Line data Source code
1 : /*-------------------------------------------------------------------------
2 : * applyparallelworker.c
3 : * Support routines for applying xact by parallel apply worker
4 : *
5 : * Copyright (c) 2023-2025, PostgreSQL Global Development Group
6 : *
7 : * IDENTIFICATION
8 : * src/backend/replication/logical/applyparallelworker.c
9 : *
10 : * This file contains the code to launch, set up, and teardown a parallel apply
11 : * worker which receives the changes from the leader worker and invokes routines
12 : * to apply those on the subscriber database. Additionally, this file contains
13 : * routines that are intended to support setting up, using, and tearing down a
14 : * ParallelApplyWorkerInfo which is required so the leader worker and parallel
15 : * apply workers can communicate with each other.
16 : *
17 : * The parallel apply workers are assigned (if available) as soon as xact's
18 : * first stream is received for subscriptions that have set their 'streaming'
19 : * option as parallel. The leader apply worker will send changes to this new
20 : * worker via shared memory. We keep this worker assigned till the transaction
21 : * commit is received and also wait for the worker to finish at commit. This
22 : * preserves commit ordering and avoid file I/O in most cases, although we
23 : * still need to spill to a file if there is no worker available. See comments
24 : * atop logical/worker to know more about streamed xacts whose changes are
25 : * spilled to disk. It is important to maintain commit order to avoid failures
26 : * due to: (a) transaction dependencies - say if we insert a row in the first
27 : * transaction and update it in the second transaction on publisher then
28 : * allowing the subscriber to apply both in parallel can lead to failure in the
29 : * update; (b) deadlocks - allowing transactions that update the same set of
30 : * rows/tables in the opposite order to be applied in parallel can lead to
31 : * deadlocks.
32 : *
33 : * A worker pool is used to avoid restarting workers for each streaming
34 : * transaction. We maintain each worker's information (ParallelApplyWorkerInfo)
35 : * in the ParallelApplyWorkerPool. After successfully launching a new worker,
36 : * its information is added to the ParallelApplyWorkerPool. Once the worker
37 : * finishes applying the transaction, it is marked as available for re-use.
38 : * Now, before starting a new worker to apply the streaming transaction, we
39 : * check the list for any available worker. Note that we retain a maximum of
40 : * half the max_parallel_apply_workers_per_subscription workers in the pool and
41 : * after that, we simply exit the worker after applying the transaction.
42 : *
43 : * XXX This worker pool threshold is arbitrary and we can provide a GUC
44 : * variable for this in the future if required.
45 : *
46 : * The leader apply worker will create a separate dynamic shared memory segment
47 : * when each parallel apply worker starts. The reason for this design is that
48 : * we cannot predict how many workers will be needed. It may be possible to
49 : * allocate enough shared memory in one segment based on the maximum number of
50 : * parallel apply workers (max_parallel_apply_workers_per_subscription), but
51 : * this would waste memory if no process is actually started.
52 : *
53 : * The dynamic shared memory segment contains: (a) a shm_mq that is used to
54 : * send changes in the transaction from leader apply worker to parallel apply
55 : * worker; (b) another shm_mq that is used to send errors (and other messages
56 : * reported via elog/ereport) from the parallel apply worker to leader apply
57 : * worker; (c) necessary information to be shared among parallel apply workers
58 : * and the leader apply worker (i.e. members of ParallelApplyWorkerShared).
59 : *
60 : * Locking Considerations
61 : * ----------------------
62 : * We have a risk of deadlock due to concurrently applying the transactions in
63 : * parallel mode that were independent on the publisher side but became
64 : * dependent on the subscriber side due to the different database structures
65 : * (like schema of subscription tables, constraints, etc.) on each side. This
66 : * can happen even without parallel mode when there are concurrent operations
67 : * on the subscriber. In order to detect the deadlocks among leader (LA) and
68 : * parallel apply (PA) workers, we used lmgr locks when the PA waits for the
69 : * next stream (set of changes) and LA waits for PA to finish the transaction.
70 : * An alternative approach could be to not allow parallelism when the schema of
71 : * tables is different between the publisher and subscriber but that would be
72 : * too restrictive and would require the publisher to send much more
73 : * information than it is currently sending.
74 : *
75 : * Consider a case where the subscribed table does not have a unique key on the
76 : * publisher and has a unique key on the subscriber. The deadlock can happen in
77 : * the following ways:
78 : *
79 : * 1) Deadlock between the leader apply worker and a parallel apply worker
80 : *
81 : * Consider that the parallel apply worker (PA) is executing TX-1 and the
82 : * leader apply worker (LA) is executing TX-2 concurrently on the subscriber.
83 : * Now, LA is waiting for PA because of the unique key constraint of the
84 : * subscribed table while PA is waiting for LA to send the next stream of
85 : * changes or transaction finish command message.
86 : *
87 : * In order for lmgr to detect this, we have LA acquire a session lock on the
88 : * remote transaction (by pa_lock_stream()) and have PA wait on the lock before
89 : * trying to receive the next stream of changes. Specifically, LA will acquire
90 : * the lock in AccessExclusive mode before sending the STREAM_STOP and will
91 : * release it if already acquired after sending the STREAM_START, STREAM_ABORT
92 : * (for toplevel transaction), STREAM_PREPARE, and STREAM_COMMIT. The PA will
93 : * acquire the lock in AccessShare mode after processing STREAM_STOP and
94 : * STREAM_ABORT (for subtransaction) and then release the lock immediately
95 : * after acquiring it.
96 : *
97 : * The lock graph for the above example will look as follows:
98 : * LA (waiting to acquire the lock on the unique index) -> PA (waiting to
99 : * acquire the stream lock) -> LA
100 : *
101 : * This way, when PA is waiting for LA for the next stream of changes, we can
102 : * have a wait-edge from PA to LA in lmgr, which will make us detect the
103 : * deadlock between LA and PA.
104 : *
105 : * 2) Deadlock between the leader apply worker and parallel apply workers
106 : *
107 : * This scenario is similar to the first case but TX-1 and TX-2 are executed by
108 : * two parallel apply workers (PA-1 and PA-2 respectively). In this scenario,
109 : * PA-2 is waiting for PA-1 to complete its transaction while PA-1 is waiting
110 : * for subsequent input from LA. Also, LA is waiting for PA-2 to complete its
111 : * transaction in order to preserve the commit order. There is a deadlock among
112 : * the three processes.
113 : *
114 : * In order for lmgr to detect this, we have PA acquire a session lock (this is
115 : * a different lock than referred in the previous case, see
116 : * pa_lock_transaction()) on the transaction being applied and have LA wait on
117 : * the lock before proceeding in the transaction finish commands. Specifically,
118 : * PA will acquire this lock in AccessExclusive mode before executing the first
119 : * message of the transaction and release it at the xact end. LA will acquire
120 : * this lock in AccessShare mode at transaction finish commands and release it
121 : * immediately.
122 : *
123 : * The lock graph for the above example will look as follows:
124 : * LA (waiting to acquire the transaction lock) -> PA-2 (waiting to acquire the
125 : * lock due to unique index constraint) -> PA-1 (waiting to acquire the stream
126 : * lock) -> LA
127 : *
128 : * This way when LA is waiting to finish the transaction end command to preserve
129 : * the commit order, we will be able to detect deadlock, if any.
130 : *
131 : * One might think we can use XactLockTableWait(), but XactLockTableWait()
132 : * considers PREPARED TRANSACTION as still in progress which means the lock
133 : * won't be released even after the parallel apply worker has prepared the
134 : * transaction.
135 : *
136 : * 3) Deadlock when the shm_mq buffer is full
137 : *
138 : * In the previous scenario (ie. PA-1 and PA-2 are executing transactions
139 : * concurrently), if the shm_mq buffer between LA and PA-2 is full, LA has to
140 : * wait to send messages, and this wait doesn't appear in lmgr.
141 : *
142 : * To avoid this wait, we use a non-blocking write and wait with a timeout. If
143 : * the timeout is exceeded, the LA will serialize all the pending messages to
144 : * a file and indicate PA-2 that it needs to read that file for the remaining
145 : * messages. Then LA will start waiting for commit as in the previous case
146 : * which will detect deadlock if any. See pa_send_data() and
147 : * enum TransApplyAction.
148 : *
149 : * Lock types
150 : * ----------
151 : * Both the stream lock and the transaction lock mentioned above are
152 : * session-level locks because both locks could be acquired outside the
153 : * transaction, and the stream lock in the leader needs to persist across
154 : * transaction boundaries i.e. until the end of the streaming transaction.
155 : *-------------------------------------------------------------------------
156 : */
157 :
158 : #include "postgres.h"
159 :
160 : #include "libpq/pqformat.h"
161 : #include "libpq/pqmq.h"
162 : #include "pgstat.h"
163 : #include "postmaster/interrupt.h"
164 : #include "replication/logicallauncher.h"
165 : #include "replication/logicalworker.h"
166 : #include "replication/origin.h"
167 : #include "replication/worker_internal.h"
168 : #include "storage/ipc.h"
169 : #include "storage/lmgr.h"
170 : #include "tcop/tcopprot.h"
171 : #include "utils/inval.h"
172 : #include "utils/memutils.h"
173 : #include "utils/syscache.h"
174 :
175 : #define PG_LOGICAL_APPLY_SHM_MAGIC 0x787ca067
176 :
177 : /*
178 : * DSM keys for parallel apply worker. Unlike other parallel execution code,
179 : * since we don't need to worry about DSM keys conflicting with plan_node_id we
180 : * can use small integers.
181 : */
182 : #define PARALLEL_APPLY_KEY_SHARED 1
183 : #define PARALLEL_APPLY_KEY_MQ 2
184 : #define PARALLEL_APPLY_KEY_ERROR_QUEUE 3
185 :
186 : /* Queue size of DSM, 16 MB for now. */
187 : #define DSM_QUEUE_SIZE (16 * 1024 * 1024)
188 :
189 : /*
190 : * Error queue size of DSM. It is desirable to make it large enough that a
191 : * typical ErrorResponse can be sent without blocking. That way, a worker that
192 : * errors out can write the whole message into the queue and terminate without
193 : * waiting for the user backend.
194 : */
195 : #define DSM_ERROR_QUEUE_SIZE (16 * 1024)
196 :
197 : /*
198 : * There are three fields in each message received by the parallel apply
199 : * worker: start_lsn, end_lsn and send_time. Because we have updated these
200 : * statistics in the leader apply worker, we can ignore these fields in the
201 : * parallel apply worker (see function LogicalRepApplyLoop).
202 : */
203 : #define SIZE_STATS_MESSAGE (2 * sizeof(XLogRecPtr) + sizeof(TimestampTz))
204 :
205 : /*
206 : * The type of session-level lock on a transaction being applied on a logical
207 : * replication subscriber.
208 : */
209 : #define PARALLEL_APPLY_LOCK_STREAM 0
210 : #define PARALLEL_APPLY_LOCK_XACT 1
211 :
212 : /*
213 : * Hash table entry to map xid to the parallel apply worker state.
214 : */
215 : typedef struct ParallelApplyWorkerEntry
216 : {
217 : TransactionId xid; /* Hash key -- must be first */
218 : ParallelApplyWorkerInfo *winfo;
219 : } ParallelApplyWorkerEntry;
220 :
221 : /*
222 : * A hash table used to cache the state of streaming transactions being applied
223 : * by the parallel apply workers.
224 : */
225 : static HTAB *ParallelApplyTxnHash = NULL;
226 :
227 : /*
228 : * A list (pool) of active parallel apply workers. The information for
229 : * the new worker is added to the list after successfully launching it. The
230 : * list entry is removed if there are already enough workers in the worker
231 : * pool at the end of the transaction. For more information about the worker
232 : * pool, see comments atop this file.
233 : */
234 : static List *ParallelApplyWorkerPool = NIL;
235 :
236 : /*
237 : * Information shared between leader apply worker and parallel apply worker.
238 : */
239 : ParallelApplyWorkerShared *MyParallelShared = NULL;
240 :
241 : /*
242 : * Is there a message sent by a parallel apply worker that the leader apply
243 : * worker needs to receive?
244 : */
245 : volatile sig_atomic_t ParallelApplyMessagePending = false;
246 :
247 : /*
248 : * Cache the parallel apply worker information required for applying the
249 : * current streaming transaction. It is used to save the cost of searching the
250 : * hash table when applying the changes between STREAM_START and STREAM_STOP.
251 : */
252 : static ParallelApplyWorkerInfo *stream_apply_worker = NULL;
253 :
254 : /* A list to maintain subtransactions, if any. */
255 : static List *subxactlist = NIL;
256 :
257 : static void pa_free_worker_info(ParallelApplyWorkerInfo *winfo);
258 : static ParallelTransState pa_get_xact_state(ParallelApplyWorkerShared *wshared);
259 : static PartialFileSetState pa_get_fileset_state(void);
260 :
261 : /*
262 : * Returns true if it is OK to start a parallel apply worker, false otherwise.
263 : */
264 : static bool
265 164 : pa_can_start(void)
266 : {
267 : /* Only leader apply workers can start parallel apply workers. */
268 164 : if (!am_leader_apply_worker())
269 54 : return false;
270 :
271 : /*
272 : * It is good to check for any change in the subscription parameter to
273 : * avoid the case where for a very long time the change doesn't get
274 : * reflected. This can happen when there is a constant flow of streaming
275 : * transactions that are handled by parallel apply workers.
276 : *
277 : * It is better to do it before the below checks so that the latest values
278 : * of subscription can be used for the checks.
279 : */
280 110 : maybe_reread_subscription();
281 :
282 : /*
283 : * Don't start a new parallel apply worker if the subscription is not
284 : * using parallel streaming mode, or if the publisher does not support
285 : * parallel apply.
286 : */
287 110 : if (!MyLogicalRepWorker->parallel_apply)
288 56 : return false;
289 :
290 : /*
291 : * Don't start a new parallel worker if user has set skiplsn as it's
292 : * possible that they want to skip the streaming transaction. For
293 : * streaming transactions, we need to serialize the transaction to a file
294 : * so that we can get the last LSN of the transaction to judge whether to
295 : * skip before starting to apply the change.
296 : *
297 : * One might think that we could allow parallelism if the first lsn of the
298 : * transaction is greater than skiplsn, but we don't send it with the
299 : * STREAM START message, and it doesn't seem worth sending the extra eight
300 : * bytes with the STREAM START to enable parallelism for this case.
301 : */
302 54 : if (!XLogRecPtrIsInvalid(MySubscription->skiplsn))
303 0 : return false;
304 :
305 : /*
306 : * For streaming transactions that are being applied using a parallel
307 : * apply worker, we cannot decide whether to apply the change for a
308 : * relation that is not in the READY state (see
309 : * should_apply_changes_for_rel) as we won't know remote_final_lsn by that
310 : * time. So, we don't start the new parallel apply worker in this case.
311 : */
312 54 : if (!AllTablesyncsReady())
313 0 : return false;
314 :
315 54 : return true;
316 : }
317 :
318 : /*
319 : * Set up a dynamic shared memory segment.
320 : *
321 : * We set up a control region that contains a fixed-size worker info
322 : * (ParallelApplyWorkerShared), a message queue, and an error queue.
323 : *
324 : * Returns true on success, false on failure.
325 : */
326 : static bool
327 20 : pa_setup_dsm(ParallelApplyWorkerInfo *winfo)
328 : {
329 : shm_toc_estimator e;
330 : Size segsize;
331 : dsm_segment *seg;
332 : shm_toc *toc;
333 : ParallelApplyWorkerShared *shared;
334 : shm_mq *mq;
335 20 : Size queue_size = DSM_QUEUE_SIZE;
336 20 : Size error_queue_size = DSM_ERROR_QUEUE_SIZE;
337 :
338 : /*
339 : * Estimate how much shared memory we need.
340 : *
341 : * Because the TOC machinery may choose to insert padding of oddly-sized
342 : * requests, we must estimate each chunk separately.
343 : *
344 : * We need one key to register the location of the header, and two other
345 : * keys to track the locations of the message queue and the error message
346 : * queue.
347 : */
348 20 : shm_toc_initialize_estimator(&e);
349 20 : shm_toc_estimate_chunk(&e, sizeof(ParallelApplyWorkerShared));
350 20 : shm_toc_estimate_chunk(&e, queue_size);
351 20 : shm_toc_estimate_chunk(&e, error_queue_size);
352 :
353 20 : shm_toc_estimate_keys(&e, 3);
354 20 : segsize = shm_toc_estimate(&e);
355 :
356 : /* Create the shared memory segment and establish a table of contents. */
357 20 : seg = dsm_create(shm_toc_estimate(&e), 0);
358 20 : if (!seg)
359 0 : return false;
360 :
361 20 : toc = shm_toc_create(PG_LOGICAL_APPLY_SHM_MAGIC, dsm_segment_address(seg),
362 : segsize);
363 :
364 : /* Set up the header region. */
365 20 : shared = shm_toc_allocate(toc, sizeof(ParallelApplyWorkerShared));
366 20 : SpinLockInit(&shared->mutex);
367 :
368 20 : shared->xact_state = PARALLEL_TRANS_UNKNOWN;
369 20 : pg_atomic_init_u32(&(shared->pending_stream_count), 0);
370 20 : shared->last_commit_end = InvalidXLogRecPtr;
371 20 : shared->fileset_state = FS_EMPTY;
372 :
373 20 : shm_toc_insert(toc, PARALLEL_APPLY_KEY_SHARED, shared);
374 :
375 : /* Set up message queue for the worker. */
376 20 : mq = shm_mq_create(shm_toc_allocate(toc, queue_size), queue_size);
377 20 : shm_toc_insert(toc, PARALLEL_APPLY_KEY_MQ, mq);
378 20 : shm_mq_set_sender(mq, MyProc);
379 :
380 : /* Attach the queue. */
381 20 : winfo->mq_handle = shm_mq_attach(mq, seg, NULL);
382 :
383 : /* Set up error queue for the worker. */
384 20 : mq = shm_mq_create(shm_toc_allocate(toc, error_queue_size),
385 : error_queue_size);
386 20 : shm_toc_insert(toc, PARALLEL_APPLY_KEY_ERROR_QUEUE, mq);
387 20 : shm_mq_set_receiver(mq, MyProc);
388 :
389 : /* Attach the queue. */
390 20 : winfo->error_mq_handle = shm_mq_attach(mq, seg, NULL);
391 :
392 : /* Return results to caller. */
393 20 : winfo->dsm_seg = seg;
394 20 : winfo->shared = shared;
395 :
396 20 : return true;
397 : }
398 :
399 : /*
400 : * Try to get a parallel apply worker from the pool. If none is available then
401 : * start a new one.
402 : */
403 : static ParallelApplyWorkerInfo *
404 54 : pa_launch_parallel_worker(void)
405 : {
406 : MemoryContext oldcontext;
407 : bool launched;
408 : ParallelApplyWorkerInfo *winfo;
409 : ListCell *lc;
410 :
411 : /* Try to get an available parallel apply worker from the worker pool. */
412 58 : foreach(lc, ParallelApplyWorkerPool)
413 : {
414 38 : winfo = (ParallelApplyWorkerInfo *) lfirst(lc);
415 :
416 38 : if (!winfo->in_use)
417 34 : return winfo;
418 : }
419 :
420 : /*
421 : * Start a new parallel apply worker.
422 : *
423 : * The worker info can be used for the lifetime of the worker process, so
424 : * create it in a permanent context.
425 : */
426 20 : oldcontext = MemoryContextSwitchTo(ApplyContext);
427 :
428 20 : winfo = (ParallelApplyWorkerInfo *) palloc0(sizeof(ParallelApplyWorkerInfo));
429 :
430 : /* Setup shared memory. */
431 20 : if (!pa_setup_dsm(winfo))
432 : {
433 0 : MemoryContextSwitchTo(oldcontext);
434 0 : pfree(winfo);
435 0 : return NULL;
436 : }
437 :
438 20 : launched = logicalrep_worker_launch(WORKERTYPE_PARALLEL_APPLY,
439 20 : MyLogicalRepWorker->dbid,
440 20 : MySubscription->oid,
441 20 : MySubscription->name,
442 20 : MyLogicalRepWorker->userid,
443 : InvalidOid,
444 : dsm_segment_handle(winfo->dsm_seg),
445 : false);
446 :
447 20 : if (launched)
448 : {
449 20 : ParallelApplyWorkerPool = lappend(ParallelApplyWorkerPool, winfo);
450 : }
451 : else
452 : {
453 0 : pa_free_worker_info(winfo);
454 0 : winfo = NULL;
455 : }
456 :
457 20 : MemoryContextSwitchTo(oldcontext);
458 :
459 20 : return winfo;
460 : }
461 :
462 : /*
463 : * Allocate a parallel apply worker that will be used for the specified xid.
464 : *
465 : * We first try to get an available worker from the pool, if any and then try
466 : * to launch a new worker. On successful allocation, remember the worker
467 : * information in the hash table so that we can get it later for processing the
468 : * streaming changes.
469 : */
470 : void
471 164 : pa_allocate_worker(TransactionId xid)
472 : {
473 : bool found;
474 164 : ParallelApplyWorkerInfo *winfo = NULL;
475 : ParallelApplyWorkerEntry *entry;
476 :
477 164 : if (!pa_can_start())
478 110 : return;
479 :
480 54 : winfo = pa_launch_parallel_worker();
481 54 : if (!winfo)
482 0 : return;
483 :
484 : /* First time through, initialize parallel apply worker state hashtable. */
485 54 : if (!ParallelApplyTxnHash)
486 : {
487 : HASHCTL ctl;
488 :
489 182 : MemSet(&ctl, 0, sizeof(ctl));
490 14 : ctl.keysize = sizeof(TransactionId);
491 14 : ctl.entrysize = sizeof(ParallelApplyWorkerEntry);
492 14 : ctl.hcxt = ApplyContext;
493 :
494 14 : ParallelApplyTxnHash = hash_create("logical replication parallel apply workers hash",
495 : 16, &ctl,
496 : HASH_ELEM | HASH_BLOBS | HASH_CONTEXT);
497 : }
498 :
499 : /* Create an entry for the requested transaction. */
500 54 : entry = hash_search(ParallelApplyTxnHash, &xid, HASH_ENTER, &found);
501 54 : if (found)
502 0 : elog(ERROR, "hash table corrupted");
503 :
504 : /* Update the transaction information in shared memory. */
505 54 : SpinLockAcquire(&winfo->shared->mutex);
506 54 : winfo->shared->xact_state = PARALLEL_TRANS_UNKNOWN;
507 54 : winfo->shared->xid = xid;
508 54 : SpinLockRelease(&winfo->shared->mutex);
509 :
510 54 : winfo->in_use = true;
511 54 : winfo->serialize_changes = false;
512 54 : entry->winfo = winfo;
513 : }
514 :
515 : /*
516 : * Find the assigned worker for the given transaction, if any.
517 : */
518 : ParallelApplyWorkerInfo *
519 520652 : pa_find_worker(TransactionId xid)
520 : {
521 : bool found;
522 : ParallelApplyWorkerEntry *entry;
523 :
524 520652 : if (!TransactionIdIsValid(xid))
525 166278 : return NULL;
526 :
527 354374 : if (!ParallelApplyTxnHash)
528 206476 : return NULL;
529 :
530 : /* Return the cached parallel apply worker if valid. */
531 147898 : if (stream_apply_worker)
532 147306 : return stream_apply_worker;
533 :
534 : /* Find an entry for the requested transaction. */
535 592 : entry = hash_search(ParallelApplyTxnHash, &xid, HASH_FIND, &found);
536 592 : if (found)
537 : {
538 : /* The worker must not have exited. */
539 : Assert(entry->winfo->in_use);
540 592 : return entry->winfo;
541 : }
542 :
543 0 : return NULL;
544 : }
545 :
546 : /*
547 : * Makes the worker available for reuse.
548 : *
549 : * This removes the parallel apply worker entry from the hash table so that it
550 : * can't be used. If there are enough workers in the pool, it stops the worker
551 : * and frees the corresponding info. Otherwise it just marks the worker as
552 : * available for reuse.
553 : *
554 : * For more information about the worker pool, see comments atop this file.
555 : */
556 : static void
557 48 : pa_free_worker(ParallelApplyWorkerInfo *winfo)
558 : {
559 : Assert(!am_parallel_apply_worker());
560 : Assert(winfo->in_use);
561 : Assert(pa_get_xact_state(winfo->shared) == PARALLEL_TRANS_FINISHED);
562 :
563 48 : if (!hash_search(ParallelApplyTxnHash, &winfo->shared->xid, HASH_REMOVE, NULL))
564 0 : elog(ERROR, "hash table corrupted");
565 :
566 : /*
567 : * Stop the worker if there are enough workers in the pool.
568 : *
569 : * XXX Additionally, we also stop the worker if the leader apply worker
570 : * serialize part of the transaction data due to a send timeout. This is
571 : * because the message could be partially written to the queue and there
572 : * is no way to clean the queue other than resending the message until it
573 : * succeeds. Instead of trying to send the data which anyway would have
574 : * been serialized and then letting the parallel apply worker deal with
575 : * the spurious message, we stop the worker.
576 : */
577 48 : if (winfo->serialize_changes ||
578 40 : list_length(ParallelApplyWorkerPool) >
579 40 : (max_parallel_apply_workers_per_subscription / 2))
580 : {
581 10 : logicalrep_pa_worker_stop(winfo);
582 10 : pa_free_worker_info(winfo);
583 :
584 10 : return;
585 : }
586 :
587 38 : winfo->in_use = false;
588 38 : winfo->serialize_changes = false;
589 : }
590 :
591 : /*
592 : * Free the parallel apply worker information and unlink the files with
593 : * serialized changes if any.
594 : */
595 : static void
596 10 : pa_free_worker_info(ParallelApplyWorkerInfo *winfo)
597 : {
598 : Assert(winfo);
599 :
600 10 : if (winfo->mq_handle)
601 10 : shm_mq_detach(winfo->mq_handle);
602 :
603 10 : if (winfo->error_mq_handle)
604 0 : shm_mq_detach(winfo->error_mq_handle);
605 :
606 : /* Unlink the files with serialized changes. */
607 10 : if (winfo->serialize_changes)
608 8 : stream_cleanup_files(MyLogicalRepWorker->subid, winfo->shared->xid);
609 :
610 10 : if (winfo->dsm_seg)
611 10 : dsm_detach(winfo->dsm_seg);
612 :
613 : /* Remove from the worker pool. */
614 10 : ParallelApplyWorkerPool = list_delete_ptr(ParallelApplyWorkerPool, winfo);
615 :
616 10 : pfree(winfo);
617 10 : }
618 :
619 : /*
620 : * Detach the error queue for all parallel apply workers.
621 : */
622 : void
623 514 : pa_detach_all_error_mq(void)
624 : {
625 : ListCell *lc;
626 :
627 524 : foreach(lc, ParallelApplyWorkerPool)
628 : {
629 10 : ParallelApplyWorkerInfo *winfo = (ParallelApplyWorkerInfo *) lfirst(lc);
630 :
631 10 : if (winfo->error_mq_handle)
632 : {
633 10 : shm_mq_detach(winfo->error_mq_handle);
634 10 : winfo->error_mq_handle = NULL;
635 : }
636 : }
637 514 : }
638 :
639 : /*
640 : * Check if there are any pending spooled messages.
641 : */
642 : static bool
643 32 : pa_has_spooled_message_pending()
644 : {
645 : PartialFileSetState fileset_state;
646 :
647 32 : fileset_state = pa_get_fileset_state();
648 :
649 32 : return (fileset_state != FS_EMPTY);
650 : }
651 :
652 : /*
653 : * Replay the spooled messages once the leader apply worker has finished
654 : * serializing changes to the file.
655 : *
656 : * Returns false if there aren't any pending spooled messages, true otherwise.
657 : */
658 : static bool
659 110 : pa_process_spooled_messages_if_required(void)
660 : {
661 : PartialFileSetState fileset_state;
662 :
663 110 : fileset_state = pa_get_fileset_state();
664 :
665 110 : if (fileset_state == FS_EMPTY)
666 94 : return false;
667 :
668 : /*
669 : * If the leader apply worker is busy serializing the partial changes then
670 : * acquire the stream lock now and wait for the leader worker to finish
671 : * serializing the changes. Otherwise, the parallel apply worker won't get
672 : * a chance to receive a STREAM_STOP (and acquire the stream lock) until
673 : * the leader had serialized all changes which can lead to undetected
674 : * deadlock.
675 : *
676 : * Note that the fileset state can be FS_SERIALIZE_DONE once the leader
677 : * worker has finished serializing the changes.
678 : */
679 16 : if (fileset_state == FS_SERIALIZE_IN_PROGRESS)
680 : {
681 0 : pa_lock_stream(MyParallelShared->xid, AccessShareLock);
682 0 : pa_unlock_stream(MyParallelShared->xid, AccessShareLock);
683 :
684 0 : fileset_state = pa_get_fileset_state();
685 : }
686 :
687 : /*
688 : * We cannot read the file immediately after the leader has serialized all
689 : * changes to the file because there may still be messages in the memory
690 : * queue. We will apply all spooled messages the next time we call this
691 : * function and that will ensure there are no messages left in the memory
692 : * queue.
693 : */
694 16 : if (fileset_state == FS_SERIALIZE_DONE)
695 : {
696 8 : pa_set_fileset_state(MyParallelShared, FS_READY);
697 : }
698 8 : else if (fileset_state == FS_READY)
699 : {
700 8 : apply_spooled_messages(&MyParallelShared->fileset,
701 8 : MyParallelShared->xid,
702 : InvalidXLogRecPtr);
703 8 : pa_set_fileset_state(MyParallelShared, FS_EMPTY);
704 : }
705 :
706 16 : return true;
707 : }
708 :
709 : /*
710 : * Interrupt handler for main loop of parallel apply worker.
711 : */
712 : static void
713 127902 : ProcessParallelApplyInterrupts(void)
714 : {
715 127902 : CHECK_FOR_INTERRUPTS();
716 :
717 127896 : if (ShutdownRequestPending)
718 : {
719 10 : ereport(LOG,
720 : (errmsg("logical replication parallel apply worker for subscription \"%s\" has finished",
721 : MySubscription->name)));
722 :
723 10 : proc_exit(0);
724 : }
725 :
726 127886 : if (ConfigReloadPending)
727 : {
728 8 : ConfigReloadPending = false;
729 8 : ProcessConfigFile(PGC_SIGHUP);
730 : }
731 127886 : }
732 :
733 : /* Parallel apply worker main loop. */
734 : static void
735 20 : LogicalParallelApplyLoop(shm_mq_handle *mqh)
736 : {
737 : shm_mq_result shmq_res;
738 : ErrorContextCallback errcallback;
739 20 : MemoryContext oldcxt = CurrentMemoryContext;
740 :
741 : /*
742 : * Init the ApplyMessageContext which we clean up after each replication
743 : * protocol message.
744 : */
745 20 : ApplyMessageContext = AllocSetContextCreate(ApplyContext,
746 : "ApplyMessageContext",
747 : ALLOCSET_DEFAULT_SIZES);
748 :
749 : /*
750 : * Push apply error context callback. Fields will be filled while applying
751 : * a change.
752 : */
753 20 : errcallback.callback = apply_error_callback;
754 20 : errcallback.previous = error_context_stack;
755 20 : error_context_stack = &errcallback;
756 :
757 : for (;;)
758 127882 : {
759 : void *data;
760 : Size len;
761 :
762 127902 : ProcessParallelApplyInterrupts();
763 :
764 : /* Ensure we are reading the data into our memory context. */
765 127886 : MemoryContextSwitchTo(ApplyMessageContext);
766 :
767 127886 : shmq_res = shm_mq_receive(mqh, &len, &data, true);
768 :
769 127886 : if (shmq_res == SHM_MQ_SUCCESS)
770 : {
771 : StringInfoData s;
772 : int c;
773 :
774 127776 : if (len == 0)
775 0 : elog(ERROR, "invalid message length");
776 :
777 127776 : initReadOnlyStringInfo(&s, data, len);
778 :
779 : /*
780 : * The first byte of messages sent from leader apply worker to
781 : * parallel apply workers can only be 'w'.
782 : */
783 127776 : c = pq_getmsgbyte(&s);
784 127776 : if (c != 'w')
785 0 : elog(ERROR, "unexpected message \"%c\"", c);
786 :
787 : /*
788 : * Ignore statistics fields that have been updated by the leader
789 : * apply worker.
790 : *
791 : * XXX We can avoid sending the statistics fields from the leader
792 : * apply worker but for that, it needs to rebuild the entire
793 : * message by removing these fields which could be more work than
794 : * simply ignoring these fields in the parallel apply worker.
795 : */
796 127776 : s.cursor += SIZE_STATS_MESSAGE;
797 :
798 127776 : apply_dispatch(&s);
799 : }
800 110 : else if (shmq_res == SHM_MQ_WOULD_BLOCK)
801 : {
802 : /* Replay the changes from the file, if any. */
803 110 : if (!pa_process_spooled_messages_if_required())
804 : {
805 : int rc;
806 :
807 : /* Wait for more work. */
808 94 : rc = WaitLatch(MyLatch,
809 : WL_LATCH_SET | WL_TIMEOUT | WL_EXIT_ON_PM_DEATH,
810 : 1000L,
811 : WAIT_EVENT_LOGICAL_PARALLEL_APPLY_MAIN);
812 :
813 94 : if (rc & WL_LATCH_SET)
814 84 : ResetLatch(MyLatch);
815 : }
816 : }
817 : else
818 : {
819 : Assert(shmq_res == SHM_MQ_DETACHED);
820 :
821 0 : ereport(ERROR,
822 : (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
823 : errmsg("lost connection to the logical replication apply worker")));
824 : }
825 :
826 127882 : MemoryContextReset(ApplyMessageContext);
827 127882 : MemoryContextSwitchTo(oldcxt);
828 : }
829 :
830 : /* Pop the error context stack. */
831 : error_context_stack = errcallback.previous;
832 :
833 : MemoryContextSwitchTo(oldcxt);
834 : }
835 :
836 : /*
837 : * Make sure the leader apply worker tries to read from our error queue one more
838 : * time. This guards against the case where we exit uncleanly without sending
839 : * an ErrorResponse, for example because some code calls proc_exit directly.
840 : *
841 : * Also explicitly detach from dsm segment to invoke on_dsm_detach callbacks,
842 : * if any. See ParallelWorkerShutdown for details.
843 : */
844 : static void
845 20 : pa_shutdown(int code, Datum arg)
846 : {
847 20 : SendProcSignal(MyLogicalRepWorker->leader_pid,
848 : PROCSIG_PARALLEL_APPLY_MESSAGE,
849 : INVALID_PROC_NUMBER);
850 :
851 20 : dsm_detach((dsm_segment *) DatumGetPointer(arg));
852 20 : }
853 :
854 : /*
855 : * Parallel apply worker entry point.
856 : */
857 : void
858 20 : ParallelApplyWorkerMain(Datum main_arg)
859 : {
860 : ParallelApplyWorkerShared *shared;
861 : dsm_handle handle;
862 : dsm_segment *seg;
863 : shm_toc *toc;
864 : shm_mq *mq;
865 : shm_mq_handle *mqh;
866 : shm_mq_handle *error_mqh;
867 : RepOriginId originid;
868 20 : int worker_slot = DatumGetInt32(main_arg);
869 : char originname[NAMEDATALEN];
870 :
871 20 : InitializingApplyWorker = true;
872 :
873 : /* Setup signal handling. */
874 20 : pqsignal(SIGHUP, SignalHandlerForConfigReload);
875 20 : pqsignal(SIGINT, SignalHandlerForShutdownRequest);
876 20 : pqsignal(SIGTERM, die);
877 20 : BackgroundWorkerUnblockSignals();
878 :
879 : /*
880 : * Attach to the dynamic shared memory segment for the parallel apply, and
881 : * find its table of contents.
882 : *
883 : * Like parallel query, we don't need resource owner by this time. See
884 : * ParallelWorkerMain.
885 : */
886 20 : memcpy(&handle, MyBgworkerEntry->bgw_extra, sizeof(dsm_handle));
887 20 : seg = dsm_attach(handle);
888 20 : if (!seg)
889 0 : ereport(ERROR,
890 : (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
891 : errmsg("could not map dynamic shared memory segment")));
892 :
893 20 : toc = shm_toc_attach(PG_LOGICAL_APPLY_SHM_MAGIC, dsm_segment_address(seg));
894 20 : if (!toc)
895 0 : ereport(ERROR,
896 : (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
897 : errmsg("invalid magic number in dynamic shared memory segment")));
898 :
899 : /* Look up the shared information. */
900 20 : shared = shm_toc_lookup(toc, PARALLEL_APPLY_KEY_SHARED, false);
901 20 : MyParallelShared = shared;
902 :
903 : /*
904 : * Attach to the message queue.
905 : */
906 20 : mq = shm_toc_lookup(toc, PARALLEL_APPLY_KEY_MQ, false);
907 20 : shm_mq_set_receiver(mq, MyProc);
908 20 : mqh = shm_mq_attach(mq, seg, NULL);
909 :
910 : /*
911 : * Primary initialization is complete. Now, we can attach to our slot.
912 : * This is to ensure that the leader apply worker does not write data to
913 : * the uninitialized memory queue.
914 : */
915 20 : logicalrep_worker_attach(worker_slot);
916 :
917 : /*
918 : * Register the shutdown callback after we are attached to the worker
919 : * slot. This is to ensure that MyLogicalRepWorker remains valid when this
920 : * callback is invoked.
921 : */
922 20 : before_shmem_exit(pa_shutdown, PointerGetDatum(seg));
923 :
924 20 : SpinLockAcquire(&MyParallelShared->mutex);
925 20 : MyParallelShared->logicalrep_worker_generation = MyLogicalRepWorker->generation;
926 20 : MyParallelShared->logicalrep_worker_slot_no = worker_slot;
927 20 : SpinLockRelease(&MyParallelShared->mutex);
928 :
929 : /*
930 : * Attach to the error queue.
931 : */
932 20 : mq = shm_toc_lookup(toc, PARALLEL_APPLY_KEY_ERROR_QUEUE, false);
933 20 : shm_mq_set_sender(mq, MyProc);
934 20 : error_mqh = shm_mq_attach(mq, seg, NULL);
935 :
936 20 : pq_redirect_to_shm_mq(seg, error_mqh);
937 20 : pq_set_parallel_leader(MyLogicalRepWorker->leader_pid,
938 : INVALID_PROC_NUMBER);
939 :
940 20 : MyLogicalRepWorker->last_send_time = MyLogicalRepWorker->last_recv_time =
941 20 : MyLogicalRepWorker->reply_time = 0;
942 :
943 20 : InitializeLogRepWorker();
944 :
945 20 : InitializingApplyWorker = false;
946 :
947 : /* Setup replication origin tracking. */
948 20 : StartTransactionCommand();
949 20 : ReplicationOriginNameForLogicalRep(MySubscription->oid, InvalidOid,
950 : originname, sizeof(originname));
951 20 : originid = replorigin_by_name(originname, false);
952 :
953 : /*
954 : * The parallel apply worker doesn't need to monopolize this replication
955 : * origin which was already acquired by its leader process.
956 : */
957 20 : replorigin_session_setup(originid, MyLogicalRepWorker->leader_pid);
958 20 : replorigin_session_origin = originid;
959 20 : CommitTransactionCommand();
960 :
961 : /*
962 : * Setup callback for syscache so that we know when something changes in
963 : * the subscription relation state.
964 : */
965 20 : CacheRegisterSyscacheCallback(SUBSCRIPTIONRELMAP,
966 : invalidate_syncing_table_states,
967 : (Datum) 0);
968 :
969 20 : set_apply_error_context_origin(originname);
970 :
971 20 : LogicalParallelApplyLoop(mqh);
972 :
973 : /*
974 : * The parallel apply worker must not get here because the parallel apply
975 : * worker will only stop when it receives a SIGTERM or SIGINT from the
976 : * leader, or when there is an error. None of these cases will allow the
977 : * code to reach here.
978 : */
979 : Assert(false);
980 0 : }
981 :
982 : /*
983 : * Handle receipt of an interrupt indicating a parallel apply worker message.
984 : *
985 : * Note: this is called within a signal handler! All we can do is set a flag
986 : * that will cause the next CHECK_FOR_INTERRUPTS() to invoke
987 : * ProcessParallelApplyMessages().
988 : */
989 : void
990 22 : HandleParallelApplyMessageInterrupt(void)
991 : {
992 22 : InterruptPending = true;
993 22 : ParallelApplyMessagePending = true;
994 22 : SetLatch(MyLatch);
995 22 : }
996 :
997 : /*
998 : * Process a single protocol message received from a single parallel apply
999 : * worker.
1000 : */
1001 : static void
1002 2 : ProcessParallelApplyMessage(StringInfo msg)
1003 : {
1004 : char msgtype;
1005 :
1006 2 : msgtype = pq_getmsgbyte(msg);
1007 :
1008 2 : switch (msgtype)
1009 : {
1010 2 : case 'E': /* ErrorResponse */
1011 : {
1012 : ErrorData edata;
1013 :
1014 : /* Parse ErrorResponse. */
1015 2 : pq_parse_errornotice(msg, &edata);
1016 :
1017 : /*
1018 : * If desired, add a context line to show that this is a
1019 : * message propagated from a parallel apply worker. Otherwise,
1020 : * it can sometimes be confusing to understand what actually
1021 : * happened.
1022 : */
1023 2 : if (edata.context)
1024 2 : edata.context = psprintf("%s\n%s", edata.context,
1025 : _("logical replication parallel apply worker"));
1026 : else
1027 0 : edata.context = pstrdup(_("logical replication parallel apply worker"));
1028 :
1029 : /*
1030 : * Context beyond that should use the error context callbacks
1031 : * that were in effect in LogicalRepApplyLoop().
1032 : */
1033 2 : error_context_stack = apply_error_context_stack;
1034 :
1035 : /*
1036 : * The actual error must have been reported by the parallel
1037 : * apply worker.
1038 : */
1039 2 : ereport(ERROR,
1040 : (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
1041 : errmsg("logical replication parallel apply worker exited due to error"),
1042 : errcontext("%s", edata.context)));
1043 : }
1044 :
1045 : /*
1046 : * Don't need to do anything about NoticeResponse and
1047 : * NotifyResponse as the logical replication worker doesn't need
1048 : * to send messages to the client.
1049 : */
1050 0 : case 'N':
1051 : case 'A':
1052 0 : break;
1053 :
1054 0 : default:
1055 0 : elog(ERROR, "unrecognized message type received from logical replication parallel apply worker: %c (message length %d bytes)",
1056 : msgtype, msg->len);
1057 : }
1058 0 : }
1059 :
1060 : /*
1061 : * Handle any queued protocol messages received from parallel apply workers.
1062 : */
1063 : void
1064 12 : ProcessParallelApplyMessages(void)
1065 : {
1066 : ListCell *lc;
1067 : MemoryContext oldcontext;
1068 :
1069 : static MemoryContext hpam_context = NULL;
1070 :
1071 : /*
1072 : * This is invoked from ProcessInterrupts(), and since some of the
1073 : * functions it calls contain CHECK_FOR_INTERRUPTS(), there is a potential
1074 : * for recursive calls if more signals are received while this runs. It's
1075 : * unclear that recursive entry would be safe, and it doesn't seem useful
1076 : * even if it is safe, so let's block interrupts until done.
1077 : */
1078 12 : HOLD_INTERRUPTS();
1079 :
1080 : /*
1081 : * Moreover, CurrentMemoryContext might be pointing almost anywhere. We
1082 : * don't want to risk leaking data into long-lived contexts, so let's do
1083 : * our work here in a private context that we can reset on each use.
1084 : */
1085 12 : if (!hpam_context) /* first time through? */
1086 10 : hpam_context = AllocSetContextCreate(TopMemoryContext,
1087 : "ProcessParallelApplyMessages",
1088 : ALLOCSET_DEFAULT_SIZES);
1089 : else
1090 2 : MemoryContextReset(hpam_context);
1091 :
1092 12 : oldcontext = MemoryContextSwitchTo(hpam_context);
1093 :
1094 12 : ParallelApplyMessagePending = false;
1095 :
1096 24 : foreach(lc, ParallelApplyWorkerPool)
1097 : {
1098 : shm_mq_result res;
1099 : Size nbytes;
1100 : void *data;
1101 14 : ParallelApplyWorkerInfo *winfo = (ParallelApplyWorkerInfo *) lfirst(lc);
1102 :
1103 : /*
1104 : * The leader will detach from the error queue and set it to NULL
1105 : * before preparing to stop all parallel apply workers, so we don't
1106 : * need to handle error messages anymore. See
1107 : * logicalrep_worker_detach.
1108 : */
1109 14 : if (!winfo->error_mq_handle)
1110 12 : continue;
1111 :
1112 4 : res = shm_mq_receive(winfo->error_mq_handle, &nbytes, &data, true);
1113 :
1114 4 : if (res == SHM_MQ_WOULD_BLOCK)
1115 2 : continue;
1116 2 : else if (res == SHM_MQ_SUCCESS)
1117 : {
1118 : StringInfoData msg;
1119 :
1120 2 : initStringInfo(&msg);
1121 2 : appendBinaryStringInfo(&msg, data, nbytes);
1122 2 : ProcessParallelApplyMessage(&msg);
1123 0 : pfree(msg.data);
1124 : }
1125 : else
1126 0 : ereport(ERROR,
1127 : (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
1128 : errmsg("lost connection to the logical replication parallel apply worker")));
1129 : }
1130 :
1131 10 : MemoryContextSwitchTo(oldcontext);
1132 :
1133 : /* Might as well clear the context on our way out */
1134 10 : MemoryContextReset(hpam_context);
1135 :
1136 10 : RESUME_INTERRUPTS();
1137 10 : }
1138 :
1139 : /*
1140 : * Send the data to the specified parallel apply worker via shared-memory
1141 : * queue.
1142 : *
1143 : * Returns false if the attempt to send data via shared memory times out, true
1144 : * otherwise.
1145 : */
1146 : bool
1147 137824 : pa_send_data(ParallelApplyWorkerInfo *winfo, Size nbytes, const void *data)
1148 : {
1149 : int rc;
1150 : shm_mq_result result;
1151 137824 : TimestampTz startTime = 0;
1152 :
1153 : Assert(!IsTransactionState());
1154 : Assert(!winfo->serialize_changes);
1155 :
1156 : /*
1157 : * We don't try to send data to parallel worker for 'immediate' mode. This
1158 : * is primarily used for testing purposes.
1159 : */
1160 137824 : if (unlikely(debug_logical_replication_streaming == DEBUG_LOGICAL_REP_STREAMING_IMMEDIATE))
1161 8 : return false;
1162 :
1163 : /*
1164 : * This timeout is a bit arbitrary but testing revealed that it is sufficient
1165 : * to send the message unless the parallel apply worker is waiting on some
1166 : * lock or there is a serious resource crunch. See the comments atop this file
1167 : * to know why we are using a non-blocking way to send the message.
1168 : */
1169 : #define SHM_SEND_RETRY_INTERVAL_MS 1000
1170 : #define SHM_SEND_TIMEOUT_MS (10000 - SHM_SEND_RETRY_INTERVAL_MS)
1171 :
1172 : for (;;)
1173 : {
1174 137816 : result = shm_mq_send(winfo->mq_handle, nbytes, data, true, true);
1175 :
1176 137816 : if (result == SHM_MQ_SUCCESS)
1177 137816 : return true;
1178 0 : else if (result == SHM_MQ_DETACHED)
1179 0 : ereport(ERROR,
1180 : (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
1181 : errmsg("could not send data to shared-memory queue")));
1182 :
1183 : Assert(result == SHM_MQ_WOULD_BLOCK);
1184 :
1185 : /* Wait before retrying. */
1186 0 : rc = WaitLatch(MyLatch,
1187 : WL_LATCH_SET | WL_TIMEOUT | WL_EXIT_ON_PM_DEATH,
1188 : SHM_SEND_RETRY_INTERVAL_MS,
1189 : WAIT_EVENT_LOGICAL_APPLY_SEND_DATA);
1190 :
1191 0 : if (rc & WL_LATCH_SET)
1192 : {
1193 0 : ResetLatch(MyLatch);
1194 0 : CHECK_FOR_INTERRUPTS();
1195 : }
1196 :
1197 0 : if (startTime == 0)
1198 0 : startTime = GetCurrentTimestamp();
1199 0 : else if (TimestampDifferenceExceeds(startTime, GetCurrentTimestamp(),
1200 : SHM_SEND_TIMEOUT_MS))
1201 0 : return false;
1202 : }
1203 : }
1204 :
1205 : /*
1206 : * Switch to PARTIAL_SERIALIZE mode for the current transaction -- this means
1207 : * that the current data and any subsequent data for this transaction will be
1208 : * serialized to a file. This is done to prevent possible deadlocks with
1209 : * another parallel apply worker (refer to the comments atop this file).
1210 : */
1211 : void
1212 8 : pa_switch_to_partial_serialize(ParallelApplyWorkerInfo *winfo,
1213 : bool stream_locked)
1214 : {
1215 8 : ereport(LOG,
1216 : (errmsg("logical replication apply worker will serialize the remaining changes of remote transaction %u to a file",
1217 : winfo->shared->xid)));
1218 :
1219 : /*
1220 : * The parallel apply worker could be stuck for some reason (say waiting
1221 : * on some lock by other backend), so stop trying to send data directly to
1222 : * it and start serializing data to the file instead.
1223 : */
1224 8 : winfo->serialize_changes = true;
1225 :
1226 : /* Initialize the stream fileset. */
1227 8 : stream_start_internal(winfo->shared->xid, true);
1228 :
1229 : /*
1230 : * Acquires the stream lock if not already to make sure that the parallel
1231 : * apply worker will wait for the leader to release the stream lock until
1232 : * the end of the transaction.
1233 : */
1234 8 : if (!stream_locked)
1235 8 : pa_lock_stream(winfo->shared->xid, AccessExclusiveLock);
1236 :
1237 8 : pa_set_fileset_state(winfo->shared, FS_SERIALIZE_IN_PROGRESS);
1238 8 : }
1239 :
1240 : /*
1241 : * Wait until the parallel apply worker's transaction state has reached or
1242 : * exceeded the given xact_state.
1243 : */
1244 : static void
1245 50 : pa_wait_for_xact_state(ParallelApplyWorkerInfo *winfo,
1246 : ParallelTransState xact_state)
1247 : {
1248 : for (;;)
1249 : {
1250 : /*
1251 : * Stop if the transaction state has reached or exceeded the given
1252 : * xact_state.
1253 : */
1254 512 : if (pa_get_xact_state(winfo->shared) >= xact_state)
1255 50 : break;
1256 :
1257 : /* Wait to be signalled. */
1258 462 : (void) WaitLatch(MyLatch,
1259 : WL_LATCH_SET | WL_TIMEOUT | WL_EXIT_ON_PM_DEATH,
1260 : 10L,
1261 : WAIT_EVENT_LOGICAL_PARALLEL_APPLY_STATE_CHANGE);
1262 :
1263 : /* Reset the latch so we don't spin. */
1264 462 : ResetLatch(MyLatch);
1265 :
1266 : /* An interrupt may have occurred while we were waiting. */
1267 462 : CHECK_FOR_INTERRUPTS();
1268 : }
1269 50 : }
1270 :
1271 : /*
1272 : * Wait until the parallel apply worker's transaction finishes.
1273 : */
1274 : static void
1275 50 : pa_wait_for_xact_finish(ParallelApplyWorkerInfo *winfo)
1276 : {
1277 : /*
1278 : * Wait until the parallel apply worker set the state to
1279 : * PARALLEL_TRANS_STARTED which means it has acquired the transaction
1280 : * lock. This is to prevent leader apply worker from acquiring the
1281 : * transaction lock earlier than the parallel apply worker.
1282 : */
1283 50 : pa_wait_for_xact_state(winfo, PARALLEL_TRANS_STARTED);
1284 :
1285 : /*
1286 : * Wait for the transaction lock to be released. This is required to
1287 : * detect deadlock among leader and parallel apply workers. Refer to the
1288 : * comments atop this file.
1289 : */
1290 50 : pa_lock_transaction(winfo->shared->xid, AccessShareLock);
1291 48 : pa_unlock_transaction(winfo->shared->xid, AccessShareLock);
1292 :
1293 : /*
1294 : * Check if the state becomes PARALLEL_TRANS_FINISHED in case the parallel
1295 : * apply worker failed while applying changes causing the lock to be
1296 : * released.
1297 : */
1298 48 : if (pa_get_xact_state(winfo->shared) != PARALLEL_TRANS_FINISHED)
1299 0 : ereport(ERROR,
1300 : (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
1301 : errmsg("lost connection to the logical replication parallel apply worker")));
1302 48 : }
1303 :
1304 : /*
1305 : * Set the transaction state for a given parallel apply worker.
1306 : */
1307 : void
1308 102 : pa_set_xact_state(ParallelApplyWorkerShared *wshared,
1309 : ParallelTransState xact_state)
1310 : {
1311 102 : SpinLockAcquire(&wshared->mutex);
1312 102 : wshared->xact_state = xact_state;
1313 102 : SpinLockRelease(&wshared->mutex);
1314 102 : }
1315 :
1316 : /*
1317 : * Get the transaction state for a given parallel apply worker.
1318 : */
1319 : static ParallelTransState
1320 560 : pa_get_xact_state(ParallelApplyWorkerShared *wshared)
1321 : {
1322 : ParallelTransState xact_state;
1323 :
1324 560 : SpinLockAcquire(&wshared->mutex);
1325 560 : xact_state = wshared->xact_state;
1326 560 : SpinLockRelease(&wshared->mutex);
1327 :
1328 560 : return xact_state;
1329 : }
1330 :
1331 : /*
1332 : * Cache the parallel apply worker information.
1333 : */
1334 : void
1335 1044 : pa_set_stream_apply_worker(ParallelApplyWorkerInfo *winfo)
1336 : {
1337 1044 : stream_apply_worker = winfo;
1338 1044 : }
1339 :
1340 : /*
1341 : * Form a unique savepoint name for the streaming transaction.
1342 : *
1343 : * Note that different subscriptions for publications on different nodes can
1344 : * receive same remote xid, so we need to use subscription id along with it.
1345 : *
1346 : * Returns the name in the supplied buffer.
1347 : */
1348 : static void
1349 54 : pa_savepoint_name(Oid suboid, TransactionId xid, char *spname, Size szsp)
1350 : {
1351 54 : snprintf(spname, szsp, "pg_sp_%u_%u", suboid, xid);
1352 54 : }
1353 :
1354 : /*
1355 : * Define a savepoint for a subxact in parallel apply worker if needed.
1356 : *
1357 : * The parallel apply worker can figure out if a new subtransaction was
1358 : * started by checking if the new change arrived with a different xid. In that
1359 : * case define a named savepoint, so that we are able to rollback to it
1360 : * if required.
1361 : */
1362 : void
1363 136788 : pa_start_subtrans(TransactionId current_xid, TransactionId top_xid)
1364 : {
1365 136788 : if (current_xid != top_xid &&
1366 104 : !list_member_xid(subxactlist, current_xid))
1367 : {
1368 : MemoryContext oldctx;
1369 : char spname[NAMEDATALEN];
1370 :
1371 34 : pa_savepoint_name(MySubscription->oid, current_xid,
1372 : spname, sizeof(spname));
1373 :
1374 34 : elog(DEBUG1, "defining savepoint %s in logical replication parallel apply worker", spname);
1375 :
1376 : /* We must be in transaction block to define the SAVEPOINT. */
1377 34 : if (!IsTransactionBlock())
1378 : {
1379 10 : if (!IsTransactionState())
1380 0 : StartTransactionCommand();
1381 :
1382 10 : BeginTransactionBlock();
1383 10 : CommitTransactionCommand();
1384 : }
1385 :
1386 34 : DefineSavepoint(spname);
1387 :
1388 : /*
1389 : * CommitTransactionCommand is needed to start a subtransaction after
1390 : * issuing a SAVEPOINT inside a transaction block (see
1391 : * StartSubTransaction()).
1392 : */
1393 34 : CommitTransactionCommand();
1394 :
1395 34 : oldctx = MemoryContextSwitchTo(TopTransactionContext);
1396 34 : subxactlist = lappend_xid(subxactlist, current_xid);
1397 34 : MemoryContextSwitchTo(oldctx);
1398 : }
1399 136788 : }
1400 :
1401 : /* Reset the list that maintains subtransactions. */
1402 : void
1403 48 : pa_reset_subtrans(void)
1404 : {
1405 : /*
1406 : * We don't need to free this explicitly as the allocated memory will be
1407 : * freed at the transaction end.
1408 : */
1409 48 : subxactlist = NIL;
1410 48 : }
1411 :
1412 : /*
1413 : * Handle STREAM ABORT message when the transaction was applied in a parallel
1414 : * apply worker.
1415 : */
1416 : void
1417 24 : pa_stream_abort(LogicalRepStreamAbortData *abort_data)
1418 : {
1419 24 : TransactionId xid = abort_data->xid;
1420 24 : TransactionId subxid = abort_data->subxid;
1421 :
1422 : /*
1423 : * Update origin state so we can restart streaming from correct position
1424 : * in case of crash.
1425 : */
1426 24 : replorigin_session_origin_lsn = abort_data->abort_lsn;
1427 24 : replorigin_session_origin_timestamp = abort_data->abort_time;
1428 :
1429 : /*
1430 : * If the two XIDs are the same, it's in fact abort of toplevel xact, so
1431 : * just free the subxactlist.
1432 : */
1433 24 : if (subxid == xid)
1434 : {
1435 4 : pa_set_xact_state(MyParallelShared, PARALLEL_TRANS_FINISHED);
1436 :
1437 : /*
1438 : * Release the lock as we might be processing an empty streaming
1439 : * transaction in which case the lock won't be released during
1440 : * transaction rollback.
1441 : *
1442 : * Note that it's ok to release the transaction lock before aborting
1443 : * the transaction because even if the parallel apply worker dies due
1444 : * to crash or some other reason, such a transaction would still be
1445 : * considered aborted.
1446 : */
1447 4 : pa_unlock_transaction(xid, AccessExclusiveLock);
1448 :
1449 4 : AbortCurrentTransaction();
1450 :
1451 4 : if (IsTransactionBlock())
1452 : {
1453 2 : EndTransactionBlock(false);
1454 2 : CommitTransactionCommand();
1455 : }
1456 :
1457 4 : pa_reset_subtrans();
1458 :
1459 4 : pgstat_report_activity(STATE_IDLE, NULL);
1460 : }
1461 : else
1462 : {
1463 : /* OK, so it's a subxact. Rollback to the savepoint. */
1464 : int i;
1465 : char spname[NAMEDATALEN];
1466 :
1467 20 : pa_savepoint_name(MySubscription->oid, subxid, spname, sizeof(spname));
1468 :
1469 20 : elog(DEBUG1, "rolling back to savepoint %s in logical replication parallel apply worker", spname);
1470 :
1471 : /*
1472 : * Search the subxactlist, determine the offset tracked for the
1473 : * subxact, and truncate the list.
1474 : *
1475 : * Note that for an empty sub-transaction we won't find the subxid
1476 : * here.
1477 : */
1478 24 : for (i = list_length(subxactlist) - 1; i >= 0; i--)
1479 : {
1480 22 : TransactionId xid_tmp = lfirst_xid(list_nth_cell(subxactlist, i));
1481 :
1482 22 : if (xid_tmp == subxid)
1483 : {
1484 18 : RollbackToSavepoint(spname);
1485 18 : CommitTransactionCommand();
1486 18 : subxactlist = list_truncate(subxactlist, i);
1487 18 : break;
1488 : }
1489 : }
1490 : }
1491 24 : }
1492 :
1493 : /*
1494 : * Set the fileset state for a particular parallel apply worker. The fileset
1495 : * will be set once the leader worker serialized all changes to the file
1496 : * so that it can be used by parallel apply worker.
1497 : */
1498 : void
1499 32 : pa_set_fileset_state(ParallelApplyWorkerShared *wshared,
1500 : PartialFileSetState fileset_state)
1501 : {
1502 32 : SpinLockAcquire(&wshared->mutex);
1503 32 : wshared->fileset_state = fileset_state;
1504 :
1505 32 : if (fileset_state == FS_SERIALIZE_DONE)
1506 : {
1507 : Assert(am_leader_apply_worker());
1508 : Assert(MyLogicalRepWorker->stream_fileset);
1509 8 : wshared->fileset = *MyLogicalRepWorker->stream_fileset;
1510 : }
1511 :
1512 32 : SpinLockRelease(&wshared->mutex);
1513 32 : }
1514 :
1515 : /*
1516 : * Get the fileset state for the current parallel apply worker.
1517 : */
1518 : static PartialFileSetState
1519 142 : pa_get_fileset_state(void)
1520 : {
1521 : PartialFileSetState fileset_state;
1522 :
1523 : Assert(am_parallel_apply_worker());
1524 :
1525 142 : SpinLockAcquire(&MyParallelShared->mutex);
1526 142 : fileset_state = MyParallelShared->fileset_state;
1527 142 : SpinLockRelease(&MyParallelShared->mutex);
1528 :
1529 142 : return fileset_state;
1530 : }
1531 :
1532 : /*
1533 : * Helper functions to acquire and release a lock for each stream block.
1534 : *
1535 : * Set locktag_field4 to PARALLEL_APPLY_LOCK_STREAM to indicate that it's a
1536 : * stream lock.
1537 : *
1538 : * Refer to the comments atop this file to see how the stream lock is used.
1539 : */
1540 : void
1541 566 : pa_lock_stream(TransactionId xid, LOCKMODE lockmode)
1542 : {
1543 566 : LockApplyTransactionForSession(MyLogicalRepWorker->subid, xid,
1544 : PARALLEL_APPLY_LOCK_STREAM, lockmode);
1545 562 : }
1546 :
1547 : void
1548 558 : pa_unlock_stream(TransactionId xid, LOCKMODE lockmode)
1549 : {
1550 558 : UnlockApplyTransactionForSession(MyLogicalRepWorker->subid, xid,
1551 : PARALLEL_APPLY_LOCK_STREAM, lockmode);
1552 558 : }
1553 :
1554 : /*
1555 : * Helper functions to acquire and release a lock for each local transaction
1556 : * apply.
1557 : *
1558 : * Set locktag_field4 to PARALLEL_APPLY_LOCK_XACT to indicate that it's a
1559 : * transaction lock.
1560 : *
1561 : * Note that all the callers must pass a remote transaction ID instead of a
1562 : * local transaction ID as xid. This is because the local transaction ID will
1563 : * only be assigned while applying the first change in the parallel apply but
1564 : * it's possible that the first change in the parallel apply worker is blocked
1565 : * by a concurrently executing transaction in another parallel apply worker. We
1566 : * can only communicate the local transaction id to the leader after applying
1567 : * the first change so it won't be able to wait after sending the xact finish
1568 : * command using this lock.
1569 : *
1570 : * Refer to the comments atop this file to see how the transaction lock is
1571 : * used.
1572 : */
1573 : void
1574 104 : pa_lock_transaction(TransactionId xid, LOCKMODE lockmode)
1575 : {
1576 104 : LockApplyTransactionForSession(MyLogicalRepWorker->subid, xid,
1577 : PARALLEL_APPLY_LOCK_XACT, lockmode);
1578 102 : }
1579 :
1580 : void
1581 96 : pa_unlock_transaction(TransactionId xid, LOCKMODE lockmode)
1582 : {
1583 96 : UnlockApplyTransactionForSession(MyLogicalRepWorker->subid, xid,
1584 : PARALLEL_APPLY_LOCK_XACT, lockmode);
1585 96 : }
1586 :
1587 : /*
1588 : * Decrement the number of pending streaming blocks and wait on the stream lock
1589 : * if there is no pending block available.
1590 : */
1591 : void
1592 520 : pa_decr_and_wait_stream_block(void)
1593 : {
1594 : Assert(am_parallel_apply_worker());
1595 :
1596 : /*
1597 : * It is only possible to not have any pending stream chunks when we are
1598 : * applying spooled messages.
1599 : */
1600 520 : if (pg_atomic_read_u32(&MyParallelShared->pending_stream_count) == 0)
1601 : {
1602 32 : if (pa_has_spooled_message_pending())
1603 32 : return;
1604 :
1605 0 : elog(ERROR, "invalid pending streaming chunk 0");
1606 : }
1607 :
1608 488 : if (pg_atomic_sub_fetch_u32(&MyParallelShared->pending_stream_count, 1) == 0)
1609 : {
1610 48 : pa_lock_stream(MyParallelShared->xid, AccessShareLock);
1611 44 : pa_unlock_stream(MyParallelShared->xid, AccessShareLock);
1612 : }
1613 : }
1614 :
1615 : /*
1616 : * Finish processing the streaming transaction in the leader apply worker.
1617 : */
1618 : void
1619 50 : pa_xact_finish(ParallelApplyWorkerInfo *winfo, XLogRecPtr remote_lsn)
1620 : {
1621 : Assert(am_leader_apply_worker());
1622 :
1623 : /*
1624 : * Unlock the shared object lock so that parallel apply worker can
1625 : * continue to receive and apply changes.
1626 : */
1627 50 : pa_unlock_stream(winfo->shared->xid, AccessExclusiveLock);
1628 :
1629 : /*
1630 : * Wait for that worker to finish. This is necessary to maintain commit
1631 : * order which avoids failures due to transaction dependencies and
1632 : * deadlocks.
1633 : */
1634 50 : pa_wait_for_xact_finish(winfo);
1635 :
1636 48 : if (!XLogRecPtrIsInvalid(remote_lsn))
1637 44 : store_flush_position(remote_lsn, winfo->shared->last_commit_end);
1638 :
1639 48 : pa_free_worker(winfo);
1640 48 : }
|
