086 78 : if (syncrep_parse_result->num_sync <= 0)
1087 : {
1088 0 : GUC_check_errmsg("number of synchronous standbys (%d) must be greater than zero",
1089 0 : syncrep_parse_result->num_sync);
1090 0 : return false;
1091 : }
1092 :
1093 : /* GUC extra value must be guc_malloc'd, not palloc'd */
1094 : pconf = (SyncRepConfigData *)
1095 78 : guc_malloc(LOG, syncrep_parse_result->config_size);
1096 78 : if (pconf == NULL)
1097 0 : return false;
1098 78 : memcpy(pconf, syncrep_parse_result, syncrep_parse_result->config_size);
1099 :
1100 78 : *extra = pconf;
1101 :
1102 : /*
1103 : * We need not explicitly clean up syncrep_parse_result. It, and any
1104 : * other cruft generated during parsing, will be freed when the
1105 : * current memory context is deleted. (This code is generally run in
1106 : * a short-lived context used for config file processing, so that will
1107 : * not be very long.)
1108 : */
1109 : }
1110 : else
1111 1201 : *extra = NULL;
1112 :
1113 1279 : return true;
1114 : }
1115 :
1116 : void
1117 1269 : assign_synchronous_standby_names(const char *newval, void *extra)
1118 : {
1119 1269 : SyncRepConfig = (SyncRepConfigData *) extra;
1120 1269 : }
1121 :
1122 : void
1123 3451 : assign_synchronous_commit(int newval, void *extra)
1124 : {
1125 3451 : switch (newval)
1126 : {
1127 0 : case SYNCHRONOUS_COMMIT_REMOTE_WRITE:
1128 0 : SyncRepWaitMode = SYNC_REP_WAIT_WRITE;
1129 0 : break;
1130 1327 : case SYNCHRONOUS_COMMIT_REMOTE_FLUSH:
1131 1327 : SyncRepWaitMode = SYNC_REP_WAIT_FLUSH;
1132 1327 : break;
1133 2 : case SYNCHRONOUS_COMMIT_REMOTE_APPLY:
1134 2 : SyncRepWaitMode = SYNC_REP_WAIT_APPLY;
1135 2 : break;
1136 2122 : default:
1137 2122 : SyncRepWaitMode = SYNC_REP_NO_WAIT;
1138 2122 : break;
1139 : }
1140 3451 : }
|
