Commit Graph

443 Commits

Author SHA1 Message Date
Bruce Momjian
8a7598091a In pg_upgrade, improve error reporting when the number of relation
objects does not match between the old and new clusters.

Backpatch to 9.2.
2012-10-02 11:53:45 -04:00
Bruce Momjian
ac96b851ec Adjust pg_upgrade query so toast tables related to system catalog schema
entries are not dumped.   This fixes an error caused by
droping/recreating the information_schema, but other failures were also
possible.

Backpatch to 9.2.
2012-10-02 11:46:08 -04:00
Bruce Momjian
b61837a49f In pg_upgrade, try to convert the locale names to canonical form before
comparison;  also report the old/new values if they don't match.

Backpatch to 9.2.
2012-10-02 11:42:34 -04:00
Peter Eisentraut
10bfe81dee pg_upgrade test: Disable fsync in initdb and postgres calls
This mirrors the behavior of pg_regress and makes the test run much
faster.
2012-09-26 22:41:57 -04:00
Peter Eisentraut
5cfd5bb15b pg_upgrade: Remove check for pg_config
It is no longer used, but was still being checked for.

bug #7548 from Reinhard Max
2012-09-18 22:04:28 -04:00
Andrew Dunstan
f8c81c5dde In pg_upgrade, try a few times to open a log file.
If we call pg_ctl stop, the server might continue and thus
hold a log file for a short time after it has deleted its pid file,
(which is when pg_ctl will exit), and so a subsequent attempt to
open the log file might fail.

We therefore try to open it a few times, sleeping one second between
tries, to give the server time to exit.

This corrects an error that was observed on the buildfarm.

Backpatched to 9.2,
2012-09-05 23:14:49 -04:00
Andrew Dunstan
f8f5cf33a3 Fix pg_upgrade test script's line end handling on Windows.
Call pg_dumpall using -f switch instead of redirection, to avoid
writing the output in text mode and generating spurious carriage
returns. Remove to carriage return ignoring hack introduced by
commit e442b0f0c6.

Backpatch to 9.2.
2012-09-05 18:00:31 -04:00
Andrew Dunstan
ea0b414a0d Fix line end mishandling in pg_upgrade on Windows.
pg_upgrade opened the output from pg_dumpall in text mode and
wrote the split files in text mode. This caused unwanted eating
of intended carriage returns on input and production of spurious
carriage returns on output. To avoid this, open all these files
in binary mode. On non-Windows platforms, this change has no
effect.

Backpatch to 9.0. On 9.0 and 9.1, we also switch from redirecting
pg_dumpall's output to using pg_dumpall's -f switch, for the same
reason.
2012-09-05 17:41:43 -04:00
Tom Lane
b98fd52a55 Silence -Wunused-result warning in contrib/pg_upgrade.
This is just neatnik-ism, but since we do it for comparable code in elog.c,
we may as well do it here.
2012-09-05 14:37:22 -04:00
Bruce Momjian
022cd22f0f In pg_upgrade, document why we can't issue \n\n in the command logfile
on Windows.  Slightly cleanup log output on Windows given this
restriction.

Backpatch to 9.2.
2012-09-05 00:01:13 -04:00
Andrew Dunstan
2042185baf Fix transcription error. 2012-09-04 09:39:49 -04:00
Andrew Dunstan
0829c7087e Fix command echoing in pg_upgade's analyze script for Windows. 2012-09-04 05:49:22 -04:00
Andrew Dunstan
2f0c7d5854 Indent fix_path_separator() header properly. 2012-09-03 22:59:19 -04:00
Andrew Dunstan
e442b0f0c6 Allow pg_upgrade "make check" to run on Windows/MSys.
Backpatch to 9.2.
2012-09-03 22:32:17 -04:00
Bruce Momjian
a80b8037cf In pg_upgrade, pull the port number from postmaster.pid, like we do for
socket location.  Also, prevent putting the socket in the current
directory for pre-9.1 servers in live check and non-live check mode,
because pre-9.1 pg_ctl -w can't handle it.

Backpatch to 9.2.
2012-09-03 22:15:09 -04:00
Andrew Dunstan
504aeea692 Use correct path separator for Windows builtin commands.
pg_upgrade produces a platform-specific script to remove the old
directory, but on Windows it has not been making sure that the
paths it writes as arguments for rmdir and del use the backslash
path separator, which will cause these scripts to fail.

The fix is backpatched to Release 9.0.
2012-09-03 18:06:47 -04:00
Andrew Dunstan
6259678f8f Fix bugs in exec.c that prevented pg_upgrade working in Windows.
Backpatch to 9.2 - code before that is quite different and should
not have these defects.
2012-09-03 15:31:26 -04:00
Tom Lane
f763b77193 Fix pg_upgrade to cope with non-default unix_socket_directory scenarios.
When starting either an old or new postmaster, force it to place its Unix
socket in the current directory.  This makes it even harder for accidental
connections to occur during pg_upgrade, and also works around some
scenarios where the default socket location isn't usable.  (For example,
if the default location is something other than "/tmp", it might not exist
during "make check".)

When checking an already-running old postmaster, find out its actual socket
directory location from postmaster.pid, if possible.  This dodges problems
with an old postmaster having a configured location different from the
default built into pg_upgrade's libpq.  We can't find that out if the old
postmaster is pre-9.1, so also document how to cope with such scenarios
manually.

In support of this, centralize handling of the connection-related command
line options passed to pg_upgrade's subsidiary programs, such as pg_dump.
This should make future changes easier.

Bruce Momjian and Tom Lane
2012-09-03 13:52:40 -04:00
Alvaro Herrera
088c065ce8 pg_upgrade: Fix exec_prog API to be less flaky
The previous signature made it very easy to pass something other than
the printf-format specifier in the corresponding position, without any
warning from the compiler.

While at it, move some of the escaping, redirecting and quoting
responsibilities from the callers into exec_prog() itself.  This makes
the callsites cleaner.
2012-08-27 14:29:14 -04:00
Alvaro Herrera
34c02044ed Fix thinko in comment
It was confusing symlinks with hard links.
2012-08-27 14:29:14 -04:00
Alvaro Herrera
ab577e63fa Remove analyze_new_cluster.sh on make clean, too 2012-08-27 14:29:14 -04:00
Peter Eisentraut
7afa8bed65 pg_upgrade: Run the created scripts in the test suite
Just to check that they actually work.
2012-08-23 02:29:21 -04:00
Bruce Momjian
85642ec00b Prevent pg_upgrade from crashing if it can't write to the current
directory.

Backpatch to 9.2.
2012-08-10 17:14:48 -04:00
Bruce Momjian
fbcfa90bb8 Fix pg_upgrade file share violation on Windows created by the commit
4741e9afb9.  This was done by adding an
optional second log file parameter to exec_prog(), and closing and
reopening the log file between system() calls.

Backpatch to 9.2.
2012-08-07 13:10:44 -04:00
Bruce Momjian
e8969c4733 In pg_upgrade, use pg_log() instead of prep_status() for
newline-terminated messages, per suggestion from Tom.

Backpatch to 9.2.
2012-08-03 12:43:37 -04:00
Bruce Momjian
4da8fc05f0 Simplify pg_upgrade's handling when returning directory listings.
Backpatch to 9.2.
2012-07-26 06:22:22 -04:00
Tom Lane
faf26bf117 Get rid of useless global variable in pg_upgrade.
Since the scandir() emulation was taken out of pg_upgrade, there's
no longer any need for scandir_file_pattern to exist as a global
variable.  Replace it with a local in the one remaining function
that was making use of it.
2012-07-18 01:23:12 -04:00
Tom Lane
3d6ec663bb Improve pg_upgrade's load_directory() function.
Error out on out-of-memory, rather than returning -1, which the sole
existing caller wasn't checking for anyway.  There doesn't seem to be
any use-case for making the caller check for failure here.

Detect failure return from readdir().

Use a less platform-dependent method of calculating the entrysize.
It's possible, but not yet confirmed, that this explains bug #6733,
in which Mike Wilson reports a pg_upgrade crash that did not occur
in 9.1.  (Note that load_directory is effectively new code in 9.2,
at least on platforms that have scandir().)

Fix up comments, avoid uselessly using two counters, reduce the number
of realloc calls to something sane.
2012-07-18 01:13:20 -04:00
Bruce Momjian
c742d1dbe7 Update pg_upgrade comments for recent configpath fix. 2012-07-06 09:39:22 -04:00
Bruce Momjian
2eeb5eb23f Fix PGDATAOLD and PGDATANEW to properly set pgconfig location, per
report from Tom.

Backpatch to 9.2.
2012-07-05 23:36:43 -04:00
Alvaro Herrera
666d494d19 pg_upgrade: abstract out copying of files from old cluster to new
Currently only pg_clog is copied, but some other directories could need
the same treatment as well, so create a subroutine to do it.

Extracted from my (somewhat larger) FOR KEY SHARE patch.
2012-07-05 11:38:42 -04:00
Alvaro Herrera
4741e9afb9 Make the pg_upgrade log files contain actual commands
Now the log file not only contains the output from commands executed by
system(), but also what command it was in the first place.  This
arrangement makes debugging a lot simpler.
2012-06-29 11:39:11 -04:00
Alvaro Herrera
9e26326ad6 pg_upgrade: fix off-by-one mistake in snprintf
snprintf counts trailing NUL towards the char limit.  Failing to account
for that was causing an invalid value to be passed to pg_resetxlog -l,
aborting the upgrade process.
2012-06-28 23:37:27 -04:00
Heikki Linnakangas
038f3a0509 Fix pg_upgrade, broken by the xlogid/segno -> 64-bit int refactoring.
The xlogid + segno representation of a particular WAL segment doesn't make
much sense in pg_resetxlog anymore, now that we don't use that anywhere
else. Use the WAL filename instead, since that's a convenient way to name a
particular WAL segment.

I did this partially for pg_resetxlog in the original xlogid/segno -> uint64
patch, but I neglected pg_upgrade and the docs. This should now be more
complete.
2012-06-26 07:49:02 +03:00
Peter Eisentraut
bb7520cc26 Make documentation of --help and --version options more consistent
Before, some places didn't document the short options (-? and -V),
some documented both, some documented nothing, and they were listed in
various orders.  Now this is hopefully more consistent and complete.
2012-06-18 02:46:59 +03:00
Bruce Momjian
877b55c619 Improve pg_upgrade wording for pg_ctl start failure; could be
connection failure.

Backpatch to 9.2.

Per report from Evan D. Hoffman
2012-06-15 19:57:01 -04:00
Bruce Momjian
b9212e379c In pg_upgrade, verify that the install user has the same oid on both
clusters, and make sure the new cluster has no additional users.

Backpatch to 9.1.
2012-06-13 12:19:18 -04:00
Bruce Momjian
927d61eeff Run pgindent on 9.2 source tree in preparation for first 9.3
commit-fest.
2012-06-10 15:20:04 -04:00
Bruce Momjian
8fcbfea788 In pg_upgrade, report pre-PG 8.1 plpython helper functions left in the
public schema that no longer point to valid shared object libraries, and
suggest a solution.
2012-06-01 11:40:04 -04:00
Bruce Momjian
4498a3407a Improve pg_upgrade C comment. 2012-05-27 06:45:29 -04:00
Bruce Momjian
20a3830b81 Add C comment explaining why we can't exclude checking functions in the
pg_catalog schema, even though they are not explicitly dumped (they are
implicitly dumped, e.g. create language plperl).
2012-05-27 06:35:10 -04:00
Bruce Momjian
d4696fa68f Add pg_update C comment about problems with plpython_call_handler(). 2012-05-27 06:29:45 -04:00
Bruce Momjian
45ca31d6a7 Have pg_upgrade only use one extra log file for Win32, not two. 2012-05-25 09:10:07 -04:00
Bruce Momjian
d226e236c6 On Windows, have pg_upgrade use different two files to log pg_ctl
start/stop output, to fix file share error reported by Edmund Horner
2012-05-24 20:30:39 -04:00
Bruce Momjian
476291bef0 Adjust pg_upgrade to output a separate log file for pg_ctl output on
Windows, to avoid opening a file by multiple processes.
2012-05-23 20:19:34 -04:00
Peter Eisentraut
c8e086795a Remove whitespace from end of lines
pgindent and perltidy should clean up the rest.
2012-05-15 22:19:41 +03:00
Peter Eisentraut
081ca7a0d1 Tweak contrib --help output to match common style
Placeholders such as OPTION are typically kept in singular.
2012-05-08 21:55:05 +03:00
Robert Haas
5d4b60f2f2 Lots of doc corrections.
Josh Kupershmidt
2012-04-23 22:43:09 -04:00
Bruce Momjian
38458e4573 Fix pg_upgrade to properly upgrade a table that is stored in the cluster
default tablespace, but part of a database that is in a user-defined
tablespace.  Caused "file not found" error during upgrade.

Per bug report from Ants Aasma.

Backpatch to 9.1 and 9.0.
2012-04-10 19:57:14 -04:00
Tom Lane
98316e211b Silence compiler warning about uninitialized variable. 2012-03-27 14:47:23 -04:00
Peter Eisentraut
1b12c3e836 pg_upgrade: Add new generated file to .gitignore 2012-03-20 20:40:31 +02:00
Bruce Momjian
02b183acb4 In pg_upgrade, remove dependency on pg_config, as that might not be in
the non-development install.  Instead, use the LOAD mechanism to check
for the pg_upgrade_support shared object, like we do for other shared
object checks.

Backpatch to 9.1.

Report from Àlvaro
2012-03-19 09:31:50 -04:00
Bruce Momjian
ced53c38d7 In pg_upgrade, move new echo quote define into include file. 2012-03-17 09:56:56 -04:00
Bruce Momjian
faec2815c8 In pg_upgrade, create a script to incrementally generate more accurate
optimizer statistics so the cluster can be made available sooner.
2012-03-16 18:54:23 -04:00
Bruce Momjian
7c8cfed78a Uppercase pg_upgrade status output title. 2012-03-14 19:22:11 -04:00
Bruce Momjian
717f6d6085 In pg_upgrade, add various logging improvements:
add ability to control permissions of created files
	have psql echo its queries for easier debugging
	output four separate log files, and delete them on success
	add -r/--retain option to keep log files after success
	make logs file append-only
	remove -g/-G/-l logging options
	sugggest tailing appropriate log file on failure
	enhance -v/--verbose behavior
2012-03-12 19:48:03 -04:00
Bruce Momjian
2127aac6ef In pg_upgrade, only lock the old cluster if link mode is used, and do it
right after we restore the schema (a common failure point), and right
before we do the link operation.

Per suggesgtions from Robert and ^!C^!^@lvaro
2012-03-05 21:20:06 -05:00
Bruce Momjian
4e6092a227 Improve wording of pg_upgrade logfile option:
-l, --logfile=FILENAME        log internal activity to file\n\
2012-02-22 18:40:32 -05:00
Bruce Momjian
7f951474dc Have pg_upgrade output 'link' or 'copy' when reporting its status.
Idea from Peter.
2012-02-15 14:23:03 -05:00
Bruce Momjian
3946c28161 Fix pg_upgrade --help text for --user to not use a plural cluster name. 2012-01-26 22:18:58 -05:00
Bruce Momjian
6d5aae7afa Now that the shared library name can be adjusted in the library test,
have pg_upgrade allocate a maximum fixed size buffer for testing the
library file name, rather than base the allocation on the library name.

Backpatch to 9.1.
2012-01-25 09:35:17 -05:00
Bruce Momjian
a7f2c79a6e Improve plpython fix comment in pg_upgrade. 2012-01-24 22:49:55 -05:00
Bruce Momjian
cfe443ab9d In pg_upgrade, when checking for the plpython library, we must check for
"plpython2" when upgrading from pre-PG 9.1.  Patch to head and 9.1.

Per report from Peter.
2012-01-24 22:42:37 -05:00
Bruce Momjian
b95aec5290 Fix new SQL tablespace location function usage in pg_upgrade to properly
check cluster version numbers, and fix missing table alias.
2012-01-24 21:51:10 -05:00
Bruce Momjian
ecf7a2ea38 Add pg_upgrade C comment about why we check all relkinds for regtypes. 2012-01-19 16:04:34 -05:00
Bruce Momjian
e126958c2e Update copyright notices for year 2012. 2012-01-01 18:01:58 -05:00
Peter Eisentraut
d383c23f6f Remove support for on_exit()
All supported platforms support the C89 standard function atexit()
(SunOS 4 probably being the last one not to), and supporting both
makes the code clumsy.
2011-12-27 20:57:59 +02:00
Peter Eisentraut
c317a3ac16 Run "make all" as a prerequisite of "make check"
This is the standard behavior but was forgotten in some places.
2011-12-27 20:27:24 +02:00
Magnus Hagander
16d8e594ac Remove spclocation field from pg_tablespace
Instead, add a function pg_tablespace_location(oid) used to return
the same information, and do this by reading the symbolic link.

Doing it this way makes it possible to relocate a tablespace when the
database is down by simply changing the symbolic link.
2011-12-07 10:37:33 +01:00
Bruce Momjian
0e8f6bf0e7 In pg_upgrade, allow tables using regclass to be upgraded because we
preserve pg_class oids since PG 9.0.
2011-12-05 16:45:19 -05:00
Tom Lane
9c6aa8cca0 Move EXTRA_CLEAN to where it actually works. 2011-11-29 23:14:16 -05:00
Peter Eisentraut
b258ceb3ad pg_upgrade: Disable installcheck
Disabled for now because some build farm members with low resources
are not prepared to handle it.
2011-11-29 06:57:09 +02:00
Peter Eisentraut
08da2d282f Add pg_upgrade test suite
It runs the regression tests, runs pg_upgrade on the populated
database, and compares the before and after dumps.  While not actually
a cross-version upgrade, this does detect omissions and bugs in the
involved tools from time to time.  It's also possible to do a
cross-version upgrade by manually supplying parameters.
2011-11-27 22:42:32 +02:00
Bruce Momjian
32fb4546e3 Remove ENABLE_SAME_CATVERSION_UPGRADES as unnecessary. 2011-11-25 19:08:41 -05:00
Bruce Momjian
5df1403b0f Add pg_upgrade ENABLE_SAME_CATVERSION_UPGRADES macro for testing to
allow upgrades of the same catalog version.  (Doesn't work for
tablespaces, as indicated by C comment.)
2011-11-23 22:51:45 -05:00
Bruce Momjian
7012b5edb7 Remove scandir() requirement in pg_upgrade; instead just use readdir()
--- we were not using the scandir pattern filtering anyway.  This also
removes the scandir requirement in configure.
2011-11-17 21:59:49 -05:00
Bruce Momjian
ee3ef8f30c Fix pg_upgrade's pg_scandir_internal() the right way. Backpatch to 9.1. 2011-11-17 13:40:44 -05:00
Bruce Momjian
988f907d24 Fix pg_upgrade's pg_scandir_internal() to properly handle a NULL
pattern, which is used on PG 9.1 and HEAD (but not pre-9.1).  Fixes
crash on Windows.

Backpatched to 9.1.

Reported by Mark Dilger
2011-11-17 13:24:54 -05:00
Bruce Momjian
b735757c32 Remove pg_upgrade function no longer called (dir_matching_filenames). 2011-11-17 13:17:33 -05:00
Robert Haas
b76c61f1e8 Silence bogus compiler warning. 2011-11-04 10:40:52 -04:00
Robert Haas
c9f48b572c Check the return value of getcwd(), instead of assuming success.
Kevin Grittner
2011-11-04 10:40:25 -04:00
Bruce Momjian
84b8fcaa92 Adjust pg_upgrade "new database skip" code, e.g. 'postgres', to more
cleanly handle old/new database mismatches.
2011-11-03 13:57:52 -04:00
Bruce Momjian
09d1174e5a Update pg_upgrade comment on missing 'postgres' database. 2011-11-01 14:34:35 -04:00
Bruce Momjian
a50d860ae1 Allow pg_upgrade to upgrade an old cluster that doesn't have a
'postgres' database.
2011-11-01 13:49:48 -04:00
Bruce Momjian
51eba98cf4 Remove pg_upgrade dependency on the 'postgres' database existing in the
new cluster.   vacuumdb, used by pg_upgrade, still has this dependency.
2011-10-28 21:19:21 -04:00
Bruce Momjian
9846dcfb9e Clarify pg_upgrade error message that the 'postgres' database must exist
in the old cluster.
2011-10-28 08:09:15 -04:00
Bruce Momjian
38f3c7c4dd Update pg_upgrade testing instructions. 2011-10-27 21:29:03 -04:00
Bruce Momjian
0dc3f57ba0 In pg_upgrade, add -o/-O options to pass parameters to the servers, and
document its use for config-only directory installs.
2011-10-10 07:44:11 -04:00
Bruce Momjian
caa1054df8 In pg_upgrade, use the new postmaster -C option to get the real data
directory, for config-only directory installs.  Only works for PG 9.2+
servers.
2011-10-07 14:40:23 -04:00
Bruce Momjian
a3996754cc In pg_upgrade, improve popen() failure detection by checking for fgets()
failures.
2011-10-06 19:37:29 -04:00
Bruce Momjian
878b74e094 In pg_upgrade, remove unnecessary local variable. 2011-10-01 10:01:43 -04:00
Bruce Momjian
5e5958428b In pg_upgrade, because toast table names can be mismatched with the heap
oid on 8.4, modify the toast name comparison test to only apply to old
9.0+ servers.  (The test was previously 8.4+.)

Backpatch to 9.1.X.
2011-09-28 22:30:44 -04:00
Peter Eisentraut
52ce20589a Add missing format attributes
Add __attribute__ decorations for printf format checking to the places that
were missing them.  Fix the resulting warnings.  Add
-Wmissing-format-attribute to the standard set of warnings for GCC, so these
don't happen again.

The warning fixes here are relatively harmless.  The one serious problem
discovered by this was already committed earlier in
cf15fb5cab.
2011-09-10 23:12:46 +03:00
Bruce Momjian
3ced32d24e In pg_upgrade, disallow migration of 8.3 clusters using contrib/ltree
because its internal format was changed in 8.4.

Backpatch to 9.0 and 9.1.

Report by depesz, diagnosis by Tom.
2011-09-07 14:43:12 -04:00
Bruce Momjian
6416a82a62 Remove unnecessary #include references, per pgrminclude script. 2011-09-01 10:04:27 -04:00
Tom Lane
731ebb64b7 Fix not-backwards-compatible pg_upgrade test for prepared transactions.
There's no reason for this test to use the undocumented pg_prepared_xact()
function, when it can use the stable API pg_prepared_xacts instead.
Fixes breakage against 8.3, as reported by Justin Arnold.
2011-08-30 17:15:00 -04:00
Bruce Momjian
f1312b5ed3 Add postgres.h to *.c files for pg_upgrade, ltree, and btree_gist, and
remove from local *.h files.

Per suggestion from Alvaro.
2011-08-26 21:16:24 -04:00
Bruce Momjian
eb013ede59 In pg_upgrade, limit schema name filter to include toast tables. Bug
introduced recently when trying to filter out temp tables.

Backpatch to 9.0 and 9.1.
2011-08-26 00:12:32 -04:00
Bruce Momjian
00a7c9014a In pg_upgrade, don't copy visibility map files from clusters that did not
have crash-safe visibility maps to clusters that expect crash-safety.

Request from Robert Haas.
2011-08-19 11:20:30 -04:00
Bruce Momjian
2411fbdac4 In pg_upgrade, avoid dumping orphaned temporary tables. This makes the
pg_upgrade schema matching pattern match pg_dump/pg_dumpall.

Fix for 9.0, 9.1, and 9.2.
2011-08-15 22:40:14 -04:00
Bruce Momjian
a31dc392d6 Fix pg_update to properly test for the data directory's existence on
Win32.

Backpatch to 9.1.
2011-07-30 01:50:55 -04:00
Bruce Momjian
081a5518c0 In pg_upgrade on Windows, check if the directory is writable by actually
creating and removing a file because access() doesn't work on that
platform.

Backpatch to 9.1 where this check was added.
2011-07-24 01:43:57 -04:00
Bruce Momjian
43aa40e155 In pg_upgrade, add C comment about why we don't try to do shared file
writes for logging.
2011-07-21 17:13:17 -04:00
Bruce Momjian
17a16eeb7c In pg_upgrade, fix the -l/log option to work on Windows.
Also, double-quote the log file name in all places, to allow (on all
platforms) log file names with spaces.

Back patch to 9.0 and 9.1.
2011-07-20 18:31:03 -04:00
Bruce Momjian
6b43fddee4 In pg_upgrade, use pg_strudup(), for consistency. 2011-07-20 16:37:17 -04:00
Peter Eisentraut
912bc4f038 Make pg_upgrade output more consistent with project style
Add errno-based output to error messages where appropriate, reformat
blocks to about 72 characters per line, use spaces instead of tabs for
indentation, and other style adjustments.
2011-07-12 07:13:51 +03:00
Peter Eisentraut
d7fb49314d Fix use of unportable %m format 2011-07-07 22:48:59 +03:00
Peter Eisentraut
76dfcb942f Postgres -> PostgreSQL in error message 2011-07-05 22:41:14 +03:00
Bruce Momjian
a88f4496b7 Change pg_upgrade to use port 50432 by default to avoid unintended
client connections during the upgrade.  Also rename data/bin/port
environment variables to being with 'PG'.  Also no longer honor PGPORT.
2011-07-01 18:17:12 -04:00
Bruce Momjian
0b44818ead In pg_upgrade, check that the binary and data directories are the same
major version.

Backpatch to 9.1.

Dan McGee
2011-06-22 20:48:34 -04:00
Bruce Momjian
3b3c2cf180 In pg_upgrade, fix odd function parameter wrapping. 2011-06-22 19:30:46 -04:00
Bruce Momjian
effbe6ecb4 Rename pg_upgrade directory validation function, for clarity. 2011-06-22 17:50:40 -04:00
Bruce Momjian
559b114dd4 Adjust pg_upgrade check for pg_upgrade_support to happen after the
binary directory has been validated.

Backpatch to 9.1.

Dan McGee
2011-06-22 17:47:23 -04:00
Bruce Momjian
b06ad7def8 Fix pg_upgrade status message capitalization mistake.
Backpatch to 9.1 and 9.0.

Dan McGee
2011-06-22 14:49:54 -04:00
Bruce Momjian
e3df3572f6 In pg_upgrade, clean up code layout in validateDirectoryOption(). 2011-06-15 17:23:02 -04:00
Bruce Momjian
07d17a73ec In pg_upgrade, check there are no prepared transactions. 2011-06-14 14:53:35 -04:00
Bruce Momjian
6560407c7d Pgindent run before 9.1 beta2. 2011-06-09 14:32:50 -04:00
Bruce Momjian
11c08c3fd7 In pg_upgrade, do case-insensitive checks of locale, encoding, and ctype
because these are often inconsistently capitalized.
2011-05-24 15:59:34 -04:00
Bruce Momjian
1f35944607 In pg_upgrade, clean up handling of invalid directory specification by
checking the stat() errno value more strictly.
2011-05-18 22:22:40 -04:00
Bruce Momjian
772a5f1e5b Improve pg_upgrade error reporting if the bin or data directories do not
exist or are not directories.
2011-05-18 18:37:25 -04:00
Peter Eisentraut
0ee391b77a /contrib/foo -> contrib/foo
Since contrib is a relative directory specification, a leading slash
is inappropriate.
2011-05-19 00:30:31 +03:00
Bruce Momjian
92ff915018 Improve pg_upgrade X_OK comment. 2011-05-18 12:25:34 -04:00
Bruce Momjian
6fbf4e0e7c Fix pg_upgrade build problem on Windows when using X_OK access
permission check on the current directory.
2011-05-18 12:13:37 -04:00
Bruce Momjian
e728701fb4 Update pg_upgrade directory check error message. 2011-05-16 12:32:36 -04:00
Bruce Momjian
bda27e502f Add pg_upgrade check to make sure the user has full access permission in
the current directory;  if not, throw an error.
2011-05-16 11:01:29 -04:00
Bruce Momjian
6c19bd96bc Allow pg_upgrade to honor libpq environment variables. Add 'local'
checks for PGHOST and PGHOSTADDR.
2011-05-16 10:48:40 -04:00
Peter Eisentraut
00fad9f6fe Format pg_upgrade --version in line with conventions 2011-05-10 21:40:29 +03:00
Peter Eisentraut
4c78846cb3 Add bug report line to pg_upgrade --help output 2011-05-10 21:37:55 +03:00
Bruce Momjian
5059cf6ebf In pg_upgrade, remove suggestion of setting pg_hba.conf to 'trust', now
that we report the libpq connection failure string.  Per suggestion from
Robert Haas.
2011-05-10 10:42:43 -04:00
Bruce Momjian
78318d63d7 In pg_upgrade, add status message about superuser check. 2011-05-09 08:55:36 -04:00
Bruce Momjian
e6a7402842 Add C comment why client encoding can be set in pg_upgrade. 2011-05-07 22:30:35 -04:00
Bruce Momjian
1609ca5adb In pg_upgrade, report non-super-user username in error message. 2011-05-07 12:17:21 -04:00
Bruce Momjian
81301b8578 Check that the pg_upgrade user specified is a super-user.
Also report the error message when the post-pg_ctl connection fails.

Per private bug report from EnterpriseDB.
2011-05-07 08:55:45 -04:00
Bruce Momjian
5c5f83507c Adjust pg_upgrade FATAL error messages to have consistent newlines.
Also adjust some error message capitalization for consistency.
2011-05-06 21:47:42 -04:00
Peter Eisentraut
d6d823c8b0 Improve formatting of pg_upgrade --help output 2011-05-05 23:19:13 +03:00
Bruce Momjian
6c4d2bd914 Now that pg_upgrade uses -w in pg_ctl, remove loop that retried testing
the connection;  also restructure the libpq connection code.

This patch also removes the unused variable postmasterPID and fixes a
libpq structure leak that was in the testing loop.
2011-04-26 10:15:18 -04:00
Bruce Momjian
44091442db In pg_upgrade, avoid one start/stop of the postmaster; use the -w
(wait) flag for pg_ctl start/stop;  remove the unused "quiet" flag in
the functions for starting/stopping the postmaster.
2011-04-25 20:18:23 -04:00
Bruce Momjian
76dd09bbec Add postmaster/postgres undocumented -b option for binary upgrades.
This option turns off autovacuum, prevents non-super-user connections,
and enables oid setting hooks in the backend.  The code continues to use
the old autoavacuum disable settings for servers with earlier catalog
versions.

This includes a catalog version bump to identify servers that support
the -b option.
2011-04-25 12:00:21 -04:00
Bruce Momjian
f6322b3191 In pg_upgrade, only compile copy_file() on non-Win32 systems.
Per report from Andrew Dunstan.
2011-04-23 20:28:29 -04:00
Bruce Momjian
0262251c33 Pg_upgrade C comment addition.
Document why we do the missing new database check during the check
phase.
2011-04-20 05:45:31 -04:00
Bruce Momjian
7228d02989 Throw error for mismatched pg_upgrade clusters
If someone removes the 'postgres' database from the old cluster and the
new cluster has a 'postgres' database, the number of databases will not
match.  We actually could upgrade such a setup, but it would violate the
1-to-1 mapping of database counts, so we throw an error instead.

Previously they got an error during the upgrade, and not at the check
stage; PG 9.0.4 does the same.
2011-04-19 21:00:29 -04:00
Bruce Momjian
0341944706 Add C comment
Add C comment about why we throw an error if the pg_upgrade old/new
database counts don't match.
2011-04-19 19:15:13 -04:00
Bruce Momjian
bf50caf105 pgindent run before PG 9.1 beta 1. 2011-04-10 11:42:00 -04:00
Peter Eisentraut
1eb2231fc4 Allow pg_upgrade with PGCLIENTENCODING set
This used to work, but since PGCLIENTENCODING is now a connection
option variable, pg_upgrade would prevent it.
2011-04-07 19:24:47 +03:00
Bruce Momjian
e69d32158c Simplify structure of query used to gather database object information
in pg_upgrade.
2011-04-07 06:23:12 -04:00
Peter Eisentraut
c75163842c Replace the confusing exit_nicely() by an atexit/on_exit hook 2011-04-06 23:00:44 +03:00
Bruce Momjian
da3418cea9 Mention pg_upgrade sets autovacuum_freeze_max_age to maximum. 2011-03-30 17:45:54 -04:00
Bruce Momjian
d609b08ead Expand comment on how pg_upgrade is turning off autovacuum. 2011-03-30 17:37:00 -04:00
Bruce Momjian
d67b0bf471 In pg_upgrade, add C comment about how autovacuum is disabled. 2011-03-29 11:11:45 -04:00
Bruce Momjian
898a14e1a0 Remove unused copy_dir() function from pg_upgrade. 2011-03-15 20:52:35 -04:00
Bruce Momjian
d367d41d66 Fix file descriptor leaks in pg_upgrade in failure code paths. 2011-03-08 21:36:17 -05:00
Bruce Momjian
05d93c38a7 Tighten pg_upgrade check for pre-8.4 toast table name matching. 2011-03-06 21:57:37 -05:00
Tom Lane
dfe18f18d2 Fix a couple more missing "static" markers. 2011-03-06 20:14:01 -05:00
Bruce Momjian
a54ba23c08 Improve pg_upgrade relation name check logic for pre-8.4 servers. 2011-03-06 06:34:58 -05:00
Bruce Momjian
8f87dcd863 Update new pg_upgrade comment about pre-8.4 TOAST tables. 2011-03-05 22:09:35 -05:00
Bruce Momjian
9e5bed2df1 Restructure pg_upgrade checks because pre-8.4 Postgres did not rename
toast file names to match the new relfilenode.
2011-03-05 21:12:21 -05:00
Bruce Momjian
f7b70dfc76 Fix pg_upgrade to print the proper database name for file transfer
failures.
2011-03-05 20:18:31 -05:00
Bruce Momjian
a3375becfa Print clearer failure message when pg_upgrade fails due to a fatal
error.
2011-03-05 19:32:53 -05:00
Bruce Momjian
497e65f822 In pg_upgrade, no need to initialize global struct values; they are
always zeros.  Also no need to free memory before we exit.
2011-02-16 15:17:28 -05:00
Alvaro Herrera
b4a7e5980b Cleanup ClusterInfo initialization in pg_upgrade 2011-02-16 14:06:17 -03:00
Bruce Momjian
1cc19cc358 Fix bug in 9.1 pg_upgrade processing of old/new relations; adjust debug
output.
2011-02-15 19:01:33 -05:00
Alvaro Herrera
8b2557553c Make pg_upgrade compile again 2011-02-15 19:52:32 -03:00
Bruce Momjian
fe8f15d5d2 Adjust pg_upgrade error message, array freeing, and add error check. 2011-02-15 15:00:47 -05:00
Tom Lane
629b3af27d Convert contrib modules to use the extension facility.
This isn't fully tested as yet, in particular I'm not sure that the
"foo--unpackaged--1.0.sql" scripts are OK.  But it's time to get some
buildfarm cycles on it.

sepgsql is not converted to an extension, mainly because it seems to
require a very nonstandard installation process.

Dimitri Fontaine and Tom Lane
2011-02-13 22:54:49 -05:00
Tom Lane
caddcb8f4b Fix pg_upgrade to handle extensions.
This follows my proposal of yesterday, namely that we try to recreate the
previous state of the extension exactly, instead of allowing CREATE
EXTENSION to run a SQL script that might create some entirely-incompatible
on-disk state.  In --binary-upgrade mode, pg_dump won't issue CREATE
EXTENSION at all, but instead uses a kluge function provided by
pg_upgrade_support to recreate the pg_extension row (and extension-level
pg_depend entries) without creating any member objects.  The member objects
are then restored in the same way as if they weren't members, in particular
using pg_upgrade's normal hacks to preserve OIDs that need to be preserved.
Then, for each member object, ALTER EXTENSION ADD is issued to recreate the
pg_depend entry that marks it as an extension member.

In passing, fix breakage in pg_upgrade's enum-type support: somebody didn't
fix it when the noise word VALUE got added to ALTER TYPE ADD.  Also,
rationalize parsetree representation of COMMENT ON DOMAIN and fix
get_object_address() to allow OBJECT_DOMAIN.
2011-02-09 19:18:08 -05:00
Tom Lane
cd7d748d51 Repair multiple breakage in Windows-specific code for appending '.exe'. 2011-02-02 20:26:43 -05:00
Bruce Momjian
2b6e2dee78 Fix wrong verb in pg_upgrade text message, per Haas. 2011-02-02 16:53:12 -05:00
Bruce Momjian
9d220fc17e Simplify pg_upgrade checking of executable permissions. 2011-02-02 15:40:52 -05:00
Bruce Momjian
2bb87a61a8 Add pg_upgrade comment for why we can't use template1 inheritance for
the support functions.
2011-02-01 12:23:08 -05:00
Bruce Momjian
6238473adb Fix pg_upgrade to create pg_authid restore functions in the 'postgres'
database, not in the os-user database, per report from Magnus.
2011-01-31 20:16:15 -05:00
Bruce Momjian
0a5f119931 A toast relid field are no longer needed in pg_upgrade's rel arrays, so
remove them.  Also other renaming.
2011-01-10 11:46:02 -05:00
Bruce Momjian
002c105a07 In pg_upgrade, remove functions that did sequential array scans looking
up relations, but rather order old/new relations and use the same array
index value for both.  This should speed up pg_upgrade for databases
with many relations.
2011-01-08 13:44:44 -05:00
Bruce Momjian
ebaf648ad2 In pg_upgrade, clarify use of install_db_support_functions(). 2011-01-08 09:11:48 -05:00
Bruce Momjian
a60b32b3dc In pg_upgrade, remove unnecessary separate handling of toast tables now
that we restore by oid;  they can be handled like regular tables when
creating the file mapping structure.
2011-01-08 08:01:52 -05:00
Bruce Momjian
133641a48e Document why pg_upgrade preserves pg_auth.oid (because they are stored
in pg_largeobject_metadata).
2011-01-07 23:02:03 -05:00
Bruce Momjian
2f657e4d72 More pg_upgrade relfilenode C comments added. 2011-01-07 22:57:30 -05:00
Bruce Momjian
3095643d37 Simplify pg_upgrade C comment about what is preserved. 2011-01-07 22:46:41 -05:00
Bruce Momjian
f75383e6e8 In pg_upgrade, track only one copy of namespace/relname in FileNameMap
because the old and new values are identical.
2011-01-07 22:36:51 -05:00
Bruce Momjian
519c008a96 Fix C indenting in pg_upgrade function. 2011-01-07 22:01:44 -05:00
Bruce Momjian
d8d3d2a4f3 Fix pg_upgrade of large object permissions by preserving pg_auth.oid,
which is stored in pg_largeobject_metadata.

No backpatch to 9.0 because you can't migrate from 9.0 to 9.0 with the
same catversion (because of tablespace conflict), and a pre-9.0
migration to 9.0 has not large object permissions to migrate.
2011-01-07 21:59:29 -05:00
Bruce Momjian
2896c87ce4 Force pg_upgrade's to preserve pg_class.oid, not pg_class.relfilenode.
Toast tables have identical pg_class.oid and pg_class.relfilenode, but
for clarity it is good to preserve the pg_class.oid.

Update comments regarding what is preserved, and do some
variable/function renaming for clarity.
2011-01-07 21:26:13 -05:00
Bruce Momjian
46d28820b6 Improve C comments about backend variables set by pg_upgrade_support
functions.
2011-01-06 22:45:36 -05:00
Bruce Momjian
56a9adcb96 Rename pg_upgrade variables, for clarity. 2011-01-05 20:14:59 -05:00
Bruce Momjian
e66e43988a Update pg_upgrade C comments. 2011-01-05 14:09:01 -05:00
Bruce Momjian
a45beaac64 Fix for recent pg_upgrade commit to fix large objects relfilenode
handling.  (metadata user ids still an open issue).
2011-01-05 13:52:44 -05:00
Bruce Momjian
b1f2a94fa2 In pg_upgrade, adjust transfer_relfile()'s parameters to be more logical. 2011-01-05 12:07:45 -05:00
Bruce Momjian
5cff5b5779 Clarify pg_upgrade's creation of the map file structure. Also clean
up pg_dump's calling of pg_upgrade_support functions.
2011-01-05 11:37:08 -05:00
Bruce Momjian
3302334b48 In pg_upgrade, copy pg_largeobject_metadata and its index for 9.0+
servers because, like pg_largeobject, it is a system table whose
contents are not dumped by pg_dump --schema-only.
2011-01-04 23:35:49 -05:00
Bruce Momjian
0857a84498 In pg_upgrade, fix backward logging display of link operations. 2011-01-04 21:33:36 -05:00
Bruce Momjian
25cc7424e3 Simplify functions and parameters used by pg_upgrade. 2011-01-04 19:11:37 -05:00
Bruce Momjian
5d950e3b0c Stamp copyrights for year 2011. 2011-01-01 13:18:15 -05:00
Bruce Momjian
67c9e4442f Furter pg_upgrade optimizations to reduce function call argument count. 2011-01-01 12:28:48 -05:00
Bruce Momjian
6e6bee987f In pg_upgrade, remove use of whichCluster, and just pass old/new cluster
pointers, which simplifies the code.  This was not possible in 9.0 because
everything was in a single nested struct, but is possible now.

Per suggestion from Tom.
2011-01-01 12:06:36 -05:00
Bruce Momjian
f82b3e58f8 In pg_upgrade, rename "CLUSTERNAME" to "CLUSTER_NAME". 2010-12-31 20:42:29 -05:00
Tom Lane
88c803457a Improve pg_upgrade's checks for required executables.
Don't insist on pg_dumpall and psql being present in the old cluster,
since they are not needed.  Do insist on pg_resetxlog being present
(in both old and new), since we need it.  Also check for pg_config,
but only in the new cluster.  Remove the useless attempt to call
pg_config in the old cluster; we don't need to know the old value of
--pkglibdir.  (In the case of a stripped-down migration installation
there might be nothing there to look at anyway, so any future change
that might reintroduce that need would have to be considered carefully.)

Per my attempts to build a minimal previous-version installation to support
pg_upgrade.
2010-12-29 13:43:53 -05:00
Bruce Momjian
16b5e08dec Use "upgrade" in preference over "migrate" in pg_upgrade messages and
documentation.  (Many were left over from the old pg_migrator naming.)
2010-12-15 07:11:31 -05:00
Tom Lane
3864afa1d1 Clean up some copied-and-pasted code in pg_upgrade.
1. Don't reimplement S_ISDIR() and S_ISREG() badly.
2. Don't reimplement access() badly.

This code appears to have been copied from ancient versions of the
corresponding backend routines, and not patched to incorporate subsequent
fixes (see my commits of 2008-03-31 and 2010-01-14 respectively).
It might be a good idea to change it to just *call* those routines,
but for now I'll just transpose these fixes over.
2010-12-11 14:17:46 -05:00
Peter Eisentraut
fc946c39ae Remove useless whitespace at end of lines 2010-11-23 22:34:55 +02:00
Robert Haas
5aa446c961 Cleanup various comparisons with the constant "true".
Itagaki Takahiro, with slight modifications.
2010-11-14 21:03:48 -05:00
Robert Haas
11e482c350 Move copydir() prototype into its own header file.
Having this in src/include/port.h makes no sense, now that copydir.c lives
in src/backend/strorage rather than src/port.  Along the way, remove an
obsolete comment from contrib/pg_upgrade that makes reference to the old
location.
2010-11-12 16:39:53 -05:00