1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
|
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>55.4. Streaming Replication Protocol</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="sasl-authentication.html" title="55.3. SASL Authentication" /><link rel="next" href="protocol-logical-replication.html" title="55.5. Logical Streaming Replication Protocol" /></head><body id="docContent" class="container-fluid col-10"><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">55.4. Streaming Replication Protocol</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="sasl-authentication.html" title="55.3. SASL Authentication">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="protocol.html" title="Chapter 55. Frontend/Backend Protocol">Up</a></td><th width="60%" align="center">Chapter 55. Frontend/Backend Protocol</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 16.2 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="protocol-logical-replication.html" title="55.5. Logical Streaming Replication Protocol">Next</a></td></tr></table><hr /></div><div class="sect1" id="PROTOCOL-REPLICATION"><div class="titlepage"><div><div><h2 class="title" style="clear: both">55.4. Streaming Replication Protocol <a href="#PROTOCOL-REPLICATION" class="id_link">#</a></h2></div></div></div><p>
To initiate streaming replication, the frontend sends the
<code class="literal">replication</code> parameter in the startup message. A Boolean
value of <code class="literal">true</code> (or <code class="literal">on</code>,
<code class="literal">yes</code>, <code class="literal">1</code>) tells the backend to go into
physical replication walsender mode, wherein a small set of replication
commands, shown below, can be issued instead of SQL statements.
</p><p>
Passing <code class="literal">database</code> as the value for the
<code class="literal">replication</code> parameter instructs the backend to go into
logical replication walsender mode, connecting to the database specified in
the <code class="literal">dbname</code> parameter. In logical replication walsender
mode, the replication commands shown below as well as normal SQL commands can
be issued.
</p><p>
In either physical replication or logical replication walsender mode, only the
simple query protocol can be used.
</p><p>
For the purpose of testing replication commands, you can make a replication
connection via <span class="application">psql</span> or any other
<span class="application">libpq</span>-using tool with a connection string including
the <code class="literal">replication</code> option,
e.g.:
</p><pre class="programlisting">
psql "dbname=postgres replication=database" -c "IDENTIFY_SYSTEM;"
</pre><p>
However, it is often more useful to use
<a class="xref" href="app-pgreceivewal.html" title="pg_receivewal"><span class="refentrytitle"><span class="application">pg_receivewal</span></span></a> (for physical replication) or
<a class="xref" href="app-pgrecvlogical.html" title="pg_recvlogical"><span class="refentrytitle"><span class="application">pg_recvlogical</span></span></a> (for logical replication).
</p><p>
Replication commands are logged in the server log when
<a class="xref" href="runtime-config-logging.html#GUC-LOG-REPLICATION-COMMANDS">log_replication_commands</a> is enabled.
</p><p>
The commands accepted in replication mode are:
</p><div class="variablelist"><dl class="variablelist"><dt id="PROTOCOL-REPLICATION-IDENTIFY-SYSTEM"><span class="term"><code class="literal">IDENTIFY_SYSTEM</code>
<a id="id-1.10.6.9.7.1.1.1.2" class="indexterm"></a>
</span> <a href="#PROTOCOL-REPLICATION-IDENTIFY-SYSTEM" class="id_link">#</a></dt><dd><p>
Requests the server to identify itself. Server replies with a result
set of a single row, containing four fields:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">systemid</code> (<code class="type">text</code>)</span></dt><dd><p>
The unique system identifier identifying the cluster. This
can be used to check that the base backup used to initialize the
standby came from the same cluster.
</p></dd><dt><span class="term"><code class="literal">timeline</code> (<code class="type">int8</code>)</span></dt><dd><p>
Current timeline ID. Also useful to check that the standby is
consistent with the primary.
</p></dd><dt><span class="term"><code class="literal">xlogpos</code> (<code class="type">text</code>)</span></dt><dd><p>
Current WAL flush location. Useful to get a known location in the
write-ahead log where streaming can start.
</p></dd><dt><span class="term"><code class="literal">dbname</code> (<code class="type">text</code>)</span></dt><dd><p>
Database connected to or null.
</p></dd></dl></div></dd><dt id="PROTOCOL-REPLICATION-SHOW"><span class="term"><code class="literal">SHOW</code> <em class="replaceable"><code>name</code></em>
<a id="id-1.10.6.9.7.1.2.1.3" class="indexterm"></a>
</span> <a href="#PROTOCOL-REPLICATION-SHOW" class="id_link">#</a></dt><dd><p>
Requests the server to send the current setting of a run-time parameter.
This is similar to the SQL command <a class="xref" href="sql-show.html" title="SHOW"><span class="refentrytitle">SHOW</span></a>.
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><em class="replaceable"><code>name</code></em></span></dt><dd><p>
The name of a run-time parameter. Available parameters are documented
in <a class="xref" href="runtime-config.html" title="Chapter 20. Server Configuration">Chapter 20</a>.
</p></dd></dl></div></dd><dt id="PROTOCOL-REPLICATION-TIMELINE-HISTORY"><span class="term"><code class="literal">TIMELINE_HISTORY</code> <em class="replaceable"><code>tli</code></em>
<a id="id-1.10.6.9.7.1.3.1.3" class="indexterm"></a>
</span> <a href="#PROTOCOL-REPLICATION-TIMELINE-HISTORY" class="id_link">#</a></dt><dd><p>
Requests the server to send over the timeline history file for timeline
<em class="replaceable"><code>tli</code></em>. Server replies with a
result set of a single row, containing two fields. While the fields
are labeled as <code class="type">text</code>, they effectively return raw bytes,
with no encoding conversion:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">filename</code> (<code class="type">text</code>)</span></dt><dd><p>
File name of the timeline history file, e.g., <code class="filename">00000002.history</code>.
</p></dd><dt><span class="term"><code class="literal">content</code> (<code class="type">text</code>)</span></dt><dd><p>
Contents of the timeline history file.
</p></dd></dl></div></dd><dt id="PROTOCOL-REPLICATION-CREATE-REPLICATION-SLOT"><span class="term"><code class="literal">CREATE_REPLICATION_SLOT</code> <em class="replaceable"><code>slot_name</code></em> [ <code class="literal">TEMPORARY</code> ] { <code class="literal">PHYSICAL</code> | <code class="literal">LOGICAL</code> <em class="replaceable"><code>output_plugin</code></em> } [ ( <em class="replaceable"><code>option</code></em> [, ...] ) ]
<a id="id-1.10.6.9.7.1.4.1.8" class="indexterm"></a>
</span> <a href="#PROTOCOL-REPLICATION-CREATE-REPLICATION-SLOT" class="id_link">#</a></dt><dd><p>
Create a physical or logical replication
slot. See <a class="xref" href="warm-standby.html#STREAMING-REPLICATION-SLOTS" title="27.2.6. Replication Slots">Section 27.2.6</a> for more about
replication slots.
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><em class="replaceable"><code>slot_name</code></em></span></dt><dd><p>
The name of the slot to create. Must be a valid replication slot
name (see <a class="xref" href="warm-standby.html#STREAMING-REPLICATION-SLOTS-MANIPULATION" title="27.2.6.1. Querying and Manipulating Replication Slots">Section 27.2.6.1</a>).
</p></dd><dt><span class="term"><em class="replaceable"><code>output_plugin</code></em></span></dt><dd><p>
The name of the output plugin used for logical decoding
(see <a class="xref" href="logicaldecoding-output-plugin.html" title="49.6. Logical Decoding Output Plugins">Section 49.6</a>).
</p></dd><dt><span class="term"><code class="literal">TEMPORARY</code></span></dt><dd><p>
Specify that this replication slot is a temporary one. Temporary
slots are not saved to disk and are automatically dropped on error
or when the session has finished.
</p></dd></dl></div><p>The following options are supported:</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">TWO_PHASE [ <em class="replaceable"><code>boolean</code></em> ]</code></span></dt><dd><p>
If true, this logical replication slot supports decoding of two-phase
commit. With this option, commands related to two-phase commit such as
<code class="literal">PREPARE TRANSACTION</code>, <code class="literal">COMMIT PREPARED</code>
and <code class="literal">ROLLBACK PREPARED</code> are decoded and transmitted.
The transaction will be decoded and transmitted at
<code class="literal">PREPARE TRANSACTION</code> time.
The default is false.
</p></dd><dt><span class="term"><code class="literal">RESERVE_WAL [ <em class="replaceable"><code>boolean</code></em> ]</code></span></dt><dd><p>
If true, this physical replication slot reserves <acronym class="acronym">WAL</acronym>
immediately. Otherwise, <acronym class="acronym">WAL</acronym> is only reserved upon
connection from a streaming replication client.
The default is false.
</p></dd><dt><span class="term"><code class="literal">SNAPSHOT { 'export' | 'use' | 'nothing' }</code></span></dt><dd><p>
Decides what to do with the snapshot created during logical slot
initialization. <code class="literal">'export'</code>, which is the default,
will export the snapshot for use in other sessions. This option can't
be used inside a transaction. <code class="literal">'use'</code> will use the
snapshot for the current transaction executing the command. This
option must be used in a transaction, and
<code class="literal">CREATE_REPLICATION_SLOT</code> must be the first command
run in that transaction. Finally, <code class="literal">'nothing'</code> will
just use the snapshot for logical decoding as normal but won't do
anything else with it.
</p></dd></dl></div><p>
In response to this command, the server will send a one-row result set
containing the following fields:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">slot_name</code> (<code class="type">text</code>)</span></dt><dd><p>
The name of the newly-created replication slot.
</p></dd><dt><span class="term"><code class="literal">consistent_point</code> (<code class="type">text</code>)</span></dt><dd><p>
The WAL location at which the slot became consistent. This is the
earliest location from which streaming can start on this replication
slot.
</p></dd><dt><span class="term"><code class="literal">snapshot_name</code> (<code class="type">text</code>)</span></dt><dd><p>
The identifier of the snapshot exported by the command. The
snapshot is valid until a new command is executed on this connection
or the replication connection is closed. Null if the created slot
is physical.
</p></dd><dt><span class="term"><code class="literal">output_plugin</code> (<code class="type">text</code>)</span></dt><dd><p>
The name of the output plugin used by the newly-created replication
slot. Null if the created slot is physical.
</p></dd></dl></div><p>
</p></dd><dt id="PROTOCOL-REPLICATION-CREATE-REPLICATION-SLOT-LEGACY"><span class="term"><code class="literal">CREATE_REPLICATION_SLOT</code> <em class="replaceable"><code>slot_name</code></em> [ <code class="literal">TEMPORARY</code> ] { <code class="literal">PHYSICAL</code> [ <code class="literal">RESERVE_WAL</code> ] | <code class="literal">LOGICAL</code> <em class="replaceable"><code>output_plugin</code></em> [ <code class="literal">EXPORT_SNAPSHOT</code> | <code class="literal">NOEXPORT_SNAPSHOT</code> | <code class="literal">USE_SNAPSHOT</code> | <code class="literal">TWO_PHASE</code> ] }
</span> <a href="#PROTOCOL-REPLICATION-CREATE-REPLICATION-SLOT-LEGACY" class="id_link">#</a></dt><dd><p>
For compatibility with older releases, this alternative syntax for
the <code class="literal">CREATE_REPLICATION_SLOT</code> command is still supported.
</p></dd><dt id="PROTOCOL-REPLICATION-READ-REPLICATION-SLOT"><span class="term"><code class="literal">READ_REPLICATION_SLOT</code> <em class="replaceable"><code>slot_name</code></em>
<a id="id-1.10.6.9.7.1.6.1.3" class="indexterm"></a>
</span> <a href="#PROTOCOL-REPLICATION-READ-REPLICATION-SLOT" class="id_link">#</a></dt><dd><p>
Read some information associated with a replication slot. Returns a tuple
with <code class="literal">NULL</code> values if the replication slot does not
exist. This command is currently only supported for physical replication
slots.
</p><p>
In response to this command, the server will return a one-row result set,
containing the following fields:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">slot_type</code> (<code class="type">text</code>)</span></dt><dd><p>
The replication slot's type, either <code class="literal">physical</code> or
<code class="literal">NULL</code>.
</p></dd><dt><span class="term"><code class="literal">restart_lsn</code> (<code class="type">text</code>)</span></dt><dd><p>
The replication slot's <code class="literal">restart_lsn</code>.
</p></dd><dt><span class="term"><code class="literal">restart_tli</code> (<code class="type">int8</code>)</span></dt><dd><p>
The timeline ID associated with <code class="literal">restart_lsn</code>,
following the current timeline history.
</p></dd></dl></div><p>
</p></dd><dt id="PROTOCOL-REPLICATION-START-REPLICATION"><span class="term"><code class="literal">START_REPLICATION</code> [ <code class="literal">SLOT</code> <em class="replaceable"><code>slot_name</code></em> ] [ <code class="literal">PHYSICAL</code> ] <em class="replaceable"><code>XXX/XXX</code></em> [ <code class="literal">TIMELINE</code> <em class="replaceable"><code>tli</code></em> ]
<a id="id-1.10.6.9.7.1.7.1.8" class="indexterm"></a>
</span> <a href="#PROTOCOL-REPLICATION-START-REPLICATION" class="id_link">#</a></dt><dd><p>
Instructs server to start streaming WAL, starting at
WAL location <em class="replaceable"><code>XXX/XXX</code></em>.
If <code class="literal">TIMELINE</code> option is specified,
streaming starts on timeline <em class="replaceable"><code>tli</code></em>;
otherwise, the server's current timeline is selected. The server can
reply with an error, for example if the requested section of WAL has already
been recycled. On success, the server responds with a CopyBothResponse
message, and then starts to stream WAL to the frontend.
</p><p>
If a slot's name is provided
via <em class="replaceable"><code>slot_name</code></em>, it will be updated
as replication progresses so that the server knows which WAL segments,
and if <code class="varname">hot_standby_feedback</code> is on which transactions,
are still needed by the standby.
</p><p>
If the client requests a timeline that's not the latest but is part of
the history of the server, the server will stream all the WAL on that
timeline starting from the requested start point up to the point where
the server switched to another timeline. If the client requests
streaming at exactly the end of an old timeline, the server skips COPY
mode entirely.
</p><p>
After streaming all the WAL on a timeline that is not the latest one,
the server will end streaming by exiting the COPY mode. When the client
acknowledges this by also exiting COPY mode, the server sends a result
set with one row and two columns, indicating the next timeline in this
server's history. The first column is the next timeline's ID (type <code class="type">int8</code>), and the
second column is the WAL location where the switch happened (type <code class="type">text</code>). Usually,
the switch position is the end of the WAL that was streamed, but there
are corner cases where the server can send some WAL from the old
timeline that it has not itself replayed before promoting. Finally, the
server sends two CommandComplete messages (one that ends the CopyData
and the other ends the <code class="literal">START_REPLICATION</code> itself), and
is ready to accept a new command.
</p><p>
WAL data is sent as a series of CopyData messages. (This allows
other information to be intermixed; in particular the server can send
an ErrorResponse message if it encounters a failure after beginning
to stream.) The payload of each CopyData message from server to the
client contains a message of one of the following formats:
</p><div class="variablelist"><dl class="variablelist"><dt id="PROTOCOL-REPLICATION-XLOGDATA"><span class="term">XLogData (B)</span> <a href="#PROTOCOL-REPLICATION-XLOGDATA" class="id_link">#</a></dt><dd><div class="variablelist"><dl class="variablelist"><dt><span class="term">Byte1('w')</span></dt><dd><p>
Identifies the message as WAL data.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The starting point of the WAL data in this message.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The current end of WAL on the server.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The server's system clock at the time of transmission, as
microseconds since midnight on 2000-01-01.
</p></dd><dt><span class="term">Byte<em class="replaceable"><code>n</code></em></span></dt><dd><p>
A section of the WAL data stream.
</p><p>
A single WAL record is never split across two XLogData messages.
When a WAL record crosses a WAL page boundary, and is therefore
already split using continuation records, it can be split at the page
boundary. In other words, the first main WAL record and its
continuation records can be sent in different XLogData messages.
</p></dd></dl></div></dd><dt id="PROTOCOL-REPLICATION-PRIMARY-KEEPALIVE-MESSAGE"><span class="term">Primary keepalive message (B)</span> <a href="#PROTOCOL-REPLICATION-PRIMARY-KEEPALIVE-MESSAGE" class="id_link">#</a></dt><dd><div class="variablelist"><dl class="variablelist"><dt><span class="term">Byte1('k')</span></dt><dd><p>
Identifies the message as a sender keepalive.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The current end of WAL on the server.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The server's system clock at the time of transmission, as
microseconds since midnight on 2000-01-01.
</p></dd><dt><span class="term">Byte1</span></dt><dd><p>
1 means that the client should reply to this message as soon as
possible, to avoid a timeout disconnect. 0 otherwise.
</p></dd></dl></div></dd></dl></div><p>
The receiving process can send replies back to the sender at any time,
using one of the following message formats (also in the payload of a
CopyData message):
</p><div class="variablelist"><dl class="variablelist"><dt id="PROTOCOL-REPLICATION-STANDBY-STATUS-UPDATE"><span class="term">Standby status update (F)</span> <a href="#PROTOCOL-REPLICATION-STANDBY-STATUS-UPDATE" class="id_link">#</a></dt><dd><div class="variablelist"><dl class="variablelist"><dt><span class="term">Byte1('r')</span></dt><dd><p>
Identifies the message as a receiver status update.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The location of the last WAL byte + 1 received and written to disk
in the standby.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The location of the last WAL byte + 1 flushed to disk in
the standby.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The location of the last WAL byte + 1 applied in the standby.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The client's system clock at the time of transmission, as
microseconds since midnight on 2000-01-01.
</p></dd><dt><span class="term">Byte1</span></dt><dd><p>
If 1, the client requests the server to reply to this message
immediately. This can be used to ping the server, to test if
the connection is still healthy.
</p></dd></dl></div></dd><dt id="PROTOCOL-REPLICATION-HOT-STANDBY-FEEDBACK-MESSAGE"><span class="term">Hot standby feedback message (F)</span> <a href="#PROTOCOL-REPLICATION-HOT-STANDBY-FEEDBACK-MESSAGE" class="id_link">#</a></dt><dd><div class="variablelist"><dl class="variablelist"><dt><span class="term">Byte1('h')</span></dt><dd><p>
Identifies the message as a hot standby feedback message.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The client's system clock at the time of transmission, as
microseconds since midnight on 2000-01-01.
</p></dd><dt><span class="term">Int32</span></dt><dd><p>
The standby's current global xmin, excluding the catalog_xmin from any
replication slots. If both this value and the following
catalog_xmin are 0 this is treated as a notification that hot standby
feedback will no longer be sent on this connection. Later non-zero
messages may reinitiate the feedback mechanism.
</p></dd><dt><span class="term">Int32</span></dt><dd><p>
The epoch of the global xmin xid on the standby.
</p></dd><dt><span class="term">Int32</span></dt><dd><p>
The lowest catalog_xmin of any replication slots on the standby. Set to 0
if no catalog_xmin exists on the standby or if hot standby feedback is being
disabled.
</p></dd><dt><span class="term">Int32</span></dt><dd><p>
The epoch of the catalog_xmin xid on the standby.
</p></dd></dl></div></dd></dl></div></dd><dt id="PROTOCOL-REPLICATION-START-REPLICATION-SLOT-LOGICAL"><span class="term"><code class="literal">START_REPLICATION</code> <code class="literal">SLOT</code> <em class="replaceable"><code>slot_name</code></em> <code class="literal">LOGICAL</code> <em class="replaceable"><code>XXX/XXX</code></em> [ ( <em class="replaceable"><code>option_name</code></em> [ <em class="replaceable"><code>option_value</code></em> ] [, ...] ) ]</span> <a href="#PROTOCOL-REPLICATION-START-REPLICATION-SLOT-LOGICAL" class="id_link">#</a></dt><dd><p>
Instructs server to start streaming WAL for logical replication,
starting at either WAL location <em class="replaceable"><code>XXX/XXX</code></em> or the slot's
<code class="literal">confirmed_flush_lsn</code> (see <a class="xref" href="view-pg-replication-slots.html" title="54.19. pg_replication_slots">Section 54.19</a>), whichever is greater. This
behavior makes it easier for clients to avoid updating their local LSN
status when there is no data to process. However, starting at a
different LSN than requested might not catch certain kinds of client
errors; so the client may wish to check that
<code class="literal">confirmed_flush_lsn</code> matches its expectations before
issuing <code class="literal">START_REPLICATION</code>.
</p><p>
The server can reply with an error, for example if the
slot does not exist. On success, the server responds with a CopyBothResponse
message, and then starts to stream WAL to the frontend.
</p><p>
The messages inside the CopyBothResponse messages are of the same format
documented for <code class="literal">START_REPLICATION ... PHYSICAL</code>, including
two CommandComplete messages.
</p><p>
The output plugin associated with the selected slot is used
to process the output for streaming.
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">SLOT</code> <em class="replaceable"><code>slot_name</code></em></span></dt><dd><p>
The name of the slot to stream changes from. This parameter is required,
and must correspond to an existing logical replication slot created
with <code class="literal">CREATE_REPLICATION_SLOT</code> in
<code class="literal">LOGICAL</code> mode.
</p></dd><dt><span class="term"><em class="replaceable"><code>XXX/XXX</code></em></span></dt><dd><p>
The WAL location to begin streaming at.
</p></dd><dt><span class="term"><em class="replaceable"><code>option_name</code></em></span></dt><dd><p>
The name of an option passed to the slot's logical decoding output
plugin. See <a class="xref" href="protocol-logical-replication.html" title="55.5. Logical Streaming Replication Protocol">Section 55.5</a> for
options that are accepted by the standard (<code class="literal">pgoutput</code>)
plugin.
</p></dd><dt><span class="term"><em class="replaceable"><code>option_value</code></em></span></dt><dd><p>
Optional value, in the form of a string constant, associated with the
specified option.
</p></dd></dl></div></dd><dt id="PROTOCOL-REPLICATION-DROP-REPLICATION-SLOT"><span class="term">
<code class="literal">DROP_REPLICATION_SLOT</code> <em class="replaceable"><code>slot_name</code></em> [<span class="optional"> <code class="literal">WAIT</code> </span>]
<a id="id-1.10.6.9.7.1.9.1.4" class="indexterm"></a>
</span> <a href="#PROTOCOL-REPLICATION-DROP-REPLICATION-SLOT" class="id_link">#</a></dt><dd><p>
Drops a replication slot, freeing any reserved server-side resources.
If the slot is a logical slot that was created in a database other than
the database the walsender is connected to, this command fails.
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><em class="replaceable"><code>slot_name</code></em></span></dt><dd><p>
The name of the slot to drop.
</p></dd><dt><span class="term"><code class="literal">WAIT</code></span></dt><dd><p>
This option causes the command to wait if the slot is active until
it becomes inactive, instead of the default behavior of raising an
error.
</p></dd></dl></div></dd><dt id="PROTOCOL-REPLICATION-BASE-BACKUP"><span class="term"><code class="literal">BASE_BACKUP</code> [ ( <em class="replaceable"><code>option</code></em> [, ...] ) ]
<a id="id-1.10.6.9.7.1.10.1.3" class="indexterm"></a>
</span> <a href="#PROTOCOL-REPLICATION-BASE-BACKUP" class="id_link">#</a></dt><dd><p>
Instructs the server to start streaming a base backup.
The system will automatically be put in backup mode before the backup
is started, and taken out of it when the backup is complete. The
following options are accepted:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">LABEL</code> <em class="replaceable"><code>'label'</code></em></span></dt><dd><p>
Sets the label of the backup. If none is specified, a backup label
of <code class="literal">base backup</code> will be used. The quoting rules
for the label are the same as a standard SQL string with
<a class="xref" href="runtime-config-compatible.html#GUC-STANDARD-CONFORMING-STRINGS">standard_conforming_strings</a> turned on.
</p></dd><dt><span class="term"><code class="literal">TARGET</code> <em class="replaceable"><code>'target'</code></em></span></dt><dd><p>
Tells the server where to send the backup. If the target is
<code class="literal">client</code>, which is the default, the backup data is
sent to the client. If it is <code class="literal">server</code>, the backup
data is written to the server at the pathname specified by the
<code class="literal">TARGET_DETAIL</code> option. If it is
<code class="literal">blackhole</code>, the backup data is not sent
anywhere; it is simply discarded.
</p><p>
The <code class="literal">server</code> target requires superuser privilege or
being granted the <code class="literal">pg_write_server_files</code> role.
</p></dd><dt><span class="term"><code class="literal">TARGET_DETAIL</code> <em class="replaceable"><code>'detail'</code></em></span></dt><dd><p>
Provides additional information about the backup target.
</p><p>
Currently, this option can only be used when the backup target is
<code class="literal">server</code>. It specifies the server directory
to which the backup should be written.
</p></dd><dt><span class="term"><code class="literal">PROGRESS [ <em class="replaceable"><code>boolean</code></em> ]</code></span></dt><dd><p>
If set to true, request information required to generate a progress
report. This will send back an approximate size in the header of each
tablespace, which can be used to calculate how far along the stream
is done. This is calculated by enumerating all the file sizes once
before the transfer is even started, and might as such have a
negative impact on the performance. In particular, it might take
longer before the first data
is streamed. Since the database files can change during the backup,
the size is only approximate and might both grow and shrink between
the time of approximation and the sending of the actual files.
The default is false.
</p></dd><dt><span class="term"><code class="literal">CHECKPOINT { 'fast' | 'spread' }</code></span></dt><dd><p>
Sets the type of checkpoint to be performed at the beginning of the
base backup. The default is <code class="literal">spread</code>.
</p></dd><dt><span class="term"><code class="literal">WAL [ <em class="replaceable"><code>boolean</code></em> ]</code></span></dt><dd><p>
If set to true, include the necessary WAL segments in the backup.
This will include all the files between start and stop backup in the
<code class="filename">pg_wal</code> directory of the base directory tar
file. The default is false.
</p></dd><dt><span class="term"><code class="literal">WAIT [ <em class="replaceable"><code>boolean</code></em> ]</code></span></dt><dd><p>
If set to true, the backup will wait until the last required WAL
segment has been archived, or emit a warning if WAL archiving is
not enabled. If false, the backup will neither wait nor warn,
leaving the client responsible for ensuring the required log is
available. The default is true.
</p></dd><dt><span class="term"><code class="literal">COMPRESSION</code> <em class="replaceable"><code>'method'</code></em></span></dt><dd><p>
Instructs the server to compress the backup using the specified
method. Currently, the supported methods are <code class="literal">gzip</code>,
<code class="literal">lz4</code>, and <code class="literal">zstd</code>.
</p></dd><dt><span class="term"><code class="literal">COMPRESSION_DETAIL</code> <em class="replaceable"><code>detail</code></em></span></dt><dd><p>
Specifies details for the chosen compression method. This should only
be used in conjunction with the <code class="literal">COMPRESSION</code>
option. If the value is an integer, it specifies the compression
level. Otherwise, it should be a comma-separated list of items,
each of the form <em class="replaceable"><code>keyword</code></em> or
<em class="replaceable"><code>keyword=value</code></em>. Currently, the supported
keywords are <code class="literal">level</code>, <code class="literal">long</code> and
<code class="literal">workers</code>.
</p><p>
The <code class="literal">level</code> keyword sets the compression level.
For <code class="literal">gzip</code> the compression level should be an
integer between <code class="literal">1</code> and <code class="literal">9</code>
(default <code class="literal">Z_DEFAULT_COMPRESSION</code> or
<code class="literal">-1</code>), for <code class="literal">lz4</code> an integer
between 1 and 12 (default <code class="literal">0</code> for fast compression
mode), and for <code class="literal">zstd</code> an integer between
<code class="literal">ZSTD_minCLevel()</code> (usually <code class="literal">-131072</code>)
and <code class="literal">ZSTD_maxCLevel()</code> (usually <code class="literal">22</code>),
(default <code class="literal">ZSTD_CLEVEL_DEFAULT</code> or
<code class="literal">3</code>).
</p><p>
The <code class="literal">long</code> keyword enables long-distance matching
mode, for improved compression ratio, at the expense of higher memory
use. Long-distance mode is supported only for
<code class="literal">zstd</code>.
</p><p>
The <code class="literal">workers</code> keyword sets the number of threads
that should be used for parallel compression. Parallel compression
is supported only for <code class="literal">zstd</code>.
</p></dd><dt><span class="term"><code class="literal">MAX_RATE</code> <em class="replaceable"><code>rate</code></em></span></dt><dd><p>
Limit (throttle) the maximum amount of data transferred from server
to client per unit of time. The expected unit is kilobytes per second.
If this option is specified, the value must either be equal to zero
or it must fall within the range from 32 kB through 1 GB (inclusive).
If zero is passed or the option is not specified, no restriction is
imposed on the transfer.
</p></dd><dt><span class="term"><code class="literal">TABLESPACE_MAP [ <em class="replaceable"><code>boolean</code></em> ]</code></span></dt><dd><p>
If true, include information about symbolic links present in the
directory <code class="filename">pg_tblspc</code> in a file named
<code class="filename">tablespace_map</code>. The tablespace map file includes
each symbolic link name as it exists in the directory
<code class="filename">pg_tblspc/</code> and the full path of that symbolic link.
The default is false.
</p></dd><dt><span class="term"><code class="literal">VERIFY_CHECKSUMS [ <em class="replaceable"><code>boolean</code></em> ]</code></span></dt><dd><p>
If true, checksums are verified during a base backup if they are
enabled. If false, this is skipped. The default is true.
</p></dd><dt><span class="term"><code class="literal">MANIFEST</code> <em class="replaceable"><code>manifest_option</code></em></span></dt><dd><p>
When this option is specified with a value of <code class="literal">yes</code>
or <code class="literal">force-encode</code>, a backup manifest is created
and sent along with the backup. The manifest is a list of every
file present in the backup with the exception of any WAL files that
may be included. It also stores the size, last modification time, and
optionally a checksum for each file.
A value of <code class="literal">force-encode</code> forces all filenames
to be hex-encoded; otherwise, this type of encoding is performed only
for files whose names are non-UTF8 octet sequences.
<code class="literal">force-encode</code> is intended primarily for testing
purposes, to be sure that clients which read the backup manifest
can handle this case. For compatibility with previous releases,
the default is <code class="literal">MANIFEST 'no'</code>.
</p></dd><dt><span class="term"><code class="literal">MANIFEST_CHECKSUMS</code> <em class="replaceable"><code>checksum_algorithm</code></em></span></dt><dd><p>
Specifies the checksum algorithm that should be applied to each file included
in the backup manifest. Currently, the available
algorithms are <code class="literal">NONE</code>, <code class="literal">CRC32C</code>,
<code class="literal">SHA224</code>, <code class="literal">SHA256</code>,
<code class="literal">SHA384</code>, and <code class="literal">SHA512</code>.
The default is <code class="literal">CRC32C</code>.
</p></dd></dl></div><p>
</p><p>
When the backup is started, the server will first send two
ordinary result sets, followed by one or more CopyOutResponse
results.
</p><p>
The first ordinary result set contains the starting position of the
backup, in a single row with two columns. The first column contains
the start position given in XLogRecPtr format, and the second column
contains the corresponding timeline ID.
</p><p>
The second ordinary result set has one row for each tablespace.
The fields in this row are:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">spcoid</code> (<code class="type">oid</code>)</span></dt><dd><p>
The OID of the tablespace, or null if it's the base
directory.
</p></dd><dt><span class="term"><code class="literal">spclocation</code> (<code class="type">text</code>)</span></dt><dd><p>
The full path of the tablespace directory, or null
if it's the base directory.
</p></dd><dt><span class="term"><code class="literal">size</code> (<code class="type">int8</code>)</span></dt><dd><p>
The approximate size of the tablespace, in kilobytes (1024 bytes),
if progress report has been requested; otherwise it's null.
</p></dd></dl></div><p>
</p><p>
After the second regular result set, a CopyOutResponse will be sent.
The payload of each CopyData message will contain a message in one of
the following formats:
</p><div class="variablelist"><dl class="variablelist"><dt><span class="term">new archive (B)</span></dt><dd><div class="variablelist"><dl class="variablelist"><dt><span class="term">Byte1('n')</span></dt><dd><p>
Identifies the message as indicating the start of a new archive.
There will be one archive for the main data directory and one
for each additional tablespace; each will use tar format
(following the <span class="quote">“<span class="quote">ustar interchange format</span>”</span> specified
in the POSIX 1003.1-2008 standard).
</p></dd><dt><span class="term">String</span></dt><dd><p>
The file name for this archive.
</p></dd><dt><span class="term">String</span></dt><dd><p>
For the main data directory, an empty string. For other
tablespaces, the full path to the directory from which this
archive was created.
</p></dd></dl></div></dd><dt><span class="term">manifest (B)</span></dt><dd><div class="variablelist"><dl class="variablelist"><dt><span class="term">Byte1('m')</span></dt><dd><p>
Identifies the message as indicating the start of the backup
manifest.
</p></dd></dl></div></dd><dt><span class="term">archive or manifest data (B)</span></dt><dd><div class="variablelist"><dl class="variablelist"><dt><span class="term">Byte1('d')</span></dt><dd><p>
Identifies the message as containing archive or manifest data.
</p></dd><dt><span class="term">Byte<em class="replaceable"><code>n</code></em></span></dt><dd><p>
Data bytes.
</p></dd></dl></div></dd><dt><span class="term">progress report (B)</span></dt><dd><div class="variablelist"><dl class="variablelist"><dt><span class="term">Byte1('p')</span></dt><dd><p>
Identifies the message as a progress report.
</p></dd><dt><span class="term">Int64</span></dt><dd><p>
The number of bytes from the current tablespace for which
processing has been completed.
</p></dd></dl></div></dd></dl></div><p>
After the CopyOutResponse, or all such responses, have been sent, a
final ordinary result set will be sent, containing the WAL end position
of the backup, in the same format as the start position.
</p><p>
The tar archive for the data directory and each tablespace will contain
all files in the directories, regardless of whether they are
<span class="productname">PostgreSQL</span> files or other files added to the same
directory. The only excluded files are:
</p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: bullet; "><li class="listitem" style="list-style-type: disc"><p>
<code class="filename">postmaster.pid</code>
</p></li><li class="listitem" style="list-style-type: disc"><p>
<code class="filename">postmaster.opts</code>
</p></li><li class="listitem" style="list-style-type: disc"><p>
<code class="filename">pg_internal.init</code> (found in multiple directories)
</p></li><li class="listitem" style="list-style-type: disc"><p>
Various temporary files and directories created during the operation
of the PostgreSQL server, such as any file or directory beginning
with <code class="filename">pgsql_tmp</code> and temporary relations.
</p></li><li class="listitem" style="list-style-type: disc"><p>
Unlogged relations, except for the init fork which is required to
recreate the (empty) unlogged relation on recovery.
</p></li><li class="listitem" style="list-style-type: disc"><p>
<code class="filename">pg_wal</code>, including subdirectories. If the backup is run
with WAL files included, a synthesized version of <code class="filename">pg_wal</code> will be
included, but it will only contain the files necessary for the
backup to work, not the rest of the contents.
</p></li><li class="listitem" style="list-style-type: disc"><p>
<code class="filename">pg_dynshmem</code>, <code class="filename">pg_notify</code>,
<code class="filename">pg_replslot</code>, <code class="filename">pg_serial</code>,
<code class="filename">pg_snapshots</code>, <code class="filename">pg_stat_tmp</code>, and
<code class="filename">pg_subtrans</code> are copied as empty directories (even if
they are symbolic links).
</p></li><li class="listitem" style="list-style-type: disc"><p>
Files other than regular files and directories, such as symbolic
links (other than for the directories listed above) and special
device files, are skipped. (Symbolic links
in <code class="filename">pg_tblspc</code> are maintained.)
</p></li></ul></div><p>
Owner, group, and file mode are set if the underlying file system on
the server supports it.
</p></dd></dl></div><p>
</p></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="sasl-authentication.html" title="55.3. SASL Authentication">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="protocol.html" title="Chapter 55. Frontend/Backend Protocol">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="protocol-logical-replication.html" title="55.5. Logical Streaming Replication Protocol">Next</a></td></tr><tr><td width="40%" align="left" valign="top">55.3. SASL Authentication </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 16.2 Documentation">Home</a></td><td width="40%" align="right" valign="top"> 55.5. Logical Streaming Replication Protocol</td></tr></table></div></body></html>
|
