This is no longer necessary, and might result in a situation where the
configuration file is reloaded (and everything seems OK) but a subsequent
restart of the database fails.
Simon Riggs [Wed, 23 Mar 2011 13:30:05 +0000 (13:30 +0000)]
Prevent intermittent hang in recovery from bgwriter interaction.
Startup process waited for cleanup lock but when hot_standby = off
the pid was not registered, so that the bgwriter would not wake
the waiting process as intended.
Tom Lane [Tue, 22 Mar 2011 21:20:50 +0000 (17:20 -0400)]
Make initdb ignore locales for client-only encodings.
While putting such entries into pg_collation is harmless (since backends
will ignore entries that don't match the database encoding), it's also
useless.
Tom Lane [Tue, 22 Mar 2011 20:55:32 +0000 (16:55 -0400)]
Improve reporting of run-time-detected indeterminate-collation errors.
pg_newlocale_from_collation does not have enough context to give an error
message that's even a little bit useful, so move the responsibility for
complaining up to its callers. Also, reword ERRCODE_INDETERMINATE_COLLATION
error messages in a less jargony, more message-style-guide-compliant
fashion.
Tom Lane [Tue, 22 Mar 2011 19:58:03 +0000 (15:58 -0400)]
Throw error for indeterminate collation of an ORDER/GROUP/DISTINCT target.
This restores a parse error that was thrown (though only in the ORDER BY
case) by the original collation patch. I had removed it in my recent
revisions because it was thrown at a place where collations now haven't
been computed yet; but I thought of another way to handle it.
Throwing the error at parse time, rather than leaving it to be done at
runtime, is good because a syntax error pointer is helpful for localizing
the problem. We can reasonably assume that the comparison function for a
collatable datatype will complain if it doesn't have a collation to use.
Now the planner might choose to implement GROUP or DISTINCT via hashing,
in which case no runtime error would actually occur, but it seems better
to throw error consistently rather than let the error depend on what the
planner chooses to do. Another possible objection is that the user might
specify a nondefault sort operator that doesn't care about collation
... but that's surely an uncommon usage, and it wouldn't hurt him to throw
in a COLLATE clause anyway. This change also makes the ORDER BY/GROUP
BY/DISTINCT case more consistent with the UNION/INTERSECT/EXCEPT case,
which was already coded to throw this error even though the same objections
could be raised there.
Tom Lane [Tue, 22 Mar 2011 17:00:24 +0000 (13:00 -0400)]
Avoid potential deadlock in InitCatCachePhase2().
Opening a catcache's index could require reading from that cache's own
catalog, which of course would acquire AccessShareLock on the catalog.
So the original coding here risks locking index before heap, which could
deadlock against another backend trying to get exclusive locks in the
normal order. Because InitCatCachePhase2 is only called when a backend
has to start up without a relcache init file, the deadlock was seldom seen
in the field. (And by the same token, there's no need to worry about any
performance disadvantage; so not much point in trying to distinguish
exactly which catalogs have the risk.)
Bug report, diagnosis, and patch by Nikhil Sontakke. Additional commentary
by me. Back-patch to all supported branches.
Tom Lane [Tue, 22 Mar 2011 04:34:31 +0000 (00:34 -0400)]
Reimplement planner's handling of MIN/MAX aggregate optimization (again).
Instead of playing cute games with pathkeys, just build a direct
representation of the intended sub-select, and feed it through
query_planner to get a Path for the index access. This is a bit slower
than 9.1's previous method, since we'll duplicate most of the overhead of
query_planner; but since the whole optimization only applies to rather
simple single-table queries, that probably won't be much of a problem in
practice. The advantage is that we get to do the right thing when there's
a partial index that needs the implicit IS NOT NULL clause to be usable.
Also, although this makes planagg.c be a bit more closely tied to the
ordering of operations in grouping_planner, we can get rid of some coupling
to lower-level parts of the planner. Per complaint from Marti Raudsepp.
When two base backups are started at the same time with pg_basebackup,
ensure that they use different checkpoints as the starting point. We use
the checkpoint redo location as a unique identifier for the base backup in
the end-of-backup record, and in the backup history file name.
The local variable "sock" can be unused depending on compilation flags.
But there seems no particular need for it, since the kernel calls can
just as easily say port->sock instead.
Tom Lane [Sun, 20 Mar 2011 16:43:39 +0000 (12:43 -0400)]
Fix up handling of C/POSIX collations.
Install just one instance of the "C" and "POSIX" collations into
pg_collation, rather than one per encoding. Make these instances exist
and do something useful even in machines without locale_t support: to wit,
it's now possible to force comparisons and case-folding functions to use C
locale in an otherwise non-C database, whether or not the platform has
support for using any additional collations.
Fix up severely broken upper/lower/initcap functions, too: the C/POSIX
fastpath now does what it is supposed to, and non-default collations are
handled correctly in single-byte database encodings.
Merge the two separate collation hashtables that were being maintained in
pg_locale.c, and be more wary of the possibility that we fail partway
through filling a cache entry.
Tom Lane [Sun, 20 Mar 2011 00:29:08 +0000 (20:29 -0400)]
Revise collation derivation method and expression-tree representation.
All expression nodes now have an explicit output-collation field, unless
they are known to only return a noncollatable data type (such as boolean
or record). Also, nodes that can invoke collation-aware functions store
a separate field that is the collation value to pass to the function.
This avoids confusion that arises when a function has collatable inputs
and noncollatable output type, or vice versa.
Also, replace the parser's on-the-fly collation assignment method with
a post-pass over the completed expression tree. This allows us to use
a more complex (and hopefully more nearly spec-compliant) assignment
rule without paying for it in extra storage in every expression node.
Fix assorted bugs in the planner's handling of collations by making
collation one of the defining properties of an EquivalenceClass and
by converting CollateExprs into discardable RelabelType nodes during
expression preprocessing.
Magnus Hagander [Sat, 19 Mar 2011 17:44:35 +0000 (18:44 +0100)]
Rename ident authentication over local connections to peer
This removes an overloading of two authentication options where
one is very secure (peer) and one is often insecure (ident). Peer
is also the name used in libpq from 9.1 to specify the same type
of authentication.
Also make initdb select peer for local connections when ident is
chosen, and ident for TCP connections when peer is chosen.
ident keyword in pg_hba.conf is still accepted and maps to peer
authentication.
Robert Haas [Sat, 19 Mar 2011 02:09:57 +0000 (22:09 -0400)]
Fix possible "tuple concurrently updated" error in ALTER TABLE.
When adding an inheritance parent to a table, an AccessShareLock on the
parent isn't strong enough to prevent trouble, so take
ShareUpdateExclusiveLock instead. Since this is a behavior change,
albeit a fairly unobtrusive one, and since we have only one report
from the field, no back-patch.
Report by Jon Nelson, analysis by Alvaro Herrera, fix by me.
Robert Haas [Sat, 19 Mar 2011 01:43:45 +0000 (21:43 -0400)]
Move synchronous_standbys_defined updates from WAL writer to BG writer.
This is advantageous because the BG writer is alive until much later in
the shutdown sequence than WAL writer; we want to make sure that it's
possible to shut off synchronous replication during a smart shutdown,
else it might not be possible to complete the shutdown at all.
Per very reasonable gripes from Fujii Masao and Simon Riggs.
Robert Haas [Fri, 18 Mar 2011 13:44:44 +0000 (09:44 -0400)]
Remove ancient -X options to pg_dump, pg_dumpall, pg_restore.
The last version in which these options were documented is now EOL, so
it's time to get rid of them for real. We now use GNU-style long
options instead.
Peter Eisentraut [Thu, 17 Mar 2011 18:19:51 +0000 (20:19 +0200)]
Raise maximum value of several timeout parameters
The maximum value of deadlock_timeout, max_standby_archive_delay,
max_standby_streaming_delay, log_min_duration_statement, and
log_autovacuum_min_duration was INT_MAX/1000 milliseconds, which is
about 35min, which is too short for some practical uses. Raise the
maximum value to INT_MAX; the code that uses the parameters already
supports that just fine.
Robert Haas [Thu, 17 Mar 2011 17:10:42 +0000 (13:10 -0400)]
Fix various possible problems with synchronous replication.
1. Don't ignore query cancel interrupts. Instead, if the user asks to
cancel the query after we've already committed it, but before it's on
the standby, just emit a warning and let the COMMIT finish.
2. Don't ignore die interrupts (pg_terminate_backend or fast shutdown).
Instead, emit a warning message and close the connection without
acknowledging the commit. Other backends will still see the effect of
the commit, but there's no getting around that; it's too late to abort
at this point, and ignoring die interrupts altogether doesn't seem like
a good idea.
3. If synchronous_standby_names becomes empty, wake up all backends
waiting for synchronous replication to complete. Without this, someone
attempting to shut synchronous replication off could easily wedge the
entire system instead.
4. Avoid depending on the assumption that if a walsender updates
MyProc->syncRepState, we'll see the change even if we read it without
holding the lock. The window for this appears to be quite narrow (and
probably doesn't exist at all on machines with strong memory ordering)
but protecting against it is practically free, so do that.
5. Remove useless state SYNC_REP_MUST_DISCONNECT, which isn't needed and
doesn't actually do anything.
There's still some further work needed here to make the behavior of fast
shutdown plausible, but that looks complex, so I'm leaving it for a
separate commit. Review by Fujii Masao.
Andrew Dunstan [Thu, 17 Mar 2011 04:06:52 +0000 (00:06 -0400)]
Use correct PATH separator for Cygwin in pg_regress.c.
This has been broken for years, and I'm not sure why it has not been
noticed before, but now a very modern Cygwin breaks on it, and the fix
is clearly correct. Backpatching to all live branches.
Tom Lane [Wed, 16 Mar 2011 01:51:36 +0000 (21:51 -0400)]
Improve handling of unknown-type literals in UNION/INTERSECT/EXCEPT.
This patch causes unknown-type Consts to be coerced to the resolved output
type of the set operation at parse time. Formerly such Consts were left
alone until late in the planning stage. The disadvantage of that approach
is that it disables some optimizations, because the planner sees the set-op
leaf query as having different output column types than the overall set-op.
We saw an example of that in a recent performance gripe from Claudio
Freire.
Fixing such a Const requires scribbling on the leaf query in
transformSetOperationTree, but that should be all right since if the leaf
query's semantics depended on that output column, it would already have
resolved the unknown to something else.
Most of the bulk of this patch is a simple adjustment of
transformSetOperationTree's API so that upper levels can get at the
TargetEntry containing a Const to be replaced: it now returns a list of
TargetEntries, instead of just the bare expressions.
Bruce Momjian [Tue, 15 Mar 2011 15:26:20 +0000 (11:26 -0400)]
Add database comments to template0 and postgres databases, and improve
the comments on the template1 database. No catalog version bump because
they are just comments.
Robert Haas [Tue, 15 Mar 2011 15:25:04 +0000 (11:25 -0400)]
Minor sync rep documentation improvements.
- Make the name of the ID tag for the GUC entry match the GUC name.
- Clarify that synchronous_replication waits for xlog flush, not receipt.
- Mention that synchronous_replication won't wait if max_wal_senders=0.
Magnus Hagander [Mon, 14 Mar 2011 18:46:52 +0000 (19:46 +0100)]
Remove special case allowing parameters to ident auth for initdb
This was required in pre-8.4 versions to allow the specification of
"ident sameuser", but sameuser is no longer required. It could be extended
to allow all parameters in the future, but should then apply to all
methods and not just ident.
Tom Lane [Mon, 14 Mar 2011 16:52:03 +0000 (12:52 -0400)]
Adjust regression test to avoid platform-dependent failure.
We have a test that verifies that max(anyarray) will cope if the array
column elements aren't all the same array type. However, it's now possible
for that to produce a collation-related error message instead of the
expected one, if the first two column elements happen to be of the same
type and it's one that expects to be given collation info. Tweak the test
to ensure this doesn't happen. Per buildfarm member pika.
Tom Lane [Sat, 12 Mar 2011 21:30:36 +0000 (16:30 -0500)]
Make all comparisons done for/with statistics use the default collation.
While this will give wrong answers when estimating selectivity for a
comparison operator that's using a non-default collation, the estimation
error probably won't be large; and anyway the former approach created
estimation errors of its own by trying to use a histogram that might have
been computed with some other collation. So we'll adopt this simplified
approach for now and perhaps improve it sometime in the future.
This patch incorporates changes from Andres Freund to make sure that
selfuncs.c passes a valid collation OID to any datatype-specific function
it calls, in case that function wants collation information. Said OID will
now always be DEFAULT_COLLATION_OID, but at least we won't get errors.
Tom Lane [Fri, 11 Mar 2011 23:18:55 +0000 (18:18 -0500)]
Put in some more safeguards against executing a division-by-zero.
Add dummy returns before every potential division-by-zero in int8.c,
because apparently further "improvements" in gcc's optimizer have
enabled it to break functions that weren't broken before.
Tom Lane [Fri, 11 Mar 2011 21:27:51 +0000 (16:27 -0500)]
Split CollateClause into separate raw and analyzed node types.
CollateClause is now used only in raw grammar output, and CollateExpr after
parse analysis. This is for clarity and to avoid carrying collation names
in post-analysis parse trees: that's both wasteful and possibly misleading,
since the collation's name could be changed while the parsetree still
exists.
Also, clean up assorted infelicities and omissions in processing of the
node type.
Tom Lane [Fri, 11 Mar 2011 18:20:11 +0000 (13:20 -0500)]
Create an explicit concept of collations that work for any encoding.
Use collencoding = -1 to represent such a collation in pg_collation.
We need this to make the "default" entry work sanely, and a later
patch will fix the C/POSIX entries to be represented this way instead
of duplicating them across all encodings. All lookup operations now
search first for an entry that's database-encoding-specific, and then
for the same name with collencoding = -1.
Also some incidental code cleanup in collationcmds.c and pg_collation.c.
In ecpg preprocessor, don't try to look up constants in the test for
variable hiding. A constant is not a variable. It worked in most cases by
accident, because we add constants to the global list of variables (why?),
but float constants like 1.23 were interpreted as struct field references,
and not found.
Backpatch to 9.0, where the test for variable hiding was added.
Bruce Momjian [Fri, 11 Mar 2011 15:23:59 +0000 (10:23 -0500)]
Adds index entries for session_user and pg_describe_object
Removes extraneous closing parenthesis from pg_describe_object
Puts pg_describe_object and has_sequence_privilege in correct
alphabetical position in function listing
Tom Lane [Thu, 10 Mar 2011 22:30:18 +0000 (17:30 -0500)]
Revert addition of third argument to format_type().
Including collation in the behavior of that function promotes a world view
we do not want. Moreover, it was producing the wrong behavior for pg_dump
anyway: what we want is to dump a COLLATE clause on attributes whose
attcollation is different from the underlying type, and likewise for
domains, and the function cannot do that for us. Doing it the hard way
in pg_dump is a bit more tedious but produces more correct output.
In passing, fix initdb so that the initial entry in pg_collation is
properly pinned. It was droppable before :-(
Robert Haas [Thu, 10 Mar 2011 20:38:39 +0000 (15:38 -0500)]
More synchronous replication tweaks.
SyncRepRequested() must check not only the value of the
synchronous_replication GUC but also whether max_wal_senders > 0.
Otherwise, we might end up waiting for sync rep even when there's no
possibility of a standby ever managing to connect. There are some
existing cross-checks to prevent this, but they're not quite sufficient:
the user can start the server with max_wal_senders=0,
synchronous_standby_names='', and synchronous_replication=off and then
subsequent make synchronous_standby_names not empty using pg_ctl reload,
and then SET synchronous_standby=on, leading to an indefinite hang.
Along the way, rename the global variable for the synchronous_replication
GUC to match the name of the GUC itself, for clarity.
Report by Fujii Masao, though I didn't use his patch.