Line data Source code
1 : /*-------------------------------------------------------------------------
2 : *
3 : * pg_createsubscriber.c
4 : * Create a new logical replica from a standby server
5 : *
6 : * Copyright (c) 2024-2025, PostgreSQL Global Development Group
7 : *
8 : * IDENTIFICATION
9 : * src/bin/pg_basebackup/pg_createsubscriber.c
10 : *
11 : *-------------------------------------------------------------------------
12 : */
13 :
14 : #include "postgres_fe.h"
15 :
16 : #include <sys/stat.h>
17 : #include <sys/time.h>
18 : #include <sys/wait.h>
19 : #include <time.h>
20 :
21 : #include "common/connect.h"
22 : #include "common/controldata_utils.h"
23 : #include "common/logging.h"
24 : #include "common/pg_prng.h"
25 : #include "common/restricted_token.h"
26 : #include "fe_utils/recovery_gen.h"
27 : #include "fe_utils/simple_list.h"
28 : #include "fe_utils/string_utils.h"
29 : #include "getopt_long.h"
30 :
31 : #define DEFAULT_SUB_PORT "50432"
32 : #define OBJECTTYPE_PUBLICATIONS 0x0001
33 :
34 : /* Command-line options */
35 : struct CreateSubscriberOptions
36 : {
37 : char *config_file; /* configuration file */
38 : char *pub_conninfo_str; /* publisher connection string */
39 : char *socket_dir; /* directory for Unix-domain socket, if any */
40 : char *sub_port; /* subscriber port number */
41 : const char *sub_username; /* subscriber username */
42 : bool two_phase; /* enable-two-phase option */
43 : SimpleStringList database_names; /* list of database names */
44 : SimpleStringList pub_names; /* list of publication names */
45 : SimpleStringList sub_names; /* list of subscription names */
46 : SimpleStringList replslot_names; /* list of replication slot names */
47 : int recovery_timeout; /* stop recovery after this time */
48 : bool all_dbs; /* all option */
49 : SimpleStringList objecttypes_to_clean; /* list of object types to cleanup */
50 : };
51 :
52 : /* per-database publication/subscription info */
53 : struct LogicalRepInfo
54 : {
55 : char *dbname; /* database name */
56 : char *pubconninfo; /* publisher connection string */
57 : char *subconninfo; /* subscriber connection string */
58 : char *pubname; /* publication name */
59 : char *subname; /* subscription name */
60 : char *replslotname; /* replication slot name */
61 :
62 : bool made_replslot; /* replication slot was created */
63 : bool made_publication; /* publication was created */
64 : };
65 :
66 : /*
67 : * Information shared across all the databases (or publications and
68 : * subscriptions).
69 : */
70 : struct LogicalRepInfos
71 : {
72 : struct LogicalRepInfo *dbinfo;
73 : bool two_phase; /* enable-two-phase option */
74 : bits32 objecttypes_to_clean; /* flags indicating which object types
75 : * to clean up on subscriber */
76 : };
77 :
78 : static void cleanup_objects_atexit(void);
79 : static void usage();
80 : static char *get_base_conninfo(const char *conninfo, char **dbname);
81 : static char *get_sub_conninfo(const struct CreateSubscriberOptions *opt);
82 : static char *get_exec_path(const char *argv0, const char *progname);
83 : static void check_data_directory(const char *datadir);
84 : static char *concat_conninfo_dbname(const char *conninfo, const char *dbname);
85 : static struct LogicalRepInfo *store_pub_sub_info(const struct CreateSubscriberOptions *opt,
86 : const char *pub_base_conninfo,
87 : const char *sub_base_conninfo);
88 : static PGconn *connect_database(const char *conninfo, bool exit_on_error);
89 : static void disconnect_database(PGconn *conn, bool exit_on_error);
90 : static uint64 get_primary_sysid(const char *conninfo);
91 : static uint64 get_standby_sysid(const char *datadir);
92 : static void modify_subscriber_sysid(const struct CreateSubscriberOptions *opt);
93 : static bool server_is_in_recovery(PGconn *conn);
94 : static char *generate_object_name(PGconn *conn);
95 : static void check_publisher(const struct LogicalRepInfo *dbinfo);
96 : static char *setup_publisher(struct LogicalRepInfo *dbinfo);
97 : static void check_subscriber(const struct LogicalRepInfo *dbinfo);
98 : static void setup_subscriber(struct LogicalRepInfo *dbinfo,
99 : const char *consistent_lsn);
100 : static void setup_recovery(const struct LogicalRepInfo *dbinfo, const char *datadir,
101 : const char *lsn);
102 : static void drop_primary_replication_slot(struct LogicalRepInfo *dbinfo,
103 : const char *slotname);
104 : static void drop_failover_replication_slots(struct LogicalRepInfo *dbinfo);
105 : static char *create_logical_replication_slot(PGconn *conn,
106 : struct LogicalRepInfo *dbinfo);
107 : static void drop_replication_slot(PGconn *conn, struct LogicalRepInfo *dbinfo,
108 : const char *slot_name);
109 : static void pg_ctl_status(const char *pg_ctl_cmd, int rc);
110 : static void start_standby_server(const struct CreateSubscriberOptions *opt,
111 : bool restricted_access,
112 : bool restrict_logical_worker);
113 : static void stop_standby_server(const char *datadir);
114 : static void wait_for_end_recovery(const char *conninfo,
115 : const struct CreateSubscriberOptions *opt);
116 : static void create_publication(PGconn *conn, struct LogicalRepInfo *dbinfo);
117 : static void drop_publication(PGconn *conn, const char *pubname,
118 : const char *dbname, bool *made_publication);
119 : static void check_and_drop_publications(PGconn *conn, struct LogicalRepInfo *dbinfo);
120 : static void create_subscription(PGconn *conn, const struct LogicalRepInfo *dbinfo);
121 : static void set_replication_progress(PGconn *conn, const struct LogicalRepInfo *dbinfo,
122 : const char *lsn);
123 : static void enable_subscription(PGconn *conn, const struct LogicalRepInfo *dbinfo);
124 : static void check_and_drop_existing_subscriptions(PGconn *conn,
125 : const struct LogicalRepInfo *dbinfo);
126 : static void drop_existing_subscriptions(PGconn *conn, const char *subname,
127 : const char *dbname);
128 : static void get_publisher_databases(struct CreateSubscriberOptions *opt,
129 : bool dbnamespecified);
130 :
131 : #define USEC_PER_SEC 1000000
132 : #define WAIT_INTERVAL 1 /* 1 second */
133 :
134 : static const char *progname;
135 :
136 : static char *primary_slot_name = NULL;
137 : static bool dry_run = false;
138 :
139 : static bool success = false;
140 :
141 : static struct LogicalRepInfos dbinfos;
142 : static int num_dbs = 0; /* number of specified databases */
143 : static int num_pubs = 0; /* number of specified publications */
144 : static int num_subs = 0; /* number of specified subscriptions */
145 : static int num_replslots = 0; /* number of specified replication slots */
146 :
147 : static pg_prng_state prng_state;
148 :
149 : static char *pg_ctl_path = NULL;
150 : static char *pg_resetwal_path = NULL;
151 :
152 : /* standby / subscriber data directory */
153 : static char *subscriber_dir = NULL;
154 :
155 : static bool recovery_ended = false;
156 : static bool standby_running = false;
157 :
158 : enum WaitPMResult
159 : {
160 : POSTMASTER_READY,
161 : POSTMASTER_STILL_STARTING
162 : };
163 :
164 :
165 : /*
166 : * Cleanup objects that were created by pg_createsubscriber if there is an
167 : * error.
168 : *
169 : * Publications and replication slots are created on primary. Depending on the
170 : * step it failed, it should remove the already created objects if it is
171 : * possible (sometimes it won't work due to a connection issue).
172 : * There is no cleanup on the target server. The steps on the target server are
173 : * executed *after* promotion, hence, at this point, a failure means recreate
174 : * the physical replica and start again.
175 : */
176 : static void
177 20 : cleanup_objects_atexit(void)
178 : {
179 20 : if (success)
180 8 : return;
181 :
182 : /*
183 : * If the server is promoted, there is no way to use the current setup
184 : * again. Warn the user that a new replication setup should be done before
185 : * trying again.
186 : */
187 12 : if (recovery_ended)
188 : {
189 0 : pg_log_warning("failed after the end of recovery");
190 0 : pg_log_warning_hint("The target server cannot be used as a physical replica anymore. "
191 : "You must recreate the physical replica before continuing.");
192 : }
193 :
194 36 : for (int i = 0; i < num_dbs; i++)
195 : {
196 24 : struct LogicalRepInfo *dbinfo = &dbinfos.dbinfo[i];
197 :
198 24 : if (dbinfo->made_publication || dbinfo->made_replslot)
199 : {
200 : PGconn *conn;
201 :
202 0 : conn = connect_database(dbinfo->pubconninfo, false);
203 0 : if (conn != NULL)
204 : {
205 0 : if (dbinfo->made_publication)
206 0 : drop_publication(conn, dbinfo->pubname, dbinfo->dbname,
207 : &dbinfo->made_publication);
208 0 : if (dbinfo->made_replslot)
209 0 : drop_replication_slot(conn, dbinfo, dbinfo->replslotname);
210 0 : disconnect_database(conn, false);
211 : }
212 : else
213 : {
214 : /*
215 : * If a connection could not be established, inform the user
216 : * that some objects were left on primary and should be
217 : * removed before trying again.
218 : */
219 0 : if (dbinfo->made_publication)
220 : {
221 0 : pg_log_warning("publication \"%s\" created in database \"%s\" on primary was left behind",
222 : dbinfo->pubname,
223 : dbinfo->dbname);
224 0 : pg_log_warning_hint("Drop this publication before trying again.");
225 : }
226 0 : if (dbinfo->made_replslot)
227 : {
228 0 : pg_log_warning("replication slot \"%s\" created in database \"%s\" on primary was left behind",
229 : dbinfo->replslotname,
230 : dbinfo->dbname);
231 0 : pg_log_warning_hint("Drop this replication slot soon to avoid retention of WAL files.");
232 : }
233 : }
234 : }
235 : }
236 :
237 12 : if (standby_running)
238 8 : stop_standby_server(subscriber_dir);
239 : }
240 :
241 : static void
242 2 : usage(void)
243 : {
244 2 : printf(_("%s creates a new logical replica from a standby server.\n\n"),
245 : progname);
246 2 : printf(_("Usage:\n"));
247 2 : printf(_(" %s [OPTION]...\n"), progname);
248 2 : printf(_("\nOptions:\n"));
249 2 : printf(_(" -a, --all create subscriptions for all databases except template\n"
250 : " databases and databases that don't allow connections\n"));
251 2 : printf(_(" -d, --database=DBNAME database in which to create a subscription\n"));
252 2 : printf(_(" -D, --pgdata=DATADIR location for the subscriber data directory\n"));
253 2 : printf(_(" -n, --dry-run dry run, just show what would be done\n"));
254 2 : printf(_(" -p, --subscriber-port=PORT subscriber port number (default %s)\n"), DEFAULT_SUB_PORT);
255 2 : printf(_(" -P, --publisher-server=CONNSTR publisher connection string\n"));
256 2 : printf(_(" -s, --socketdir=DIR socket directory to use (default current dir.)\n"));
257 2 : printf(_(" -t, --recovery-timeout=SECS seconds to wait for recovery to end\n"));
258 2 : printf(_(" -T, --enable-two-phase enable two-phase commit for all subscriptions\n"));
259 2 : printf(_(" -U, --subscriber-username=NAME user name for subscriber connection\n"));
260 2 : printf(_(" -v, --verbose output verbose messages\n"));
261 2 : printf(_(" --clean=OBJECTTYPE drop all objects of the specified type from specified\n"
262 : " databases on the subscriber; accepts: \"%s\"\n"), "publications");
263 2 : printf(_(" --config-file=FILENAME use specified main server configuration\n"
264 : " file when running target cluster\n"));
265 2 : printf(_(" --publication=NAME publication name\n"));
266 2 : printf(_(" --replication-slot=NAME replication slot name\n"));
267 2 : printf(_(" --subscription=NAME subscription name\n"));
268 2 : printf(_(" -V, --version output version information, then exit\n"));
269 2 : printf(_(" -?, --help show this help, then exit\n"));
270 2 : printf(_("\nReport bugs to <%s>.\n"), PACKAGE_BUGREPORT);
271 2 : printf(_("%s home page: <%s>\n"), PACKAGE_NAME, PACKAGE_URL);
272 2 : }
273 :
274 : /*
275 : * Subroutine to append "keyword=value" to a connection string,
276 : * with proper quoting of the value. (We assume keywords don't need that.)
277 : */
278 : static void
279 214 : appendConnStrItem(PQExpBuffer buf, const char *keyword, const char *val)
280 : {
281 214 : if (buf->len > 0)
282 158 : appendPQExpBufferChar(buf, ' ');
283 214 : appendPQExpBufferStr(buf, keyword);
284 214 : appendPQExpBufferChar(buf, '=');
285 214 : appendConnStrVal(buf, val);
286 214 : }
287 :
288 : /*
289 : * Validate a connection string. Returns a base connection string that is a
290 : * connection string without a database name.
291 : *
292 : * Since we might process multiple databases, each database name will be
293 : * appended to this base connection string to provide a final connection
294 : * string. If the second argument (dbname) is not null, returns dbname if the
295 : * provided connection string contains it.
296 : *
297 : * It is the caller's responsibility to free the returned connection string and
298 : * dbname.
299 : */
300 : static char *
301 28 : get_base_conninfo(const char *conninfo, char **dbname)
302 : {
303 : PQExpBuffer buf;
304 : PQconninfoOption *conn_opts;
305 : PQconninfoOption *conn_opt;
306 28 : char *errmsg = NULL;
307 : char *ret;
308 :
309 28 : conn_opts = PQconninfoParse(conninfo, &errmsg);
310 28 : if (conn_opts == NULL)
311 : {
312 0 : pg_log_error("could not parse connection string: %s", errmsg);
313 0 : PQfreemem(errmsg);
314 0 : return NULL;
315 : }
316 :
317 28 : buf = createPQExpBuffer();
318 1456 : for (conn_opt = conn_opts; conn_opt->keyword != NULL; conn_opt++)
319 : {
320 1428 : if (conn_opt->val != NULL && conn_opt->val[0] != '\0')
321 : {
322 66 : if (strcmp(conn_opt->keyword, "dbname") == 0)
323 : {
324 18 : if (dbname)
325 18 : *dbname = pg_strdup(conn_opt->val);
326 18 : continue;
327 : }
328 48 : appendConnStrItem(buf, conn_opt->keyword, conn_opt->val);
329 : }
330 : }
331 :
332 28 : ret = pg_strdup(buf->data);
333 :
334 28 : destroyPQExpBuffer(buf);
335 28 : PQconninfoFree(conn_opts);
336 :
337 28 : return ret;
338 : }
339 :
340 : /*
341 : * Build a subscriber connection string. Only a few parameters are supported
342 : * since it starts a server with restricted access.
343 : */
344 : static char *
345 28 : get_sub_conninfo(const struct CreateSubscriberOptions *opt)
346 : {
347 28 : PQExpBuffer buf = createPQExpBuffer();
348 : char *ret;
349 :
350 28 : appendConnStrItem(buf, "port", opt->sub_port);
351 : #if !defined(WIN32)
352 28 : appendConnStrItem(buf, "host", opt->socket_dir);
353 : #endif
354 28 : if (opt->sub_username != NULL)
355 0 : appendConnStrItem(buf, "user", opt->sub_username);
356 28 : appendConnStrItem(buf, "fallback_application_name", progname);
357 :
358 28 : ret = pg_strdup(buf->data);
359 :
360 28 : destroyPQExpBuffer(buf);
361 :
362 28 : return ret;
363 : }
364 :
365 : /*
366 : * Verify if a PostgreSQL binary (progname) is available in the same directory as
367 : * pg_createsubscriber and it has the same version. It returns the absolute
368 : * path of the progname.
369 : */
370 : static char *
371 40 : get_exec_path(const char *argv0, const char *progname)
372 : {
373 : char *versionstr;
374 : char *exec_path;
375 : int ret;
376 :
377 40 : versionstr = psprintf("%s (PostgreSQL) %s\n", progname, PG_VERSION);
378 40 : exec_path = pg_malloc(MAXPGPATH);
379 40 : ret = find_other_exec(argv0, progname, versionstr, exec_path);
380 :
381 40 : if (ret < 0)
382 : {
383 : char full_path[MAXPGPATH];
384 :
385 0 : if (find_my_exec(argv0, full_path) < 0)
386 0 : strlcpy(full_path, progname, sizeof(full_path));
387 :
388 0 : if (ret == -1)
389 0 : pg_fatal("program \"%s\" is needed by %s but was not found in the same directory as \"%s\"",
390 : progname, "pg_createsubscriber", full_path);
391 : else
392 0 : pg_fatal("program \"%s\" was found by \"%s\" but was not the same version as %s",
393 : progname, full_path, "pg_createsubscriber");
394 : }
395 :
396 40 : pg_log_debug("%s path is: %s", progname, exec_path);
397 :
398 40 : return exec_path;
399 : }
400 :
401 : /*
402 : * Is it a cluster directory? These are preliminary checks. It is far from
403 : * making an accurate check. If it is not a clone from the publisher, it will
404 : * eventually fail in a future step.
405 : */
406 : static void
407 20 : check_data_directory(const char *datadir)
408 : {
409 : struct stat statbuf;
410 : char versionfile[MAXPGPATH];
411 :
412 20 : pg_log_info("checking if directory \"%s\" is a cluster data directory",
413 : datadir);
414 :
415 20 : if (stat(datadir, &statbuf) != 0)
416 : {
417 0 : if (errno == ENOENT)
418 0 : pg_fatal("data directory \"%s\" does not exist", datadir);
419 : else
420 0 : pg_fatal("could not access directory \"%s\": %m", datadir);
421 : }
422 :
423 20 : snprintf(versionfile, MAXPGPATH, "%s/PG_VERSION", datadir);
424 20 : if (stat(versionfile, &statbuf) != 0 && errno == ENOENT)
425 : {
426 0 : pg_fatal("directory \"%s\" is not a database cluster directory",
427 : datadir);
428 : }
429 20 : }
430 :
431 : /*
432 : * Append database name into a base connection string.
433 : *
434 : * dbname is the only parameter that changes so it is not included in the base
435 : * connection string. This function concatenates dbname to build a "real"
436 : * connection string.
437 : */
438 : static char *
439 82 : concat_conninfo_dbname(const char *conninfo, const char *dbname)
440 : {
441 82 : PQExpBuffer buf = createPQExpBuffer();
442 : char *ret;
443 :
444 : Assert(conninfo != NULL);
445 :
446 82 : appendPQExpBufferStr(buf, conninfo);
447 82 : appendConnStrItem(buf, "dbname", dbname);
448 :
449 82 : ret = pg_strdup(buf->data);
450 82 : destroyPQExpBuffer(buf);
451 :
452 82 : return ret;
453 : }
454 :
455 : /*
456 : * Store publication and subscription information.
457 : *
458 : * If publication, replication slot and subscription names were specified,
459 : * store it here. Otherwise, a generated name will be assigned to the object in
460 : * setup_publisher().
461 : */
462 : static struct LogicalRepInfo *
463 20 : store_pub_sub_info(const struct CreateSubscriberOptions *opt,
464 : const char *pub_base_conninfo,
465 : const char *sub_base_conninfo)
466 : {
467 : struct LogicalRepInfo *dbinfo;
468 20 : SimpleStringListCell *pubcell = NULL;
469 20 : SimpleStringListCell *subcell = NULL;
470 20 : SimpleStringListCell *replslotcell = NULL;
471 20 : int i = 0;
472 :
473 20 : dbinfo = pg_malloc_array(struct LogicalRepInfo, num_dbs);
474 :
475 20 : if (num_pubs > 0)
476 4 : pubcell = opt->pub_names.head;
477 20 : if (num_subs > 0)
478 2 : subcell = opt->sub_names.head;
479 20 : if (num_replslots > 0)
480 4 : replslotcell = opt->replslot_names.head;
481 :
482 60 : for (SimpleStringListCell *cell = opt->database_names.head; cell; cell = cell->next)
483 : {
484 : char *conninfo;
485 :
486 : /* Fill publisher attributes */
487 40 : conninfo = concat_conninfo_dbname(pub_base_conninfo, cell->val);
488 40 : dbinfo[i].pubconninfo = conninfo;
489 40 : dbinfo[i].dbname = cell->val;
490 40 : if (num_pubs > 0)
491 8 : dbinfo[i].pubname = pubcell->val;
492 : else
493 32 : dbinfo[i].pubname = NULL;
494 40 : if (num_replslots > 0)
495 6 : dbinfo[i].replslotname = replslotcell->val;
496 : else
497 34 : dbinfo[i].replslotname = NULL;
498 40 : dbinfo[i].made_replslot = false;
499 40 : dbinfo[i].made_publication = false;
500 : /* Fill subscriber attributes */
501 40 : conninfo = concat_conninfo_dbname(sub_base_conninfo, cell->val);
502 40 : dbinfo[i].subconninfo = conninfo;
503 40 : if (num_subs > 0)
504 4 : dbinfo[i].subname = subcell->val;
505 : else
506 36 : dbinfo[i].subname = NULL;
507 : /* Other fields will be filled later */
508 :
509 40 : pg_log_debug("publisher(%d): publication: %s ; replication slot: %s ; connection string: %s", i,
510 : dbinfo[i].pubname ? dbinfo[i].pubname : "(auto)",
511 : dbinfo[i].replslotname ? dbinfo[i].replslotname : "(auto)",
512 : dbinfo[i].pubconninfo);
513 40 : pg_log_debug("subscriber(%d): subscription: %s ; connection string: %s, two_phase: %s", i,
514 : dbinfo[i].subname ? dbinfo[i].subname : "(auto)",
515 : dbinfo[i].subconninfo,
516 : dbinfos.two_phase ? "true" : "false");
517 :
518 40 : if (num_pubs > 0)
519 8 : pubcell = pubcell->next;
520 40 : if (num_subs > 0)
521 4 : subcell = subcell->next;
522 40 : if (num_replslots > 0)
523 6 : replslotcell = replslotcell->next;
524 :
525 40 : i++;
526 : }
527 :
528 20 : return dbinfo;
529 : }
530 :
531 : /*
532 : * Open a new connection. If exit_on_error is true, it has an undesired
533 : * condition and it should exit immediately.
534 : */
535 : static PGconn *
536 114 : connect_database(const char *conninfo, bool exit_on_error)
537 : {
538 : PGconn *conn;
539 : PGresult *res;
540 :
541 114 : conn = PQconnectdb(conninfo);
542 114 : if (PQstatus(conn) != CONNECTION_OK)
543 : {
544 0 : pg_log_error("connection to database failed: %s",
545 : PQerrorMessage(conn));
546 0 : PQfinish(conn);
547 :
548 0 : if (exit_on_error)
549 0 : exit(1);
550 0 : return NULL;
551 : }
552 :
553 : /* Secure search_path */
554 114 : res = PQexec(conn, ALWAYS_SECURE_SEARCH_PATH_SQL);
555 114 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
556 : {
557 0 : pg_log_error("could not clear \"search_path\": %s",
558 : PQresultErrorMessage(res));
559 0 : PQclear(res);
560 0 : PQfinish(conn);
561 :
562 0 : if (exit_on_error)
563 0 : exit(1);
564 0 : return NULL;
565 : }
566 114 : PQclear(res);
567 :
568 114 : return conn;
569 : }
570 :
571 : /*
572 : * Close the connection. If exit_on_error is true, it has an undesired
573 : * condition and it should exit immediately.
574 : */
575 : static void
576 114 : disconnect_database(PGconn *conn, bool exit_on_error)
577 : {
578 : Assert(conn != NULL);
579 :
580 114 : PQfinish(conn);
581 :
582 114 : if (exit_on_error)
583 4 : exit(1);
584 110 : }
585 :
586 : /*
587 : * Obtain the system identifier using the provided connection. It will be used
588 : * to compare if a data directory is a clone of another one.
589 : */
590 : static uint64
591 20 : get_primary_sysid(const char *conninfo)
592 : {
593 : PGconn *conn;
594 : PGresult *res;
595 : uint64 sysid;
596 :
597 20 : pg_log_info("getting system identifier from publisher");
598 :
599 20 : conn = connect_database(conninfo, true);
600 :
601 20 : res = PQexec(conn, "SELECT system_identifier FROM pg_catalog.pg_control_system()");
602 20 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
603 : {
604 0 : pg_log_error("could not get system identifier: %s",
605 : PQresultErrorMessage(res));
606 0 : disconnect_database(conn, true);
607 : }
608 20 : if (PQntuples(res) != 1)
609 : {
610 0 : pg_log_error("could not get system identifier: got %d rows, expected %d row",
611 : PQntuples(res), 1);
612 0 : disconnect_database(conn, true);
613 : }
614 :
615 20 : sysid = strtou64(PQgetvalue(res, 0, 0), NULL, 10);
616 :
617 20 : pg_log_info("system identifier is %" PRIu64 " on publisher", sysid);
618 :
619 20 : PQclear(res);
620 20 : disconnect_database(conn, false);
621 :
622 20 : return sysid;
623 : }
624 :
625 : /*
626 : * Obtain the system identifier from control file. It will be used to compare
627 : * if a data directory is a clone of another one. This routine is used locally
628 : * and avoids a connection.
629 : */
630 : static uint64
631 20 : get_standby_sysid(const char *datadir)
632 : {
633 : ControlFileData *cf;
634 : bool crc_ok;
635 : uint64 sysid;
636 :
637 20 : pg_log_info("getting system identifier from subscriber");
638 :
639 20 : cf = get_controlfile(datadir, &crc_ok);
640 20 : if (!crc_ok)
641 0 : pg_fatal("control file appears to be corrupt");
642 :
643 20 : sysid = cf->system_identifier;
644 :
645 20 : pg_log_info("system identifier is %" PRIu64 " on subscriber", sysid);
646 :
647 20 : pg_free(cf);
648 :
649 20 : return sysid;
650 : }
651 :
652 : /*
653 : * Modify the system identifier. Since a standby server preserves the system
654 : * identifier, it makes sense to change it to avoid situations in which WAL
655 : * files from one of the systems might be used in the other one.
656 : */
657 : static void
658 8 : modify_subscriber_sysid(const struct CreateSubscriberOptions *opt)
659 : {
660 : ControlFileData *cf;
661 : bool crc_ok;
662 : struct timeval tv;
663 :
664 : char *cmd_str;
665 :
666 8 : pg_log_info("modifying system identifier of subscriber");
667 :
668 8 : cf = get_controlfile(subscriber_dir, &crc_ok);
669 8 : if (!crc_ok)
670 0 : pg_fatal("control file appears to be corrupt");
671 :
672 : /*
673 : * Select a new system identifier.
674 : *
675 : * XXX this code was extracted from BootStrapXLOG().
676 : */
677 8 : gettimeofday(&tv, NULL);
678 8 : cf->system_identifier = ((uint64) tv.tv_sec) << 32;
679 8 : cf->system_identifier |= ((uint64) tv.tv_usec) << 12;
680 8 : cf->system_identifier |= getpid() & 0xFFF;
681 :
682 8 : if (!dry_run)
683 2 : update_controlfile(subscriber_dir, cf, true);
684 :
685 8 : pg_log_info("system identifier is %" PRIu64 " on subscriber",
686 : cf->system_identifier);
687 :
688 8 : pg_log_info("running pg_resetwal on the subscriber");
689 :
690 8 : cmd_str = psprintf("\"%s\" -D \"%s\" > \"%s\"", pg_resetwal_path,
691 : subscriber_dir, DEVNULL);
692 :
693 8 : pg_log_debug("pg_resetwal command is: %s", cmd_str);
694 :
695 8 : if (!dry_run)
696 : {
697 2 : int rc = system(cmd_str);
698 :
699 2 : if (rc == 0)
700 2 : pg_log_info("subscriber successfully changed the system identifier");
701 : else
702 0 : pg_fatal("could not change system identifier of subscriber: %s", wait_result_to_str(rc));
703 : }
704 :
705 8 : pg_free(cf);
706 8 : }
707 :
708 : /*
709 : * Generate an object name using a prefix, database oid and a random integer.
710 : * It is used in case the user does not specify an object name (publication,
711 : * subscription, replication slot).
712 : */
713 : static char *
714 16 : generate_object_name(PGconn *conn)
715 : {
716 : PGresult *res;
717 : Oid oid;
718 : uint32 rand;
719 : char *objname;
720 :
721 16 : res = PQexec(conn,
722 : "SELECT oid FROM pg_catalog.pg_database "
723 : "WHERE datname = pg_catalog.current_database()");
724 16 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
725 : {
726 0 : pg_log_error("could not obtain database OID: %s",
727 : PQresultErrorMessage(res));
728 0 : disconnect_database(conn, true);
729 : }
730 :
731 16 : if (PQntuples(res) != 1)
732 : {
733 0 : pg_log_error("could not obtain database OID: got %d rows, expected %d row",
734 : PQntuples(res), 1);
735 0 : disconnect_database(conn, true);
736 : }
737 :
738 : /* Database OID */
739 16 : oid = strtoul(PQgetvalue(res, 0, 0), NULL, 10);
740 :
741 16 : PQclear(res);
742 :
743 : /* Random unsigned integer */
744 16 : rand = pg_prng_uint32(&prng_state);
745 :
746 : /*
747 : * Build the object name. The name must not exceed NAMEDATALEN - 1. This
748 : * current schema uses a maximum of 40 characters (20 + 10 + 1 + 8 +
749 : * '\0').
750 : */
751 16 : objname = psprintf("pg_createsubscriber_%u_%x", oid, rand);
752 :
753 16 : return objname;
754 : }
755 :
756 : /*
757 : * Create the publications and replication slots in preparation for logical
758 : * replication. Returns the LSN from latest replication slot. It will be the
759 : * replication start point that is used to adjust the subscriptions (see
760 : * set_replication_progress).
761 : */
762 : static char *
763 8 : setup_publisher(struct LogicalRepInfo *dbinfo)
764 : {
765 8 : char *lsn = NULL;
766 :
767 8 : pg_prng_seed(&prng_state, (uint64) (getpid() ^ time(NULL)));
768 :
769 24 : for (int i = 0; i < num_dbs; i++)
770 : {
771 : PGconn *conn;
772 16 : char *genname = NULL;
773 :
774 16 : conn = connect_database(dbinfo[i].pubconninfo, true);
775 :
776 : /*
777 : * If an object name was not specified as command-line options, assign
778 : * a generated object name. The replication slot has a different rule.
779 : * The subscription name is assigned to the replication slot name if
780 : * no replication slot is specified. It follows the same rule as
781 : * CREATE SUBSCRIPTION.
782 : */
783 16 : if (num_pubs == 0 || num_subs == 0 || num_replslots == 0)
784 16 : genname = generate_object_name(conn);
785 16 : if (num_pubs == 0)
786 8 : dbinfo[i].pubname = pg_strdup(genname);
787 16 : if (num_subs == 0)
788 12 : dbinfo[i].subname = pg_strdup(genname);
789 16 : if (num_replslots == 0)
790 10 : dbinfo[i].replslotname = pg_strdup(dbinfo[i].subname);
791 :
792 : /*
793 : * Create publication on publisher. This step should be executed
794 : * *before* promoting the subscriber to avoid any transactions between
795 : * consistent LSN and the new publication rows (such transactions
796 : * wouldn't see the new publication rows resulting in an error).
797 : */
798 16 : create_publication(conn, &dbinfo[i]);
799 :
800 : /* Create replication slot on publisher */
801 16 : if (lsn)
802 2 : pg_free(lsn);
803 16 : lsn = create_logical_replication_slot(conn, &dbinfo[i]);
804 16 : if (lsn != NULL || dry_run)
805 16 : pg_log_info("create replication slot \"%s\" on publisher",
806 : dbinfo[i].replslotname);
807 : else
808 0 : exit(1);
809 :
810 : /*
811 : * Since we are using the LSN returned by the last replication slot as
812 : * recovery_target_lsn, this LSN is ahead of the current WAL position
813 : * and the recovery waits until the publisher writes a WAL record to
814 : * reach the target and ends the recovery. On idle systems, this wait
815 : * time is unpredictable and could lead to failure in promoting the
816 : * subscriber. To avoid that, insert a harmless WAL record.
817 : */
818 16 : if (i == num_dbs - 1 && !dry_run)
819 : {
820 : PGresult *res;
821 :
822 2 : res = PQexec(conn, "SELECT pg_log_standby_snapshot()");
823 2 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
824 : {
825 0 : pg_log_error("could not write an additional WAL record: %s",
826 : PQresultErrorMessage(res));
827 0 : disconnect_database(conn, true);
828 : }
829 2 : PQclear(res);
830 : }
831 :
832 16 : disconnect_database(conn, false);
833 : }
834 :
835 8 : return lsn;
836 : }
837 :
838 : /*
839 : * Is recovery still in progress?
840 : */
841 : static bool
842 36 : server_is_in_recovery(PGconn *conn)
843 : {
844 : PGresult *res;
845 : int ret;
846 :
847 36 : res = PQexec(conn, "SELECT pg_catalog.pg_is_in_recovery()");
848 :
849 36 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
850 : {
851 0 : pg_log_error("could not obtain recovery progress: %s",
852 : PQresultErrorMessage(res));
853 0 : disconnect_database(conn, true);
854 : }
855 :
856 :
857 36 : ret = strcmp("t", PQgetvalue(res, 0, 0));
858 :
859 36 : PQclear(res);
860 :
861 36 : return ret == 0;
862 : }
863 :
864 : /*
865 : * Is the primary server ready for logical replication?
866 : *
867 : * XXX Does it not allow a synchronous replica?
868 : */
869 : static void
870 12 : check_publisher(const struct LogicalRepInfo *dbinfo)
871 : {
872 : PGconn *conn;
873 : PGresult *res;
874 12 : bool failed = false;
875 :
876 : char *wal_level;
877 : int max_repslots;
878 : int cur_repslots;
879 : int max_walsenders;
880 : int cur_walsenders;
881 : int max_prepared_transactions;
882 : char *max_slot_wal_keep_size;
883 :
884 12 : pg_log_info("checking settings on publisher");
885 :
886 12 : conn = connect_database(dbinfo[0].pubconninfo, true);
887 :
888 : /*
889 : * If the primary server is in recovery (i.e. cascading replication),
890 : * objects (publication) cannot be created because it is read only.
891 : */
892 12 : if (server_is_in_recovery(conn))
893 : {
894 2 : pg_log_error("primary server cannot be in recovery");
895 2 : disconnect_database(conn, true);
896 : }
897 :
898 : /*------------------------------------------------------------------------
899 : * Logical replication requires a few parameters to be set on publisher.
900 : * Since these parameters are not a requirement for physical replication,
901 : * we should check it to make sure it won't fail.
902 : *
903 : * - wal_level = logical
904 : * - max_replication_slots >= current + number of dbs to be converted
905 : * - max_wal_senders >= current + number of dbs to be converted
906 : * - max_slot_wal_keep_size = -1 (to prevent deletion of required WAL files)
907 : * -----------------------------------------------------------------------
908 : */
909 10 : res = PQexec(conn,
910 : "SELECT pg_catalog.current_setting('wal_level'),"
911 : " pg_catalog.current_setting('max_replication_slots'),"
912 : " (SELECT count(*) FROM pg_catalog.pg_replication_slots),"
913 : " pg_catalog.current_setting('max_wal_senders'),"
914 : " (SELECT count(*) FROM pg_catalog.pg_stat_activity WHERE backend_type = 'walsender'),"
915 : " pg_catalog.current_setting('max_prepared_transactions'),"
916 : " pg_catalog.current_setting('max_slot_wal_keep_size')");
917 :
918 10 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
919 : {
920 0 : pg_log_error("could not obtain publisher settings: %s",
921 : PQresultErrorMessage(res));
922 0 : disconnect_database(conn, true);
923 : }
924 :
925 10 : wal_level = pg_strdup(PQgetvalue(res, 0, 0));
926 10 : max_repslots = atoi(PQgetvalue(res, 0, 1));
927 10 : cur_repslots = atoi(PQgetvalue(res, 0, 2));
928 10 : max_walsenders = atoi(PQgetvalue(res, 0, 3));
929 10 : cur_walsenders = atoi(PQgetvalue(res, 0, 4));
930 10 : max_prepared_transactions = atoi(PQgetvalue(res, 0, 5));
931 10 : max_slot_wal_keep_size = pg_strdup(PQgetvalue(res, 0, 6));
932 :
933 10 : PQclear(res);
934 :
935 10 : pg_log_debug("publisher: wal_level: %s", wal_level);
936 10 : pg_log_debug("publisher: max_replication_slots: %d", max_repslots);
937 10 : pg_log_debug("publisher: current replication slots: %d", cur_repslots);
938 10 : pg_log_debug("publisher: max_wal_senders: %d", max_walsenders);
939 10 : pg_log_debug("publisher: current wal senders: %d", cur_walsenders);
940 10 : pg_log_debug("publisher: max_prepared_transactions: %d",
941 : max_prepared_transactions);
942 10 : pg_log_debug("publisher: max_slot_wal_keep_size: %s",
943 : max_slot_wal_keep_size);
944 :
945 10 : disconnect_database(conn, false);
946 :
947 10 : if (strcmp(wal_level, "logical") != 0)
948 : {
949 2 : pg_log_error("publisher requires \"wal_level\" >= \"logical\"");
950 2 : failed = true;
951 : }
952 :
953 10 : if (max_repslots - cur_repslots < num_dbs)
954 : {
955 2 : pg_log_error("publisher requires %d replication slots, but only %d remain",
956 : num_dbs, max_repslots - cur_repslots);
957 2 : pg_log_error_hint("Increase the configuration parameter \"%s\" to at least %d.",
958 : "max_replication_slots", cur_repslots + num_dbs);
959 2 : failed = true;
960 : }
961 :
962 10 : if (max_walsenders - cur_walsenders < num_dbs)
963 : {
964 2 : pg_log_error("publisher requires %d WAL sender processes, but only %d remain",
965 : num_dbs, max_walsenders - cur_walsenders);
966 2 : pg_log_error_hint("Increase the configuration parameter \"%s\" to at least %d.",
967 : "max_wal_senders", cur_walsenders + num_dbs);
968 2 : failed = true;
969 : }
970 :
971 10 : if (max_prepared_transactions != 0 && !dbinfos.two_phase)
972 : {
973 0 : pg_log_warning("two_phase option will not be enabled for replication slots");
974 0 : pg_log_warning_detail("Subscriptions will be created with the two_phase option disabled. "
975 : "Prepared transactions will be replicated at COMMIT PREPARED.");
976 0 : pg_log_warning_hint("You can use the command-line option --enable-two-phase to enable two_phase.");
977 : }
978 :
979 : /*
980 : * Validate 'max_slot_wal_keep_size'. If this parameter is set to a
981 : * non-default value, it may cause replication failures due to required
982 : * WAL files being prematurely removed.
983 : */
984 10 : if (dry_run && (strcmp(max_slot_wal_keep_size, "-1") != 0))
985 : {
986 0 : pg_log_warning("required WAL could be removed from the publisher");
987 0 : pg_log_warning_hint("Set the configuration parameter \"%s\" to -1 to ensure that required WAL files are not prematurely removed.",
988 : "max_slot_wal_keep_size");
989 : }
990 :
991 10 : pg_free(wal_level);
992 :
993 10 : if (failed)
994 2 : exit(1);
995 8 : }
996 :
997 : /*
998 : * Is the standby server ready for logical replication?
999 : *
1000 : * XXX Does it not allow a time-delayed replica?
1001 : *
1002 : * XXX In a cascaded replication scenario (P -> S -> C), if the target server
1003 : * is S, it cannot detect there is a replica (server C) because server S starts
1004 : * accepting only local connections and server C cannot connect to it. Hence,
1005 : * there is not a reliable way to provide a suitable error saying the server C
1006 : * will be broken at the end of this process (due to pg_resetwal).
1007 : */
1008 : static void
1009 16 : check_subscriber(const struct LogicalRepInfo *dbinfo)
1010 : {
1011 : PGconn *conn;
1012 : PGresult *res;
1013 16 : bool failed = false;
1014 :
1015 : int max_lrworkers;
1016 : int max_reporigins;
1017 : int max_wprocs;
1018 :
1019 16 : pg_log_info("checking settings on subscriber");
1020 :
1021 16 : conn = connect_database(dbinfo[0].subconninfo, true);
1022 :
1023 : /* The target server must be a standby */
1024 16 : if (!server_is_in_recovery(conn))
1025 : {
1026 2 : pg_log_error("target server must be a standby");
1027 2 : disconnect_database(conn, true);
1028 : }
1029 :
1030 : /*------------------------------------------------------------------------
1031 : * Logical replication requires a few parameters to be set on subscriber.
1032 : * Since these parameters are not a requirement for physical replication,
1033 : * we should check it to make sure it won't fail.
1034 : *
1035 : * - max_active_replication_origins >= number of dbs to be converted
1036 : * - max_logical_replication_workers >= number of dbs to be converted
1037 : * - max_worker_processes >= 1 + number of dbs to be converted
1038 : *------------------------------------------------------------------------
1039 : */
1040 14 : res = PQexec(conn,
1041 : "SELECT setting FROM pg_catalog.pg_settings WHERE name IN ("
1042 : "'max_logical_replication_workers', "
1043 : "'max_active_replication_origins', "
1044 : "'max_worker_processes', "
1045 : "'primary_slot_name') "
1046 : "ORDER BY name");
1047 :
1048 14 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
1049 : {
1050 0 : pg_log_error("could not obtain subscriber settings: %s",
1051 : PQresultErrorMessage(res));
1052 0 : disconnect_database(conn, true);
1053 : }
1054 :
1055 14 : max_reporigins = atoi(PQgetvalue(res, 0, 0));
1056 14 : max_lrworkers = atoi(PQgetvalue(res, 1, 0));
1057 14 : max_wprocs = atoi(PQgetvalue(res, 2, 0));
1058 14 : if (strcmp(PQgetvalue(res, 3, 0), "") != 0)
1059 12 : primary_slot_name = pg_strdup(PQgetvalue(res, 3, 0));
1060 :
1061 14 : pg_log_debug("subscriber: max_logical_replication_workers: %d",
1062 : max_lrworkers);
1063 14 : pg_log_debug("subscriber: max_active_replication_origins: %d", max_reporigins);
1064 14 : pg_log_debug("subscriber: max_worker_processes: %d", max_wprocs);
1065 14 : if (primary_slot_name)
1066 12 : pg_log_debug("subscriber: primary_slot_name: %s", primary_slot_name);
1067 :
1068 14 : PQclear(res);
1069 :
1070 14 : disconnect_database(conn, false);
1071 :
1072 14 : if (max_reporigins < num_dbs)
1073 : {
1074 2 : pg_log_error("subscriber requires %d active replication origins, but only %d remain",
1075 : num_dbs, max_reporigins);
1076 2 : pg_log_error_hint("Increase the configuration parameter \"%s\" to at least %d.",
1077 : "max_active_replication_origins", num_dbs);
1078 2 : failed = true;
1079 : }
1080 :
1081 14 : if (max_lrworkers < num_dbs)
1082 : {
1083 2 : pg_log_error("subscriber requires %d logical replication workers, but only %d remain",
1084 : num_dbs, max_lrworkers);
1085 2 : pg_log_error_hint("Increase the configuration parameter \"%s\" to at least %d.",
1086 : "max_logical_replication_workers", num_dbs);
1087 2 : failed = true;
1088 : }
1089 :
1090 14 : if (max_wprocs < num_dbs + 1)
1091 : {
1092 2 : pg_log_error("subscriber requires %d worker processes, but only %d remain",
1093 : num_dbs + 1, max_wprocs);
1094 2 : pg_log_error_hint("Increase the configuration parameter \"%s\" to at least %d.",
1095 : "max_worker_processes", num_dbs + 1);
1096 2 : failed = true;
1097 : }
1098 :
1099 14 : if (failed)
1100 2 : exit(1);
1101 12 : }
1102 :
1103 : /*
1104 : * Drop a specified subscription. This is to avoid duplicate subscriptions on
1105 : * the primary (publisher node) and the newly created subscriber. We
1106 : * shouldn't drop the associated slot as that would be used by the publisher
1107 : * node.
1108 : */
1109 : static void
1110 8 : drop_existing_subscriptions(PGconn *conn, const char *subname, const char *dbname)
1111 : {
1112 8 : PQExpBuffer query = createPQExpBuffer();
1113 : PGresult *res;
1114 :
1115 : Assert(conn != NULL);
1116 :
1117 : /*
1118 : * Construct a query string. These commands are allowed to be executed
1119 : * within a transaction.
1120 : */
1121 8 : appendPQExpBuffer(query, "ALTER SUBSCRIPTION %s DISABLE;",
1122 : subname);
1123 8 : appendPQExpBuffer(query, " ALTER SUBSCRIPTION %s SET (slot_name = NONE);",
1124 : subname);
1125 8 : appendPQExpBuffer(query, " DROP SUBSCRIPTION %s;", subname);
1126 :
1127 8 : pg_log_info("dropping subscription \"%s\" in database \"%s\"",
1128 : subname, dbname);
1129 :
1130 8 : if (!dry_run)
1131 : {
1132 2 : res = PQexec(conn, query->data);
1133 :
1134 2 : if (PQresultStatus(res) != PGRES_COMMAND_OK)
1135 : {
1136 0 : pg_log_error("could not drop subscription \"%s\": %s",
1137 : subname, PQresultErrorMessage(res));
1138 0 : disconnect_database(conn, true);
1139 : }
1140 :
1141 2 : PQclear(res);
1142 : }
1143 :
1144 8 : destroyPQExpBuffer(query);
1145 8 : }
1146 :
1147 : /*
1148 : * Retrieve and drop the pre-existing subscriptions.
1149 : */
1150 : static void
1151 16 : check_and_drop_existing_subscriptions(PGconn *conn,
1152 : const struct LogicalRepInfo *dbinfo)
1153 : {
1154 16 : PQExpBuffer query = createPQExpBuffer();
1155 : char *dbname;
1156 : PGresult *res;
1157 :
1158 : Assert(conn != NULL);
1159 :
1160 16 : dbname = PQescapeLiteral(conn, dbinfo->dbname, strlen(dbinfo->dbname));
1161 :
1162 16 : appendPQExpBuffer(query,
1163 : "SELECT s.subname FROM pg_catalog.pg_subscription s "
1164 : "INNER JOIN pg_catalog.pg_database d ON (s.subdbid = d.oid) "
1165 : "WHERE d.datname = %s",
1166 : dbname);
1167 16 : res = PQexec(conn, query->data);
1168 :
1169 16 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
1170 : {
1171 0 : pg_log_error("could not obtain pre-existing subscriptions: %s",
1172 : PQresultErrorMessage(res));
1173 0 : disconnect_database(conn, true);
1174 : }
1175 :
1176 24 : for (int i = 0; i < PQntuples(res); i++)
1177 8 : drop_existing_subscriptions(conn, PQgetvalue(res, i, 0),
1178 8 : dbinfo->dbname);
1179 :
1180 16 : PQclear(res);
1181 16 : destroyPQExpBuffer(query);
1182 16 : PQfreemem(dbname);
1183 16 : }
1184 :
1185 : /*
1186 : * Create the subscriptions, adjust the initial location for logical
1187 : * replication and enable the subscriptions. That's the last step for logical
1188 : * replication setup.
1189 : */
1190 : static void
1191 8 : setup_subscriber(struct LogicalRepInfo *dbinfo, const char *consistent_lsn)
1192 : {
1193 24 : for (int i = 0; i < num_dbs; i++)
1194 : {
1195 : PGconn *conn;
1196 :
1197 : /* Connect to subscriber. */
1198 16 : conn = connect_database(dbinfo[i].subconninfo, true);
1199 :
1200 : /*
1201 : * We don't need the pre-existing subscriptions on the newly formed
1202 : * subscriber. They can connect to other publisher nodes and either
1203 : * get some unwarranted data or can lead to ERRORs in connecting to
1204 : * such nodes.
1205 : */
1206 16 : check_and_drop_existing_subscriptions(conn, &dbinfo[i]);
1207 :
1208 : /* Check and drop the required publications in the given database. */
1209 16 : check_and_drop_publications(conn, &dbinfo[i]);
1210 :
1211 16 : create_subscription(conn, &dbinfo[i]);
1212 :
1213 : /* Set the replication progress to the correct LSN */
1214 16 : set_replication_progress(conn, &dbinfo[i], consistent_lsn);
1215 :
1216 : /* Enable subscription */
1217 16 : enable_subscription(conn, &dbinfo[i]);
1218 :
1219 16 : disconnect_database(conn, false);
1220 : }
1221 8 : }
1222 :
1223 : /*
1224 : * Write the required recovery parameters.
1225 : */
1226 : static void
1227 8 : setup_recovery(const struct LogicalRepInfo *dbinfo, const char *datadir, const char *lsn)
1228 : {
1229 : PGconn *conn;
1230 : PQExpBuffer recoveryconfcontents;
1231 :
1232 : /*
1233 : * Despite of the recovery parameters will be written to the subscriber,
1234 : * use a publisher connection. The primary_conninfo is generated using the
1235 : * connection settings.
1236 : */
1237 8 : conn = connect_database(dbinfo[0].pubconninfo, true);
1238 :
1239 : /*
1240 : * Write recovery parameters.
1241 : *
1242 : * The subscriber is not running yet. In dry run mode, the recovery
1243 : * parameters *won't* be written. An invalid LSN is used for printing
1244 : * purposes. Additional recovery parameters are added here. It avoids
1245 : * unexpected behavior such as end of recovery as soon as a consistent
1246 : * state is reached (recovery_target) and failure due to multiple recovery
1247 : * targets (name, time, xid, LSN).
1248 : */
1249 8 : recoveryconfcontents = GenerateRecoveryConfig(conn, NULL, NULL);
1250 8 : appendPQExpBufferStr(recoveryconfcontents, "recovery_target = ''\n");
1251 8 : appendPQExpBufferStr(recoveryconfcontents,
1252 : "recovery_target_timeline = 'latest'\n");
1253 :
1254 : /*
1255 : * Set recovery_target_inclusive = false to avoid reapplying the
1256 : * transaction committed at 'lsn' after subscription is enabled. This is
1257 : * because the provided 'lsn' is also used as the replication start point
1258 : * for the subscription. So, the server can send the transaction committed
1259 : * at that 'lsn' after replication is started which can lead to applying
1260 : * the same transaction twice if we keep recovery_target_inclusive = true.
1261 : */
1262 8 : appendPQExpBufferStr(recoveryconfcontents,
1263 : "recovery_target_inclusive = false\n");
1264 8 : appendPQExpBufferStr(recoveryconfcontents,
1265 : "recovery_target_action = promote\n");
1266 8 : appendPQExpBufferStr(recoveryconfcontents, "recovery_target_name = ''\n");
1267 8 : appendPQExpBufferStr(recoveryconfcontents, "recovery_target_time = ''\n");
1268 8 : appendPQExpBufferStr(recoveryconfcontents, "recovery_target_xid = ''\n");
1269 :
1270 8 : if (dry_run)
1271 : {
1272 6 : appendPQExpBufferStr(recoveryconfcontents, "# dry run mode");
1273 6 : appendPQExpBuffer(recoveryconfcontents,
1274 : "recovery_target_lsn = '%X/%08X'\n",
1275 6 : LSN_FORMAT_ARGS((XLogRecPtr) InvalidXLogRecPtr));
1276 : }
1277 : else
1278 : {
1279 2 : appendPQExpBuffer(recoveryconfcontents, "recovery_target_lsn = '%s'\n",
1280 : lsn);
1281 2 : WriteRecoveryConfig(conn, datadir, recoveryconfcontents);
1282 : }
1283 8 : disconnect_database(conn, false);
1284 :
1285 8 : pg_log_debug("recovery parameters:\n%s", recoveryconfcontents->data);
1286 8 : }
1287 :
1288 : /*
1289 : * Drop physical replication slot on primary if the standby was using it. After
1290 : * the transformation, it has no use.
1291 : *
1292 : * XXX we might not fail here. Instead, we provide a warning so the user
1293 : * eventually drops this replication slot later.
1294 : */
1295 : static void
1296 8 : drop_primary_replication_slot(struct LogicalRepInfo *dbinfo, const char *slotname)
1297 : {
1298 : PGconn *conn;
1299 :
1300 : /* Replication slot does not exist, do nothing */
1301 8 : if (!primary_slot_name)
1302 0 : return;
1303 :
1304 8 : conn = connect_database(dbinfo[0].pubconninfo, false);
1305 8 : if (conn != NULL)
1306 : {
1307 8 : drop_replication_slot(conn, &dbinfo[0], slotname);
1308 8 : disconnect_database(conn, false);
1309 : }
1310 : else
1311 : {
1312 0 : pg_log_warning("could not drop replication slot \"%s\" on primary",
1313 : slotname);
1314 0 : pg_log_warning_hint("Drop this replication slot soon to avoid retention of WAL files.");
1315 : }
1316 : }
1317 :
1318 : /*
1319 : * Drop failover replication slots on subscriber. After the transformation,
1320 : * they have no use.
1321 : *
1322 : * XXX We do not fail here. Instead, we provide a warning so the user can drop
1323 : * them later.
1324 : */
1325 : static void
1326 8 : drop_failover_replication_slots(struct LogicalRepInfo *dbinfo)
1327 : {
1328 : PGconn *conn;
1329 : PGresult *res;
1330 :
1331 8 : conn = connect_database(dbinfo[0].subconninfo, false);
1332 8 : if (conn != NULL)
1333 : {
1334 : /* Get failover replication slot names */
1335 8 : res = PQexec(conn,
1336 : "SELECT slot_name FROM pg_catalog.pg_replication_slots WHERE failover");
1337 :
1338 8 : if (PQresultStatus(res) == PGRES_TUPLES_OK)
1339 : {
1340 : /* Remove failover replication slots from subscriber */
1341 16 : for (int i = 0; i < PQntuples(res); i++)
1342 8 : drop_replication_slot(conn, &dbinfo[0], PQgetvalue(res, i, 0));
1343 : }
1344 : else
1345 : {
1346 0 : pg_log_warning("could not obtain failover replication slot information: %s",
1347 : PQresultErrorMessage(res));
1348 0 : pg_log_warning_hint("Drop the failover replication slots on subscriber soon to avoid retention of WAL files.");
1349 : }
1350 :
1351 8 : PQclear(res);
1352 8 : disconnect_database(conn, false);
1353 : }
1354 : else
1355 : {
1356 0 : pg_log_warning("could not drop failover replication slot");
1357 0 : pg_log_warning_hint("Drop the failover replication slots on subscriber soon to avoid retention of WAL files.");
1358 : }
1359 8 : }
1360 :
1361 : /*
1362 : * Create a logical replication slot and returns a LSN.
1363 : *
1364 : * CreateReplicationSlot() is not used because it does not provide the one-row
1365 : * result set that contains the LSN.
1366 : */
1367 : static char *
1368 16 : create_logical_replication_slot(PGconn *conn, struct LogicalRepInfo *dbinfo)
1369 : {
1370 16 : PQExpBuffer str = createPQExpBuffer();
1371 16 : PGresult *res = NULL;
1372 16 : const char *slot_name = dbinfo->replslotname;
1373 : char *slot_name_esc;
1374 16 : char *lsn = NULL;
1375 :
1376 : Assert(conn != NULL);
1377 :
1378 16 : pg_log_info("creating the replication slot \"%s\" in database \"%s\"",
1379 : slot_name, dbinfo->dbname);
1380 :
1381 16 : slot_name_esc = PQescapeLiteral(conn, slot_name, strlen(slot_name));
1382 :
1383 16 : appendPQExpBuffer(str,
1384 : "SELECT lsn FROM pg_catalog.pg_create_logical_replication_slot(%s, 'pgoutput', false, %s, false)",
1385 : slot_name_esc,
1386 16 : dbinfos.two_phase ? "true" : "false");
1387 :
1388 16 : PQfreemem(slot_name_esc);
1389 :
1390 16 : pg_log_debug("command is: %s", str->data);
1391 :
1392 16 : if (!dry_run)
1393 : {
1394 4 : res = PQexec(conn, str->data);
1395 4 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
1396 : {
1397 0 : pg_log_error("could not create replication slot \"%s\" in database \"%s\": %s",
1398 : slot_name, dbinfo->dbname,
1399 : PQresultErrorMessage(res));
1400 0 : PQclear(res);
1401 0 : destroyPQExpBuffer(str);
1402 0 : return NULL;
1403 : }
1404 :
1405 4 : lsn = pg_strdup(PQgetvalue(res, 0, 0));
1406 4 : PQclear(res);
1407 : }
1408 :
1409 : /* For cleanup purposes */
1410 16 : dbinfo->made_replslot = true;
1411 :
1412 16 : destroyPQExpBuffer(str);
1413 :
1414 16 : return lsn;
1415 : }
1416 :
1417 : static void
1418 16 : drop_replication_slot(PGconn *conn, struct LogicalRepInfo *dbinfo,
1419 : const char *slot_name)
1420 : {
1421 16 : PQExpBuffer str = createPQExpBuffer();
1422 : char *slot_name_esc;
1423 : PGresult *res;
1424 :
1425 : Assert(conn != NULL);
1426 :
1427 16 : pg_log_info("dropping the replication slot \"%s\" in database \"%s\"",
1428 : slot_name, dbinfo->dbname);
1429 :
1430 16 : slot_name_esc = PQescapeLiteral(conn, slot_name, strlen(slot_name));
1431 :
1432 16 : appendPQExpBuffer(str, "SELECT pg_catalog.pg_drop_replication_slot(%s)", slot_name_esc);
1433 :
1434 16 : PQfreemem(slot_name_esc);
1435 :
1436 16 : pg_log_debug("command is: %s", str->data);
1437 :
1438 16 : if (!dry_run)
1439 : {
1440 4 : res = PQexec(conn, str->data);
1441 4 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
1442 : {
1443 0 : pg_log_error("could not drop replication slot \"%s\" in database \"%s\": %s",
1444 : slot_name, dbinfo->dbname, PQresultErrorMessage(res));
1445 0 : dbinfo->made_replslot = false; /* don't try again. */
1446 : }
1447 :
1448 4 : PQclear(res);
1449 : }
1450 :
1451 16 : destroyPQExpBuffer(str);
1452 16 : }
1453 :
1454 : /*
1455 : * Reports a suitable message if pg_ctl fails.
1456 : */
1457 : static void
1458 48 : pg_ctl_status(const char *pg_ctl_cmd, int rc)
1459 : {
1460 48 : if (rc != 0)
1461 : {
1462 0 : if (WIFEXITED(rc))
1463 : {
1464 0 : pg_log_error("pg_ctl failed with exit code %d", WEXITSTATUS(rc));
1465 : }
1466 0 : else if (WIFSIGNALED(rc))
1467 : {
1468 : #if defined(WIN32)
1469 : pg_log_error("pg_ctl was terminated by exception 0x%X",
1470 : WTERMSIG(rc));
1471 : pg_log_error_detail("See C include file \"ntstatus.h\" for a description of the hexadecimal value.");
1472 : #else
1473 0 : pg_log_error("pg_ctl was terminated by signal %d: %s",
1474 : WTERMSIG(rc), pg_strsignal(WTERMSIG(rc)));
1475 : #endif
1476 : }
1477 : else
1478 : {
1479 0 : pg_log_error("pg_ctl exited with unrecognized status %d", rc);
1480 : }
1481 :
1482 0 : pg_log_error_detail("The failed command was: %s", pg_ctl_cmd);
1483 0 : exit(1);
1484 : }
1485 48 : }
1486 :
1487 : static void
1488 24 : start_standby_server(const struct CreateSubscriberOptions *opt, bool restricted_access,
1489 : bool restrict_logical_worker)
1490 : {
1491 24 : PQExpBuffer pg_ctl_cmd = createPQExpBuffer();
1492 : int rc;
1493 :
1494 24 : appendPQExpBuffer(pg_ctl_cmd, "\"%s\" start -D ", pg_ctl_path);
1495 24 : appendShellString(pg_ctl_cmd, subscriber_dir);
1496 24 : appendPQExpBufferStr(pg_ctl_cmd, " -s -o \"-c sync_replication_slots=off\"");
1497 :
1498 : /* Prevent unintended slot invalidation */
1499 24 : appendPQExpBufferStr(pg_ctl_cmd, " -o \"-c idle_replication_slot_timeout=0\"");
1500 :
1501 24 : if (restricted_access)
1502 : {
1503 24 : appendPQExpBuffer(pg_ctl_cmd, " -o \"-p %s\"", opt->sub_port);
1504 : #if !defined(WIN32)
1505 :
1506 : /*
1507 : * An empty listen_addresses list means the server does not listen on
1508 : * any IP interfaces; only Unix-domain sockets can be used to connect
1509 : * to the server. Prevent external connections to minimize the chance
1510 : * of failure.
1511 : */
1512 24 : appendPQExpBufferStr(pg_ctl_cmd, " -o \"-c listen_addresses='' -c unix_socket_permissions=0700");
1513 24 : if (opt->socket_dir)
1514 24 : appendPQExpBuffer(pg_ctl_cmd, " -c unix_socket_directories='%s'",
1515 24 : opt->socket_dir);
1516 24 : appendPQExpBufferChar(pg_ctl_cmd, '"');
1517 : #endif
1518 : }
1519 24 : if (opt->config_file != NULL)
1520 0 : appendPQExpBuffer(pg_ctl_cmd, " -o \"-c config_file=%s\"",
1521 0 : opt->config_file);
1522 :
1523 : /* Suppress to start logical replication if requested */
1524 24 : if (restrict_logical_worker)
1525 8 : appendPQExpBufferStr(pg_ctl_cmd, " -o \"-c max_logical_replication_workers=0\"");
1526 :
1527 24 : pg_log_debug("pg_ctl command is: %s", pg_ctl_cmd->data);
1528 24 : rc = system(pg_ctl_cmd->data);
1529 24 : pg_ctl_status(pg_ctl_cmd->data, rc);
1530 24 : standby_running = true;
1531 24 : destroyPQExpBuffer(pg_ctl_cmd);
1532 24 : pg_log_info("server was started");
1533 24 : }
1534 :
1535 : static void
1536 24 : stop_standby_server(const char *datadir)
1537 : {
1538 : char *pg_ctl_cmd;
1539 : int rc;
1540 :
1541 24 : pg_ctl_cmd = psprintf("\"%s\" stop -D \"%s\" -s", pg_ctl_path,
1542 : datadir);
1543 24 : pg_log_debug("pg_ctl command is: %s", pg_ctl_cmd);
1544 24 : rc = system(pg_ctl_cmd);
1545 24 : pg_ctl_status(pg_ctl_cmd, rc);
1546 24 : standby_running = false;
1547 24 : pg_log_info("server was stopped");
1548 24 : }
1549 :
1550 : /*
1551 : * Returns after the server finishes the recovery process.
1552 : *
1553 : * If recovery_timeout option is set, terminate abnormally without finishing
1554 : * the recovery process. By default, it waits forever.
1555 : *
1556 : * XXX Is the recovery process still in progress? When recovery process has a
1557 : * better progress reporting mechanism, it should be added here.
1558 : */
1559 : static void
1560 8 : wait_for_end_recovery(const char *conninfo, const struct CreateSubscriberOptions *opt)
1561 : {
1562 : PGconn *conn;
1563 8 : int status = POSTMASTER_STILL_STARTING;
1564 8 : int timer = 0;
1565 :
1566 8 : pg_log_info("waiting for the target server to reach the consistent state");
1567 :
1568 8 : conn = connect_database(conninfo, true);
1569 :
1570 : for (;;)
1571 0 : {
1572 8 : bool in_recovery = server_is_in_recovery(conn);
1573 :
1574 : /*
1575 : * Does the recovery process finish? In dry run mode, there is no
1576 : * recovery mode. Bail out as the recovery process has ended.
1577 : */
1578 8 : if (!in_recovery || dry_run)
1579 : {
1580 8 : status = POSTMASTER_READY;
1581 8 : recovery_ended = true;
1582 8 : break;
1583 : }
1584 :
1585 : /* Bail out after recovery_timeout seconds if this option is set */
1586 0 : if (opt->recovery_timeout > 0 && timer >= opt->recovery_timeout)
1587 : {
1588 0 : stop_standby_server(subscriber_dir);
1589 0 : pg_log_error("recovery timed out");
1590 0 : disconnect_database(conn, true);
1591 : }
1592 :
1593 : /* Keep waiting */
1594 0 : pg_usleep(WAIT_INTERVAL * USEC_PER_SEC);
1595 :
1596 0 : timer += WAIT_INTERVAL;
1597 : }
1598 :
1599 8 : disconnect_database(conn, false);
1600 :
1601 8 : if (status == POSTMASTER_STILL_STARTING)
1602 0 : pg_fatal("server did not end recovery");
1603 :
1604 8 : pg_log_info("target server reached the consistent state");
1605 8 : pg_log_info_hint("If pg_createsubscriber fails after this point, you must recreate the physical replica before continuing.");
1606 8 : }
1607 :
1608 : /*
1609 : * Create a publication that includes all tables in the database.
1610 : */
1611 : static void
1612 16 : create_publication(PGconn *conn, struct LogicalRepInfo *dbinfo)
1613 : {
1614 16 : PQExpBuffer str = createPQExpBuffer();
1615 : PGresult *res;
1616 : char *ipubname_esc;
1617 : char *spubname_esc;
1618 :
1619 : Assert(conn != NULL);
1620 :
1621 16 : ipubname_esc = PQescapeIdentifier(conn, dbinfo->pubname, strlen(dbinfo->pubname));
1622 16 : spubname_esc = PQescapeLiteral(conn, dbinfo->pubname, strlen(dbinfo->pubname));
1623 :
1624 : /* Check if the publication already exists */
1625 16 : appendPQExpBuffer(str,
1626 : "SELECT 1 FROM pg_catalog.pg_publication "
1627 : "WHERE pubname = %s",
1628 : spubname_esc);
1629 16 : res = PQexec(conn, str->data);
1630 16 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
1631 : {
1632 0 : pg_log_error("could not obtain publication information: %s",
1633 : PQresultErrorMessage(res));
1634 0 : disconnect_database(conn, true);
1635 : }
1636 :
1637 16 : if (PQntuples(res) == 1)
1638 : {
1639 : /*
1640 : * Unfortunately, if it reaches this code path, it will always fail
1641 : * (unless you decide to change the existing publication name). That's
1642 : * bad but it is very unlikely that the user will choose a name with
1643 : * pg_createsubscriber_ prefix followed by the exact database oid and
1644 : * a random number.
1645 : */
1646 0 : pg_log_error("publication \"%s\" already exists", dbinfo->pubname);
1647 0 : pg_log_error_hint("Consider renaming this publication before continuing.");
1648 0 : disconnect_database(conn, true);
1649 : }
1650 :
1651 16 : PQclear(res);
1652 16 : resetPQExpBuffer(str);
1653 :
1654 16 : pg_log_info("creating publication \"%s\" in database \"%s\"",
1655 : dbinfo->pubname, dbinfo->dbname);
1656 :
1657 16 : appendPQExpBuffer(str, "CREATE PUBLICATION %s FOR ALL TABLES",
1658 : ipubname_esc);
1659 :
1660 16 : pg_log_debug("command is: %s", str->data);
1661 :
1662 16 : if (!dry_run)
1663 : {
1664 4 : res = PQexec(conn, str->data);
1665 4 : if (PQresultStatus(res) != PGRES_COMMAND_OK)
1666 : {
1667 0 : pg_log_error("could not create publication \"%s\" in database \"%s\": %s",
1668 : dbinfo->pubname, dbinfo->dbname, PQresultErrorMessage(res));
1669 0 : disconnect_database(conn, true);
1670 : }
1671 4 : PQclear(res);
1672 : }
1673 :
1674 : /* For cleanup purposes */
1675 16 : dbinfo->made_publication = true;
1676 :
1677 16 : PQfreemem(ipubname_esc);
1678 16 : PQfreemem(spubname_esc);
1679 16 : destroyPQExpBuffer(str);
1680 16 : }
1681 :
1682 : /*
1683 : * Drop the specified publication in the given database.
1684 : */
1685 : static void
1686 20 : drop_publication(PGconn *conn, const char *pubname, const char *dbname,
1687 : bool *made_publication)
1688 : {
1689 20 : PQExpBuffer str = createPQExpBuffer();
1690 : PGresult *res;
1691 : char *pubname_esc;
1692 :
1693 : Assert(conn != NULL);
1694 :
1695 20 : pubname_esc = PQescapeIdentifier(conn, pubname, strlen(pubname));
1696 :
1697 20 : pg_log_info("dropping publication \"%s\" in database \"%s\"",
1698 : pubname, dbname);
1699 :
1700 20 : appendPQExpBuffer(str, "DROP PUBLICATION %s", pubname_esc);
1701 :
1702 20 : PQfreemem(pubname_esc);
1703 :
1704 20 : pg_log_debug("command is: %s", str->data);
1705 :
1706 20 : if (!dry_run)
1707 : {
1708 8 : res = PQexec(conn, str->data);
1709 8 : if (PQresultStatus(res) != PGRES_COMMAND_OK)
1710 : {
1711 0 : pg_log_error("could not drop publication \"%s\" in database \"%s\": %s",
1712 : pubname, dbname, PQresultErrorMessage(res));
1713 0 : *made_publication = false; /* don't try again. */
1714 :
1715 : /*
1716 : * Don't disconnect and exit here. This routine is used by primary
1717 : * (cleanup publication / replication slot due to an error) and
1718 : * subscriber (remove the replicated publications). In both cases,
1719 : * it can continue and provide instructions for the user to remove
1720 : * it later if cleanup fails.
1721 : */
1722 : }
1723 8 : PQclear(res);
1724 : }
1725 :
1726 20 : destroyPQExpBuffer(str);
1727 20 : }
1728 :
1729 : /*
1730 : * Retrieve and drop the publications.
1731 : *
1732 : * Since the publications were created before the consistent LSN, they
1733 : * remain on the subscriber even after the physical replica is
1734 : * promoted. Remove these publications from the subscriber because
1735 : * they have no use. Additionally, if requested, drop all pre-existing
1736 : * publications.
1737 : */
1738 : static void
1739 16 : check_and_drop_publications(PGconn *conn, struct LogicalRepInfo *dbinfo)
1740 : {
1741 : PGresult *res;
1742 16 : bool drop_all_pubs = dbinfos.objecttypes_to_clean & OBJECTTYPE_PUBLICATIONS;
1743 :
1744 : Assert(conn != NULL);
1745 :
1746 16 : if (drop_all_pubs)
1747 : {
1748 4 : pg_log_info("dropping all existing publications in database \"%s\"",
1749 : dbinfo->dbname);
1750 :
1751 : /* Fetch all publication names */
1752 4 : res = PQexec(conn, "SELECT pubname FROM pg_catalog.pg_publication;");
1753 4 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
1754 : {
1755 0 : pg_log_error("could not obtain publication information: %s",
1756 : PQresultErrorMessage(res));
1757 0 : PQclear(res);
1758 0 : disconnect_database(conn, true);
1759 : }
1760 :
1761 : /* Drop each publication */
1762 12 : for (int i = 0; i < PQntuples(res); i++)
1763 8 : drop_publication(conn, PQgetvalue(res, i, 0), dbinfo->dbname,
1764 : &dbinfo->made_publication);
1765 :
1766 4 : PQclear(res);
1767 : }
1768 :
1769 : /*
1770 : * In dry-run mode, we don't create publications, but we still try to drop
1771 : * those to provide necessary information to the user.
1772 : */
1773 16 : if (!drop_all_pubs || dry_run)
1774 12 : drop_publication(conn, dbinfo->pubname, dbinfo->dbname,
1775 : &dbinfo->made_publication);
1776 16 : }
1777 :
1778 : /*
1779 : * Create a subscription with some predefined options.
1780 : *
1781 : * A replication slot was already created in a previous step. Let's use it. It
1782 : * is not required to copy data. The subscription will be created but it will
1783 : * not be enabled now. That's because the replication progress must be set and
1784 : * the replication origin name (one of the function arguments) contains the
1785 : * subscription OID in its name. Once the subscription is created,
1786 : * set_replication_progress() can obtain the chosen origin name and set up its
1787 : * initial location.
1788 : */
1789 : static void
1790 16 : create_subscription(PGconn *conn, const struct LogicalRepInfo *dbinfo)
1791 : {
1792 16 : PQExpBuffer str = createPQExpBuffer();
1793 : PGresult *res;
1794 : char *pubname_esc;
1795 : char *subname_esc;
1796 : char *pubconninfo_esc;
1797 : char *replslotname_esc;
1798 :
1799 : Assert(conn != NULL);
1800 :
1801 16 : pubname_esc = PQescapeIdentifier(conn, dbinfo->pubname, strlen(dbinfo->pubname));
1802 16 : subname_esc = PQescapeIdentifier(conn, dbinfo->subname, strlen(dbinfo->subname));
1803 16 : pubconninfo_esc = PQescapeLiteral(conn, dbinfo->pubconninfo, strlen(dbinfo->pubconninfo));
1804 16 : replslotname_esc = PQescapeLiteral(conn, dbinfo->replslotname, strlen(dbinfo->replslotname));
1805 :
1806 16 : pg_log_info("creating subscription \"%s\" in database \"%s\"",
1807 : dbinfo->subname, dbinfo->dbname);
1808 :
1809 16 : appendPQExpBuffer(str,
1810 : "CREATE SUBSCRIPTION %s CONNECTION %s PUBLICATION %s "
1811 : "WITH (create_slot = false, enabled = false, "
1812 : "slot_name = %s, copy_data = false, two_phase = %s)",
1813 : subname_esc, pubconninfo_esc, pubname_esc, replslotname_esc,
1814 16 : dbinfos.two_phase ? "true" : "false");
1815 :
1816 16 : PQfreemem(pubname_esc);
1817 16 : PQfreemem(subname_esc);
1818 16 : PQfreemem(pubconninfo_esc);
1819 16 : PQfreemem(replslotname_esc);
1820 :
1821 16 : pg_log_debug("command is: %s", str->data);
1822 :
1823 16 : if (!dry_run)
1824 : {
1825 4 : res = PQexec(conn, str->data);
1826 4 : if (PQresultStatus(res) != PGRES_COMMAND_OK)
1827 : {
1828 0 : pg_log_error("could not create subscription \"%s\" in database \"%s\": %s",
1829 : dbinfo->subname, dbinfo->dbname, PQresultErrorMessage(res));
1830 0 : disconnect_database(conn, true);
1831 : }
1832 4 : PQclear(res);
1833 : }
1834 :
1835 16 : destroyPQExpBuffer(str);
1836 16 : }
1837 :
1838 : /*
1839 : * Sets the replication progress to the consistent LSN.
1840 : *
1841 : * The subscriber caught up to the consistent LSN provided by the last
1842 : * replication slot that was created. The goal is to set up the initial
1843 : * location for the logical replication that is the exact LSN that the
1844 : * subscriber was promoted. Once the subscription is enabled it will start
1845 : * streaming from that location onwards. In dry run mode, the subscription OID
1846 : * and LSN are set to invalid values for printing purposes.
1847 : */
1848 : static void
1849 16 : set_replication_progress(PGconn *conn, const struct LogicalRepInfo *dbinfo, const char *lsn)
1850 : {
1851 16 : PQExpBuffer str = createPQExpBuffer();
1852 : PGresult *res;
1853 : Oid suboid;
1854 : char *subname;
1855 : char *dbname;
1856 : char *originname;
1857 : char *lsnstr;
1858 :
1859 : Assert(conn != NULL);
1860 :
1861 16 : subname = PQescapeLiteral(conn, dbinfo->subname, strlen(dbinfo->subname));
1862 16 : dbname = PQescapeLiteral(conn, dbinfo->dbname, strlen(dbinfo->dbname));
1863 :
1864 16 : appendPQExpBuffer(str,
1865 : "SELECT s.oid FROM pg_catalog.pg_subscription s "
1866 : "INNER JOIN pg_catalog.pg_database d ON (s.subdbid = d.oid) "
1867 : "WHERE s.subname = %s AND d.datname = %s",
1868 : subname, dbname);
1869 :
1870 16 : res = PQexec(conn, str->data);
1871 16 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
1872 : {
1873 0 : pg_log_error("could not obtain subscription OID: %s",
1874 : PQresultErrorMessage(res));
1875 0 : disconnect_database(conn, true);
1876 : }
1877 :
1878 16 : if (PQntuples(res) != 1 && !dry_run)
1879 : {
1880 0 : pg_log_error("could not obtain subscription OID: got %d rows, expected %d row",
1881 : PQntuples(res), 1);
1882 0 : disconnect_database(conn, true);
1883 : }
1884 :
1885 16 : if (dry_run)
1886 : {
1887 12 : suboid = InvalidOid;
1888 12 : lsnstr = psprintf("%X/%08X", LSN_FORMAT_ARGS((XLogRecPtr) InvalidXLogRecPtr));
1889 : }
1890 : else
1891 : {
1892 4 : suboid = strtoul(PQgetvalue(res, 0, 0), NULL, 10);
1893 4 : lsnstr = psprintf("%s", lsn);
1894 : }
1895 :
1896 16 : PQclear(res);
1897 :
1898 : /*
1899 : * The origin name is defined as pg_%u. %u is the subscription OID. See
1900 : * ApplyWorkerMain().
1901 : */
1902 16 : originname = psprintf("pg_%u", suboid);
1903 :
1904 16 : pg_log_info("setting the replication progress (node name \"%s\", LSN %s) in database \"%s\"",
1905 : originname, lsnstr, dbinfo->dbname);
1906 :
1907 16 : resetPQExpBuffer(str);
1908 16 : appendPQExpBuffer(str,
1909 : "SELECT pg_catalog.pg_replication_origin_advance('%s', '%s')",
1910 : originname, lsnstr);
1911 :
1912 16 : pg_log_debug("command is: %s", str->data);
1913 :
1914 16 : if (!dry_run)
1915 : {
1916 4 : res = PQexec(conn, str->data);
1917 4 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
1918 : {
1919 0 : pg_log_error("could not set replication progress for subscription \"%s\": %s",
1920 : dbinfo->subname, PQresultErrorMessage(res));
1921 0 : disconnect_database(conn, true);
1922 : }
1923 4 : PQclear(res);
1924 : }
1925 :
1926 16 : PQfreemem(subname);
1927 16 : PQfreemem(dbname);
1928 16 : pg_free(originname);
1929 16 : pg_free(lsnstr);
1930 16 : destroyPQExpBuffer(str);
1931 16 : }
1932 :
1933 : /*
1934 : * Enables the subscription.
1935 : *
1936 : * The subscription was created in a previous step but it was disabled. After
1937 : * adjusting the initial logical replication location, enable the subscription.
1938 : */
1939 : static void
1940 16 : enable_subscription(PGconn *conn, const struct LogicalRepInfo *dbinfo)
1941 : {
1942 16 : PQExpBuffer str = createPQExpBuffer();
1943 : PGresult *res;
1944 : char *subname;
1945 :
1946 : Assert(conn != NULL);
1947 :
1948 16 : subname = PQescapeIdentifier(conn, dbinfo->subname, strlen(dbinfo->subname));
1949 :
1950 16 : pg_log_info("enabling subscription \"%s\" in database \"%s\"",
1951 : dbinfo->subname, dbinfo->dbname);
1952 :
1953 16 : appendPQExpBuffer(str, "ALTER SUBSCRIPTION %s ENABLE", subname);
1954 :
1955 16 : pg_log_debug("command is: %s", str->data);
1956 :
1957 16 : if (!dry_run)
1958 : {
1959 4 : res = PQexec(conn, str->data);
1960 4 : if (PQresultStatus(res) != PGRES_COMMAND_OK)
1961 : {
1962 0 : pg_log_error("could not enable subscription \"%s\": %s",
1963 : dbinfo->subname, PQresultErrorMessage(res));
1964 0 : disconnect_database(conn, true);
1965 : }
1966 :
1967 4 : PQclear(res);
1968 : }
1969 :
1970 16 : PQfreemem(subname);
1971 16 : destroyPQExpBuffer(str);
1972 16 : }
1973 :
1974 : /*
1975 : * Fetch a list of all connectable non-template databases from the source server
1976 : * and form a list such that they appear as if the user has specified multiple
1977 : * --database options, one for each source database.
1978 : */
1979 : static void
1980 2 : get_publisher_databases(struct CreateSubscriberOptions *opt,
1981 : bool dbnamespecified)
1982 : {
1983 : PGconn *conn;
1984 : PGresult *res;
1985 :
1986 : /* If a database name was specified, just connect to it. */
1987 2 : if (dbnamespecified)
1988 0 : conn = connect_database(opt->pub_conninfo_str, true);
1989 : else
1990 : {
1991 : /* Otherwise, try postgres first and then template1. */
1992 : char *conninfo;
1993 :
1994 2 : conninfo = concat_conninfo_dbname(opt->pub_conninfo_str, "postgres");
1995 2 : conn = connect_database(conninfo, false);
1996 2 : pg_free(conninfo);
1997 2 : if (!conn)
1998 : {
1999 0 : conninfo = concat_conninfo_dbname(opt->pub_conninfo_str, "template1");
2000 0 : conn = connect_database(conninfo, true);
2001 0 : pg_free(conninfo);
2002 : }
2003 : }
2004 :
2005 2 : res = PQexec(conn, "SELECT datname FROM pg_database WHERE datistemplate = false AND datallowconn AND datconnlimit <> -2 ORDER BY 1");
2006 2 : if (PQresultStatus(res) != PGRES_TUPLES_OK)
2007 : {
2008 0 : pg_log_error("could not obtain a list of databases: %s", PQresultErrorMessage(res));
2009 0 : PQclear(res);
2010 0 : disconnect_database(conn, true);
2011 : }
2012 :
2013 8 : for (int i = 0; i < PQntuples(res); i++)
2014 : {
2015 6 : const char *dbname = PQgetvalue(res, i, 0);
2016 :
2017 6 : simple_string_list_append(&opt->database_names, dbname);
2018 :
2019 : /* Increment num_dbs to reflect multiple --database options */
2020 6 : num_dbs++;
2021 : }
2022 :
2023 2 : PQclear(res);
2024 2 : disconnect_database(conn, false);
2025 2 : }
2026 :
2027 : int
2028 46 : main(int argc, char **argv)
2029 : {
2030 : static struct option long_options[] =
2031 : {
2032 : {"all", no_argument, NULL, 'a'},
2033 : {"database", required_argument, NULL, 'd'},
2034 : {"pgdata", required_argument, NULL, 'D'},
2035 : {"dry-run", no_argument, NULL, 'n'},
2036 : {"subscriber-port", required_argument, NULL, 'p'},
2037 : {"publisher-server", required_argument, NULL, 'P'},
2038 : {"socketdir", required_argument, NULL, 's'},
2039 : {"recovery-timeout", required_argument, NULL, 't'},
2040 : {"enable-two-phase", no_argument, NULL, 'T'},
2041 : {"subscriber-username", required_argument, NULL, 'U'},
2042 : {"verbose", no_argument, NULL, 'v'},
2043 : {"version", no_argument, NULL, 'V'},
2044 : {"help", no_argument, NULL, '?'},
2045 : {"config-file", required_argument, NULL, 1},
2046 : {"publication", required_argument, NULL, 2},
2047 : {"replication-slot", required_argument, NULL, 3},
2048 : {"subscription", required_argument, NULL, 4},
2049 : {"clean", required_argument, NULL, 5},
2050 : {NULL, 0, NULL, 0}
2051 : };
2052 :
2053 46 : struct CreateSubscriberOptions opt = {0};
2054 :
2055 : int c;
2056 : int option_index;
2057 :
2058 : char *pub_base_conninfo;
2059 : char *sub_base_conninfo;
2060 46 : char *dbname_conninfo = NULL;
2061 :
2062 : uint64 pub_sysid;
2063 : uint64 sub_sysid;
2064 : struct stat statbuf;
2065 :
2066 : char *consistent_lsn;
2067 :
2068 : char pidfile[MAXPGPATH];
2069 :
2070 46 : pg_logging_init(argv[0]);
2071 46 : pg_logging_set_level(PG_LOG_WARNING);
2072 46 : progname = get_progname(argv[0]);
2073 46 : set_pglocale_pgservice(argv[0], PG_TEXTDOMAIN("pg_basebackup"));
2074 :
2075 46 : if (argc > 1)
2076 : {
2077 44 : if (strcmp(argv[1], "--help") == 0 || strcmp(argv[1], "-?") == 0)
2078 : {
2079 2 : usage();
2080 2 : exit(0);
2081 : }
2082 42 : else if (strcmp(argv[1], "-V") == 0
2083 42 : || strcmp(argv[1], "--version") == 0)
2084 : {
2085 2 : puts("pg_createsubscriber (PostgreSQL) " PG_VERSION);
2086 2 : exit(0);
2087 : }
2088 : }
2089 :
2090 : /* Default settings */
2091 42 : subscriber_dir = NULL;
2092 42 : opt.config_file = NULL;
2093 42 : opt.pub_conninfo_str = NULL;
2094 42 : opt.socket_dir = NULL;
2095 42 : opt.sub_port = DEFAULT_SUB_PORT;
2096 42 : opt.sub_username = NULL;
2097 42 : opt.two_phase = false;
2098 42 : opt.database_names = (SimpleStringList)
2099 : {
2100 : 0
2101 : };
2102 42 : opt.recovery_timeout = 0;
2103 42 : opt.all_dbs = false;
2104 :
2105 : /*
2106 : * Don't allow it to be run as root. It uses pg_ctl which does not allow
2107 : * it either.
2108 : */
2109 : #ifndef WIN32
2110 42 : if (geteuid() == 0)
2111 : {
2112 0 : pg_log_error("cannot be executed by \"root\"");
2113 0 : pg_log_error_hint("You must run %s as the PostgreSQL superuser.",
2114 : progname);
2115 0 : exit(1);
2116 : }
2117 : #endif
2118 :
2119 42 : get_restricted_token();
2120 :
2121 324 : while ((c = getopt_long(argc, argv, "ad:D:np:P:s:t:TU:v",
2122 324 : long_options, &option_index)) != -1)
2123 : {
2124 288 : switch (c)
2125 : {
2126 6 : case 'a':
2127 6 : opt.all_dbs = true;
2128 6 : break;
2129 50 : case 'd':
2130 50 : if (!simple_string_list_member(&opt.database_names, optarg))
2131 : {
2132 48 : simple_string_list_append(&opt.database_names, optarg);
2133 48 : num_dbs++;
2134 : }
2135 : else
2136 2 : pg_fatal("database \"%s\" specified more than once for -d/--database", optarg);
2137 48 : break;
2138 38 : case 'D':
2139 38 : subscriber_dir = pg_strdup(optarg);
2140 38 : canonicalize_path(subscriber_dir);
2141 38 : break;
2142 18 : case 'n':
2143 18 : dry_run = true;
2144 18 : break;
2145 24 : case 'p':
2146 24 : opt.sub_port = pg_strdup(optarg);
2147 24 : break;
2148 36 : case 'P':
2149 36 : opt.pub_conninfo_str = pg_strdup(optarg);
2150 36 : break;
2151 24 : case 's':
2152 24 : opt.socket_dir = pg_strdup(optarg);
2153 24 : canonicalize_path(opt.socket_dir);
2154 24 : break;
2155 6 : case 't':
2156 6 : opt.recovery_timeout = atoi(optarg);
2157 6 : break;
2158 2 : case 'T':
2159 2 : opt.two_phase = true;
2160 2 : break;
2161 0 : case 'U':
2162 0 : opt.sub_username = pg_strdup(optarg);
2163 0 : break;
2164 38 : case 'v':
2165 38 : pg_logging_increase_verbosity();
2166 38 : break;
2167 0 : case 1:
2168 0 : opt.config_file = pg_strdup(optarg);
2169 0 : break;
2170 24 : case 2:
2171 24 : if (!simple_string_list_member(&opt.pub_names, optarg))
2172 : {
2173 22 : simple_string_list_append(&opt.pub_names, optarg);
2174 22 : num_pubs++;
2175 : }
2176 : else
2177 2 : pg_fatal("publication \"%s\" specified more than once for --publication", optarg);
2178 22 : break;
2179 8 : case 3:
2180 8 : if (!simple_string_list_member(&opt.replslot_names, optarg))
2181 : {
2182 8 : simple_string_list_append(&opt.replslot_names, optarg);
2183 8 : num_replslots++;
2184 : }
2185 : else
2186 0 : pg_fatal("replication slot \"%s\" specified more than once for --replication-slot", optarg);
2187 8 : break;
2188 10 : case 4:
2189 10 : if (!simple_string_list_member(&opt.sub_names, optarg))
2190 : {
2191 10 : simple_string_list_append(&opt.sub_names, optarg);
2192 10 : num_subs++;
2193 : }
2194 : else
2195 0 : pg_fatal("subscription \"%s\" specified more than once for --subscription", optarg);
2196 10 : break;
2197 2 : case 5:
2198 2 : if (!simple_string_list_member(&opt.objecttypes_to_clean, optarg))
2199 2 : simple_string_list_append(&opt.objecttypes_to_clean, optarg);
2200 : else
2201 0 : pg_fatal("object type \"%s\" specified more than once for --clean", optarg);
2202 2 : break;
2203 2 : default:
2204 : /* getopt_long already emitted a complaint */
2205 2 : pg_log_error_hint("Try \"%s --help\" for more information.", progname);
2206 2 : exit(1);
2207 : }
2208 : }
2209 :
2210 : /* Validate that --all is not used with incompatible options */
2211 36 : if (opt.all_dbs)
2212 : {
2213 6 : char *bad_switch = NULL;
2214 :
2215 6 : if (num_dbs > 0)
2216 2 : bad_switch = "--database";
2217 4 : else if (num_pubs > 0)
2218 2 : bad_switch = "--publication";
2219 2 : else if (num_replslots > 0)
2220 0 : bad_switch = "--replication-slot";
2221 2 : else if (num_subs > 0)
2222 0 : bad_switch = "--subscription";
2223 :
2224 6 : if (bad_switch)
2225 : {
2226 4 : pg_log_error("options %s and -a/--all cannot be used together", bad_switch);
2227 4 : pg_log_error_hint("Try \"%s --help\" for more information.", progname);
2228 4 : exit(1);
2229 : }
2230 : }
2231 :
2232 : /* Any non-option arguments? */
2233 32 : if (optind < argc)
2234 : {
2235 0 : pg_log_error("too many command-line arguments (first is \"%s\")",
2236 : argv[optind]);
2237 0 : pg_log_error_hint("Try \"%s --help\" for more information.", progname);
2238 0 : exit(1);
2239 : }
2240 :
2241 : /* Required arguments */
2242 32 : if (subscriber_dir == NULL)
2243 : {
2244 2 : pg_log_error("no subscriber data directory specified");
2245 2 : pg_log_error_hint("Try \"%s --help\" for more information.", progname);
2246 2 : exit(1);
2247 : }
2248 :
2249 : /* If socket directory is not provided, use the current directory */
2250 30 : if (opt.socket_dir == NULL)
2251 : {
2252 : char cwd[MAXPGPATH];
2253 :
2254 10 : if (!getcwd(cwd, MAXPGPATH))
2255 0 : pg_fatal("could not determine current directory");
2256 10 : opt.socket_dir = pg_strdup(cwd);
2257 10 : canonicalize_path(opt.socket_dir);
2258 : }
2259 :
2260 : /*
2261 : * Parse connection string. Build a base connection string that might be
2262 : * reused by multiple databases.
2263 : */
2264 30 : if (opt.pub_conninfo_str == NULL)
2265 : {
2266 : /*
2267 : * TODO use primary_conninfo (if available) from subscriber and
2268 : * extract publisher connection string. Assume that there are
2269 : * identical entries for physical and logical replication. If there is
2270 : * not, we would fail anyway.
2271 : */
2272 2 : pg_log_error("no publisher connection string specified");
2273 2 : pg_log_error_hint("Try \"%s --help\" for more information.", progname);
2274 2 : exit(1);
2275 : }
2276 28 : pg_log_info("validating publisher connection string");
2277 28 : pub_base_conninfo = get_base_conninfo(opt.pub_conninfo_str,
2278 : &dbname_conninfo);
2279 28 : if (pub_base_conninfo == NULL)
2280 0 : exit(1);
2281 :
2282 28 : pg_log_info("validating subscriber connection string");
2283 28 : sub_base_conninfo = get_sub_conninfo(&opt);
2284 :
2285 : /*
2286 : * Fetch all databases from the source (publisher) and treat them as if
2287 : * the user specified has multiple --database options, one for each source
2288 : * database.
2289 : */
2290 28 : if (opt.all_dbs)
2291 : {
2292 2 : bool dbnamespecified = (dbname_conninfo != NULL);
2293 :
2294 2 : get_publisher_databases(&opt, dbnamespecified);
2295 : }
2296 :
2297 28 : if (opt.database_names.head == NULL)
2298 : {
2299 4 : pg_log_info("no database was specified");
2300 :
2301 : /*
2302 : * Try to obtain the dbname from the publisher conninfo. If dbname
2303 : * parameter is not available, error out.
2304 : */
2305 4 : if (dbname_conninfo)
2306 : {
2307 2 : simple_string_list_append(&opt.database_names, dbname_conninfo);
2308 2 : num_dbs++;
2309 :
2310 2 : pg_log_info("database name \"%s\" was extracted from the publisher connection string",
2311 : dbname_conninfo);
2312 : }
2313 : else
2314 : {
2315 2 : pg_log_error("no database name specified");
2316 2 : pg_log_error_hint("Try \"%s --help\" for more information.",
2317 : progname);
2318 2 : exit(1);
2319 : }
2320 : }
2321 :
2322 : /* Number of object names must match number of databases */
2323 26 : if (num_pubs > 0 && num_pubs != num_dbs)
2324 : {
2325 2 : pg_log_error("wrong number of publication names specified");
2326 2 : pg_log_error_detail("The number of specified publication names (%d) must match the number of specified database names (%d).",
2327 : num_pubs, num_dbs);
2328 2 : exit(1);
2329 : }
2330 24 : if (num_subs > 0 && num_subs != num_dbs)
2331 : {
2332 2 : pg_log_error("wrong number of subscription names specified");
2333 2 : pg_log_error_detail("The number of specified subscription names (%d) must match the number of specified database names (%d).",
2334 : num_subs, num_dbs);
2335 2 : exit(1);
2336 : }
2337 22 : if (num_replslots > 0 && num_replslots != num_dbs)
2338 : {
2339 2 : pg_log_error("wrong number of replication slot names specified");
2340 2 : pg_log_error_detail("The number of specified replication slot names (%d) must match the number of specified database names (%d).",
2341 : num_replslots, num_dbs);
2342 2 : exit(1);
2343 : }
2344 :
2345 : /* Verify the object types specified for removal from the subscriber */
2346 22 : for (SimpleStringListCell *cell = opt.objecttypes_to_clean.head; cell; cell = cell->next)
2347 : {
2348 2 : if (pg_strcasecmp(cell->val, "publications") == 0)
2349 2 : dbinfos.objecttypes_to_clean |= OBJECTTYPE_PUBLICATIONS;
2350 : else
2351 : {
2352 0 : pg_log_error("invalid object type \"%s\" specified for --clean", cell->val);
2353 0 : pg_log_error_hint("The valid value is: \"%s\"", "publications");
2354 0 : exit(1);
2355 : }
2356 : }
2357 :
2358 : /* Get the absolute path of pg_ctl and pg_resetwal on the subscriber */
2359 20 : pg_ctl_path = get_exec_path(argv[0], "pg_ctl");
2360 20 : pg_resetwal_path = get_exec_path(argv[0], "pg_resetwal");
2361 :
2362 : /* Rudimentary check for a data directory */
2363 20 : check_data_directory(subscriber_dir);
2364 :
2365 20 : dbinfos.two_phase = opt.two_phase;
2366 :
2367 : /*
2368 : * Store database information for publisher and subscriber. It should be
2369 : * called before atexit() because its return is used in the
2370 : * cleanup_objects_atexit().
2371 : */
2372 20 : dbinfos.dbinfo = store_pub_sub_info(&opt, pub_base_conninfo, sub_base_conninfo);
2373 :
2374 : /* Register a function to clean up objects in case of failure */
2375 20 : atexit(cleanup_objects_atexit);
2376 :
2377 : /*
2378 : * Check if the subscriber data directory has the same system identifier
2379 : * than the publisher data directory.
2380 : */
2381 20 : pub_sysid = get_primary_sysid(dbinfos.dbinfo[0].pubconninfo);
2382 20 : sub_sysid = get_standby_sysid(subscriber_dir);
2383 20 : if (pub_sysid != sub_sysid)
2384 2 : pg_fatal("subscriber data directory is not a copy of the source database cluster");
2385 :
2386 : /* Subscriber PID file */
2387 18 : snprintf(pidfile, MAXPGPATH, "%s/postmaster.pid", subscriber_dir);
2388 :
2389 : /*
2390 : * The standby server must not be running. If the server is started under
2391 : * service manager and pg_createsubscriber stops it, the service manager
2392 : * might react to this action and start the server again. Therefore,
2393 : * refuse to proceed if the server is running to avoid possible failures.
2394 : */
2395 18 : if (stat(pidfile, &statbuf) == 0)
2396 : {
2397 2 : pg_log_error("standby server is running");
2398 2 : pg_log_error_hint("Stop the standby server and try again.");
2399 2 : exit(1);
2400 : }
2401 :
2402 : /*
2403 : * Start a short-lived standby server with temporary parameters (provided
2404 : * by command-line options). The goal is to avoid connections during the
2405 : * transformation steps.
2406 : */
2407 16 : pg_log_info("starting the standby server with command-line options");
2408 16 : start_standby_server(&opt, true, false);
2409 :
2410 : /* Check if the standby server is ready for logical replication */
2411 16 : check_subscriber(dbinfos.dbinfo);
2412 :
2413 : /* Check if the primary server is ready for logical replication */
2414 12 : check_publisher(dbinfos.dbinfo);
2415 :
2416 : /*
2417 : * Stop the target server. The recovery process requires that the server
2418 : * reaches a consistent state before targeting the recovery stop point.
2419 : * Make sure a consistent state is reached (stop the target server
2420 : * guarantees it) *before* creating the replication slots in
2421 : * setup_publisher().
2422 : */
2423 8 : pg_log_info("stopping the subscriber");
2424 8 : stop_standby_server(subscriber_dir);
2425 :
2426 : /* Create the required objects for each database on publisher */
2427 8 : consistent_lsn = setup_publisher(dbinfos.dbinfo);
2428 :
2429 : /* Write the required recovery parameters */
2430 8 : setup_recovery(dbinfos.dbinfo, subscriber_dir, consistent_lsn);
2431 :
2432 : /*
2433 : * Start subscriber so the recovery parameters will take effect. Wait
2434 : * until accepting connections. We don't want to start logical replication
2435 : * during setup.
2436 : */
2437 8 : pg_log_info("starting the subscriber");
2438 8 : start_standby_server(&opt, true, true);
2439 :
2440 : /* Waiting the subscriber to be promoted */
2441 8 : wait_for_end_recovery(dbinfos.dbinfo[0].subconninfo, &opt);
2442 :
2443 : /*
2444 : * Create the subscription for each database on subscriber. It does not
2445 : * enable it immediately because it needs to adjust the replication start
2446 : * point to the LSN reported by setup_publisher(). It also cleans up
2447 : * publications created by this tool and replication to the standby.
2448 : */
2449 8 : setup_subscriber(dbinfos.dbinfo, consistent_lsn);
2450 :
2451 : /* Remove primary_slot_name if it exists on primary */
2452 8 : drop_primary_replication_slot(dbinfos.dbinfo, primary_slot_name);
2453 :
2454 : /* Remove failover replication slots if they exist on subscriber */
2455 8 : drop_failover_replication_slots(dbinfos.dbinfo);
2456 :
2457 : /* Stop the subscriber */
2458 8 : pg_log_info("stopping the subscriber");
2459 8 : stop_standby_server(subscriber_dir);
2460 :
2461 : /* Change system identifier from subscriber */
2462 8 : modify_subscriber_sysid(&opt);
2463 :
2464 8 : success = true;
2465 :
2466 8 : pg_log_info("Done!");
2467 :
2468 8 : return 0;
2469 : }
|
