1PG_UPGRADE(1) PostgreSQL 16.1 Documentation PG_UPGRADE(1)
2
6 pg_upgrade - upgrade a PostgreSQL server instance
7
9 pg_upgrade -b oldbindir [-B newbindir] -d oldconfigdir -D newconfigdir
10 [option...]
11
13 pg_upgrade (formerly called pg_migrator) allows data stored in
14 PostgreSQL data files to be upgraded to a later PostgreSQL major
15 version without the data dump/restore typically required for major
16 version upgrades, e.g., from 12.14 to 13.10 or from 14.9 to 15.5. It is
17 not required for minor version upgrades, e.g., from 12.7 to 12.8 or
18 from 14.1 to 14.5.
19 Major PostgreSQL releases regularly add new features that often change
21 the layout of the system tables, but the internal data storage format
22 rarely changes. pg_upgrade uses this fact to perform rapid upgrades by
23 creating new system tables and simply reusing the old user data files.
24 If a future major release ever changes the data storage format in a way
25 that makes the old data format unreadable, pg_upgrade will not be
26 usable for such upgrades. (The community will attempt to avoid such
27 situations.)
28 pg_upgrade does its best to make sure the old and new clusters are
30 binary-compatible, e.g., by checking for compatible compile-time
31 settings, including 32/64-bit binaries. It is important that any
32 external modules are also binary compatible, though this cannot be
33 checked by pg_upgrade.
34 pg_upgrade supports upgrades from 9.2.X and later to the current major
36 release of PostgreSQL, including snapshot and beta releases.
37
39 pg_upgrade accepts the following command-line arguments:
40 -b bindir
42 --old-bindir=bindir
43 the old PostgreSQL executable directory; environment variable
44 PGBINOLD
45 -B bindir
47 --new-bindir=bindir
48 the new PostgreSQL executable directory; default is the directory
49 where pg_upgrade resides; environment variable PGBINNEW
50 -c
52 --check
53 check clusters only, don't change any data
54 -d configdir
56 --old-datadir=configdir
57 the old database cluster configuration directory; environment
58 variable PGDATAOLD
59 -D configdir
61 --new-datadir=configdir
62 the new database cluster configuration directory; environment
63 variable PGDATANEW
64 -j njobs
66 --jobs=njobs
67 number of simultaneous processes or threads to use
68 -k
70 --link
71 use hard links instead of copying files to the new cluster
72 -N
74 --no-sync
75 By default, pg_upgrade will wait for all files of the upgraded
76 cluster to be written safely to disk. This option causes pg_upgrade
77 to return without waiting, which is faster, but means that a
78 subsequent operating system crash can leave the data directory
79 corrupt. Generally, this option is useful for testing but should
80 not be used on a production installation.
81 -o options
83 --old-options options
84 options to be passed directly to the old postgres command; multiple
85 option invocations are appended
86 -O options
88 --new-options options
89 options to be passed directly to the new postgres command; multiple
90 option invocations are appended
91 -p port
93 --old-port=port
94 the old cluster port number; environment variable PGPORTOLD
95 -P port
97 --new-port=port
98 the new cluster port number; environment variable PGPORTNEW
99 -r
101 --retain
102 retain SQL and log files even after successful completion
103 -s dir
105 --socketdir=dir
106 directory to use for postmaster sockets during upgrade; default is
107 current working directory; environment variable PGSOCKETDIR
108 -U username
110 --username=username
111 cluster's install user name; environment variable PGUSER
112 -v
114 --verbose
115 enable verbose internal logging
116 -V
118 --version
119 display version information, then exit
120 --clone
122 Use efficient file cloning (also known as “reflinks” on some
123 systems) instead of copying files to the new cluster. This can
124 result in near-instantaneous copying of the data files, giving the
125 speed advantages of -k/--link while leaving the old cluster
126 untouched.
127 File cloning is only supported on some operating systems and file
129 systems. If it is selected but not supported, the pg_upgrade run
130 will error. At present, it is supported on Linux (kernel 4.5 or
131 later) with Btrfs and XFS (on file systems created with reflink
132 support), and on macOS with APFS.
133 --copy
135 Copy files to the new cluster. This is the default. (See also
136 --link and --clone.)
137 -?
139 --help
140 show help, then exit
141
143 These are the steps to perform an upgrade with pg_upgrade:
144 1. Optionally move the old cluster: If you are using a
146 version-specific installation directory, e.g., /opt/PostgreSQL/16,
147 you do not need to move the old cluster. The graphical installers
148 all use version-specific installation directories.
149 If your installation directory is not version-specific, e.g.,
151 /usr/local/pgsql, it is necessary to move the current PostgreSQL
152 install directory so it does not interfere with the new PostgreSQL
153 installation. Once the current PostgreSQL server is shut down, it
154 is safe to rename the PostgreSQL installation directory; assuming
155 the old directory is /usr/local/pgsql, you can do:
156 mv /usr/local/pgsql /usr/local/pgsql.old
158 to rename the directory.
160 2. For source installs, build the new version: Build the new
162 PostgreSQL source with configure flags that are compatible with the
163 old cluster. pg_upgrade will check pg_controldata to make sure all
164 settings are compatible before starting the upgrade.
165 3. Install the new PostgreSQL binaries: Install the new server's
167 binaries and support files. pg_upgrade is included in a default
168 installation.
169 For source installs, if you wish to install the new server in a
171 custom location, use the prefix variable:
172 make prefix=/usr/local/pgsql.new install
174 4. Initialize the new PostgreSQL cluster: Initialize the new cluster
176 using initdb. Again, use compatible initdb flags that match the old
177 cluster. Many prebuilt installers do this step automatically. There
178 is no need to start the new cluster.
179 5. Install extension shared object files: Many extensions and custom
181 modules, whether from contrib or another source, use shared object
182 files (or DLLs), e.g., pgcrypto.so. If the old cluster used these,
183 shared object files matching the new server binary must be
184 installed in the new cluster, usually via operating system
185 commands. Do not load the schema definitions, e.g., CREATE
186 EXTENSION pgcrypto, because these will be duplicated from the old
187 cluster. If extension updates are available, pg_upgrade will report
188 this and create a script that can be run later to update them.
189 6. Copy custom full-text search files: Copy any custom full text
191 search files (dictionary, synonym, thesaurus, stop words) from the
192 old to the new cluster.
193 7. Adjust authentication: pg_upgrade will connect to the old and new
195 servers several times, so you might want to set authentication to
196 peer in pg_hba.conf or use a ~/.pgpass file (see Section 34.16).
197 8. Stop both servers: Make sure both database servers are stopped
199 using, on Unix, e.g.:
200 pg_ctl -D /opt/PostgreSQL/12 stop
202 pg_ctl -D /opt/PostgreSQL/16 stop
203 or on Windows, using the proper service names:
205 NET STOP postgresql-12
207 NET STOP postgresql-16
208 Streaming replication and log-shipping standby servers must be
210 running during this shutdown so they receive all changes.
211 9. Prepare for standby server upgrades: If you are upgrading standby
213 servers using methods outlined in section Step 11, verify that the
214 old standby servers are caught up by running pg_controldata against
215 the old primary and standby clusters. Verify that the “Latest
216 checkpoint location” values match in all clusters. Also, make sure
217 wal_level is not set to minimal in the postgresql.conf file on the
218 new primary cluster.
219 10. Run pg_upgrade: Always run the pg_upgrade binary of the new
221 server, not the old one. pg_upgrade requires the specification of
222 the old and new cluster's data and executable (bin) directories.
223 You can also specify user and port values, and whether you want the
224 data files linked or cloned instead of the default copy behavior.
225 If you use link mode, the upgrade will be much faster (no file
227 copying) and use less disk space, but you will not be able to
228 access your old cluster once you start the new cluster after the
229 upgrade. Link mode also requires that the old and new cluster data
230 directories be in the same file system. (Tablespaces and pg_wal can
231 be on different file systems.) Clone mode provides the same speed
232 and disk space advantages but does not cause the old cluster to be
233 unusable once the new cluster is started. Clone mode also requires
234 that the old and new data directories be in the same file system.
235 This mode is only available on certain operating systems and file
236 systems.
237 The --jobs option allows multiple CPU cores to be used for
239 copying/linking of files and to dump and restore database schemas
240 in parallel; a good place to start is the maximum of the number of
241 CPU cores and tablespaces. This option can dramatically reduce the
242 time to upgrade a multi-database server running on a multiprocessor
243 machine.
244 For Windows users, you must be logged into an administrative
246 account, and then start a shell as the postgres user and set the
247 proper path:
248 RUNAS /USER:postgres "CMD.EXE"
250 SET PATH=%PATH%;C:\Program Files\PostgreSQL\16\bin;
251 and then run pg_upgrade with quoted directories, e.g.:
253 pg_upgrade.exe
255 --old-datadir "C:/Program Files/PostgreSQL/12/data"
256 --new-datadir "C:/Program Files/PostgreSQL/16/data"
257 --old-bindir "C:/Program Files/PostgreSQL/12/bin"
258 --new-bindir "C:/Program Files/PostgreSQL/16/bin"
259 Once started, pg_upgrade will verify the two clusters are
261 compatible and then do the upgrade. You can use pg_upgrade --check
262 to perform only the checks, even if the old server is still
263 running. pg_upgrade --check will also outline any manual
264 adjustments you will need to make after the upgrade. If you are
265 going to be using link or clone mode, you should use the option
266 --link or --clone with --check to enable mode-specific checks.
267 pg_upgrade requires write permission in the current directory.
268 Obviously, no one should be accessing the clusters during the
270 upgrade. pg_upgrade defaults to running servers on port 50432 to
271 avoid unintended client connections. You can use the same port
272 number for both clusters when doing an upgrade because the old and
273 new clusters will not be running at the same time. However, when
274 checking an old running server, the old and new port numbers must
275 be different.
276 If an error occurs while restoring the database schema, pg_upgrade
278 will exit and you will have to revert to the old cluster as
279 outlined in Step 17 below. To try pg_upgrade again, you will need
280 to modify the old cluster so the pg_upgrade schema restore
281 succeeds. If the problem is a contrib module, you might need to
282 uninstall the contrib module from the old cluster and install it in
283 the new cluster after the upgrade, assuming the module is not being
284 used to store user data.
285 11. Upgrade streaming replication and log-shipping standby servers: If
287 you used link mode and have Streaming Replication (see
288 Section 27.2.5) or Log-Shipping (see Section 27.2) standby servers,
289 you can follow these steps to quickly upgrade them. You will not be
290 running pg_upgrade on the standby servers, but rather rsync on the
291 primary. Do not start any servers yet.
292 If you did not use link mode, do not have or do not want to use
294 rsync, or want an easier solution, skip the instructions in this
295 section and simply recreate the standby servers once pg_upgrade
296 completes and the new primary is running.
297 1. Install the new PostgreSQL binaries on standby servers: Make
299 sure the new binaries and support files are installed on all
300 standby servers.
301 2. Make sure the new standby data directories do not exist: Make
303 sure the new standby data directories do not exist or are
304 empty. If initdb was run, delete the standby servers' new data
305 directories.
306 3. Install extension shared object files: Install the same
308 extension shared object files on the new standbys that you
309 installed in the new primary cluster.
310 4. Stop standby servers: If the standby servers are still running,
312 stop them now using the above instructions.
313 5. Save configuration files: Save any configuration files from the
315 old standbys' configuration directories you need to keep, e.g.,
316 postgresql.conf (and any files included by it),
317 postgresql.auto.conf, pg_hba.conf, because these will be
318 overwritten or removed in the next step.
319 6. Run rsync: When using link mode, standby servers can be quickly
321 upgraded using rsync. To accomplish this, from a directory on
322 the primary server that is above the old and new database
323 cluster directories, run this on the primary for each standby
324 server:
325 rsync --archive --delete --hard-links --size-only --no-inc-recursive old_cluster new_cluster remote_dir
327 where old_cluster and new_cluster are relative to the current
329 directory on the primary, and remote_dir is above the old and
330 new cluster directories on the standby. The directory structure
331 under the specified directories on the primary and standbys
332 must match. Consult the rsync manual page for details on
333 specifying the remote directory, e.g.,
334 rsync --archive --delete --hard-links --size-only --no-inc-recursive /opt/PostgreSQL/12 \
336 /opt/PostgreSQL/16 standby.example.com:/opt/PostgreSQL
337 You can verify what the command will do using rsync's --dry-run
339 option. While rsync must be run on the primary for at least one
340 standby, it is possible to run rsync on an upgraded standby to
341 upgrade other standbys, as long as the upgraded standby has not
342 been started.
343 What this does is to record the links created by pg_upgrade's
345 link mode that connect files in the old and new clusters on the
346 primary server. It then finds matching files in the standby's
347 old cluster and creates links for them in the standby's new
348 cluster. Files that were not linked on the primary are copied
349 from the primary to the standby. (They are usually small.) This
350 provides rapid standby upgrades. Unfortunately, rsync
351 needlessly copies files associated with temporary and unlogged
352 tables because these files don't normally exist on standby
353 servers.
354 If you have tablespaces, you will need to run a similar rsync
356 command for each tablespace directory, e.g.:
357 rsync --archive --delete --hard-links --size-only --no-inc-recursive /vol1/pg_tblsp/PG_12_201909212 \
359 /vol1/pg_tblsp/PG_16_202307071 standby.example.com:/vol1/pg_tblsp
360 If you have relocated pg_wal outside the data directories,
362 rsync must be run on those directories too.
363 7. Configure streaming replication and log-shipping standby
365 servers: Configure the servers for log shipping. (You do not
366 need to run pg_backup_start() and pg_backup_stop() or take a
367 file system backup as the standbys are still synchronized with
368 the primary.) Replication slots are not copied and must be
369 recreated.
370 12. Restore pg_hba.conf: If you modified pg_hba.conf, restore its
372 original settings. It might also be necessary to adjust other
373 configuration files in the new cluster to match the old cluster,
374 e.g., postgresql.conf (and any files included by it),
375 postgresql.auto.conf.
376 13. Start the new server: The new server can now be safely started,
378 and then any rsync'ed standby servers.
379 14. Post-upgrade processing: If any post-upgrade processing is
381 required, pg_upgrade will issue warnings as it completes. It will
382 also generate script files that must be run by the administrator.
383 The script files will connect to each database that needs
384 post-upgrade processing. Each script should be run using:
385 psql --username=postgres --file=script.sql postgres
387 The scripts can be run in any order and can be deleted once they
389 have been run.
390 Caution
392 In general it is unsafe to access tables referenced in rebuild
393 scripts until the rebuild scripts have run to completion; doing
394 so could yield incorrect results or poor performance. Tables
395 not referenced in rebuild scripts can be accessed immediately.
396 15. Statistics: Because optimizer statistics are not transferred by
398 pg_upgrade, you will be instructed to run a command to regenerate
399 that information at the end of the upgrade. You might need to set
400 connection parameters to match your new cluster.
401 16. Delete old cluster: Once you are satisfied with the upgrade, you
403 can delete the old cluster's data directories by running the script
404 mentioned when pg_upgrade completes. (Automatic deletion is not
405 possible if you have user-defined tablespaces inside the old data
406 directory.) You can also delete the old installation directories
407 (e.g., bin, share).
408 17. Reverting to old cluster: If, after running pg_upgrade, you wish
410 to revert to the old cluster, there are several options:
411 • If the --check option was used, the old cluster was unmodified;
413 it can be restarted.
414 • If the --link option was not used, the old cluster was
416 unmodified; it can be restarted.
417 • If the --link option was used, the data files might be shared
419 between the old and new cluster:
420 • If pg_upgrade aborted before linking started, the old
422 cluster was unmodified; it can be restarted.
423 • If you did not start the new cluster, the old cluster was
425 unmodified except that, when linking started, a .old suffix
426 was appended to $PGDATA/global/pg_control. To reuse the old
427 cluster, remove the .old suffix from
428 $PGDATA/global/pg_control; you can then restart the old
429 cluster.
430 • If you did start the new cluster, it has written to shared
432 files and it is unsafe to use the old cluster. The old
433 cluster will need to be restored from backup in this case.
434
436 pg_upgrade creates various working files, such as schema dumps, stored
437 within pg_upgrade_output.d in the directory of the new cluster. Each
438 run creates a new subdirectory named with a timestamp formatted as per
439 ISO 8601 (%Y%m%dT%H%M%S), where all its generated files are stored.
440 pg_upgrade_output.d and its contained files will be removed
441 automatically if pg_upgrade completes successfully; but in the event of
442 trouble, the files there may provide useful debugging information.
443 pg_upgrade launches short-lived postmasters in the old and new data
445 directories. Temporary Unix socket files for communication with these
446 postmasters are, by default, made in the current working directory. In
447 some situations the path name for the current directory might be too
448 long to be a valid socket name. In that case you can use the -s option
449 to put the socket files in some directory with a shorter path name. For
450 security, be sure that that directory is not readable or writable by
451 any other users. (This is not supported on Windows.)
452 All failure, rebuild, and reindex cases will be reported by pg_upgrade
454 if they affect your installation; post-upgrade scripts to rebuild
455 tables and indexes will be generated automatically. If you are trying
456 to automate the upgrade of many clusters, you should find that clusters
457 with identical database schemas require the same post-upgrade steps for
458 all cluster upgrades; this is because the post-upgrade steps are based
459 on the database schemas, and not user data.
460 For deployment testing, create a schema-only copy of the old cluster,
462 insert dummy data, and upgrade that.
463 pg_upgrade does not support upgrading of databases containing table
465 columns using these reg* OID-referencing system data types:
466 regcollation
467 regconfig
468 regdictionary
469 regnamespace
470 regoper
471 regoperator
472 regproc
473 regprocedure
474 (regclass, regrole, and regtype can be upgraded.)
475 If you want to use link mode and you do not want your old cluster to be
477 modified when the new cluster is started, consider using the clone
478 mode. If that is not available, make a copy of the old cluster and
479 upgrade that in link mode. To make a valid copy of the old cluster, use
480 rsync to create a dirty copy of the old cluster while the server is
481 running, then shut down the old server and run rsync --checksum again
482 to update the copy with any changes to make it consistent. (--checksum
483 is necessary because rsync only has file modification-time granularity
484 of one second.) You might want to exclude some files, e.g.,
485 postmaster.pid, as documented in Section 26.3.3. If your file system
486 supports file system snapshots or copy-on-write file copies, you can
487 use that to make a backup of the old cluster and tablespaces, though
488 the snapshot and copies must be created simultaneously or while the
489 database server is down.
490
492 initdb(1), pg_ctl(1), pg_dump(1), postgres(1)
493PostgreSQL 16.1 2023 PG_UPGRADE(1)