Michael Paquier [Sun, 23 Dec 2018 07:42:22 +0000 (16:42 +0900)]
Disable WAL-skipping optimization for COPY on views and foreign tables
COPY can skip writing WAL when loading data on a table which has been
created in the same transaction as the one loading the data, however
this cannot work on views or foreign table as this would result in
trying to flush relation files which do not exist. So disable the
optimization so as commands are able to work the same way with any
configuration of wal_level.
Tests are added to cover the different cases, which need to have
wal_level set to minimal to allow the problem to show up, and that is
not the default configuration.
Reported-by: Luis M. Carril, Etsuro Fujita
Author: Amit Langote, Michael Paquier Reviewed-by: Etsuro Fujita
Discussion: https://postgr.es/m/15552-c64aa14c5c22f63c@postgresql.org
Backpatch-through: 10, where support for COPY on views has been added,
while v11 has added support for COPY on foreign tables.
Peter Eisentraut [Sat, 22 Dec 2018 05:53:37 +0000 (06:53 +0100)]
Add WRITE_*_ARRAY macros
Add WRITE_ATTRNUMBER_ARRAY, WRITE_OID_ARRAY, WRITE_INT_ARRAY,
WRITE_BOOL_ARRAY macros to outfuncs.c, mirroring the existing
READ_*_ARRAY macros in readfuncs.c.
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://www.postgresql.org/message-id/flat/8f2ebc67-e75f-9478-f5a5-bbbf090b1f8d%402ndquadrant.com
Check for conflicting queries during replay of gistvacuumpage()
013ebc0a7b implements so-called GiST microvacuum. That is gistgettuple() marks
index tuples as dead when kill_prior_tuple is set. Later, when new tuple
insertion claims page space, those dead index tuples are physically deleted
from page. When this deletion is replayed on standby, it might conflict with
read-only queries. But 013ebc0a7b doesn't handle this. That may lead to
disappearance of some tuples from read-only snapshots on standby.
This commit implements resolving of conflicts between replay of GiST microvacuum
and standby queries. On the master we implement new WAL record type
XLOG_GIST_DELETE, which comprises necessary information. On stable releases
we've to be tricky to keep WAL compatibility. Information required for conflict
processing is just appended to data of XLOG_GIST_PAGE_UPDATE record. So,
PostgreSQL version, which doesn't know about conflict processing, will just
ignore that.
Reported-by: Andres Freund Diagnosed-by: Andres Freund
Discussion: https://postgr.es/m/20181212224524.scafnlyjindmrbe6%40alap3.anarazel.de
Author: Alexander Korotkov
Backpatch-through: 9.6
Tom Lane [Thu, 20 Dec 2018 21:21:51 +0000 (16:21 -0500)]
Base information_schema.sql_identifier domain on name, not varchar.
The SQL spec says that sql_identifier is a domain over varchar,
but it also says that that domain is supposed to represent the set
of valid identifiers for the implementation, in particular applying
a length limit matching the implementation's identifier length limit.
We were declaring sql_identifier as just "character varying", thus
duplicating what the spec says about base type, but entirely failing
at the rest of it.
Instead, let's declare sql_identifier as a domain over type "name".
(We can drop the COLLATE "C" added by commit 6b0faf723, since that's
now implicit in "name".) With the recent improvements to name's
comparison support, there's not a lot of functional difference between
name and varchar. So although in principle this is a spec deviation,
it's a pretty minor one. And correctly enforcing PG's name length limit
is a good thing; on balance this seems closer to the intent of the spec
than what we had.
But that's all just language-lawyering. The *real* reason to do this is
that it makes sql_identifier columns exposed by information_schema views
be just direct representations of the underlying "name" catalog columns,
eliminating a semantic mismatch that was disastrous for performance of
typical queries on the information_schema. In combination with the
recent change to allow dropping no-op CoerceToDomain nodes, this allows
(for example) queries such as
select ... from information_schema.tables where table_name = 'foo';
to produce an indexscan rather than a seqscan on pg_class.
Tom Lane [Thu, 20 Dec 2018 20:58:58 +0000 (15:58 -0500)]
Avoid producing over-length specific_name outputs in information_schema.
information_schema output columns that are declared as being type
sql_identifier are supposed to conform to the implementation's rules
for valid identifiers, in particular the identifier length limit.
Several places potentially violated this limit by concatenating a
function's name and OID. (The OID is added to ensure name uniqueness
within a schema, since the spec doesn't expect function name overloading.)
Simply truncating the concatenation result to fit in "name" won't do,
since losing part of the OID might wind up giving non-unique results.
Instead, let's truncate the function name as necessary.
The most practical way to do that is to do it in a C function; the
information_schema.sql script doesn't have easy access to the value
of NAMEDATALEN, nor does it have an easy way to truncate on the basis
of resulting byte-length rather than number of characters.
(There are still a couple of places that cast concatenation results to
sql_identifier, but as far as I can see they are guaranteed not to produce
over-length strings, at least with the normal value of NAMEDATALEN.)
Alvaro Herrera [Thu, 20 Dec 2018 19:42:13 +0000 (16:42 -0300)]
Fix lock level used for partition when detaching it
For probably bogus reasons, we acquire only AccessShareLock on the
partition when we try to detach it from its parent partitioned table.
This can cause ugly things to happen if another transaction is doing
any sort of DDL to the partition concurrently.
Upgrade that lock to ShareUpdateExclusiveLock, which per discussion
seems to be the minimum needed.
Tom Lane [Thu, 20 Dec 2018 18:55:11 +0000 (13:55 -0500)]
Doc: fix ancient mistake in search_path documentation.
"$user" in a search_path string is replaced by CURRENT_USER not
SESSION_USER. (It actually was SESSION_USER in the initial implementation,
but we changed it shortly later, and evidently forgot to fix the docs to
match.)
Tom Lane [Thu, 20 Dec 2018 17:19:07 +0000 (12:19 -0500)]
Make bitmapset.c use 64-bit bitmap words on 64-bit machines.
Using the full width of the CPU's native word size shouldn't cost
anything in typical cases. When working with large bitmapsets,
this halves the number of operations needed for many common BMS
operations. On the right sort of test case, a measurable improvement
is obtained.
Alvaro Herrera [Thu, 20 Dec 2018 13:58:22 +0000 (10:58 -0300)]
DETACH PARTITION: hold locks on indexes until end of transaction
When a partition is detached from its parent, we acquire locks on all
attached indexes to also detach them ... but we release those locks
immediately. This is a violation of the policy of keeping locks on user
objects to the end of the transaction. Bug introduced in 8b08f7d4820f.
It's unclear that there are any ill effects possible, but it's clearly
wrong nonetheless. It's likely that bad behavior *is* possible, but
mostly because the relation that the index is for is only locked with
AccessShareLock, which is an older bug that shall be fixed separately.
While touching that line of code, close the index opened with
index_open() using index_close() instead of relation_close().
No difference in practice, but let's be consistent.
Michael Paquier [Thu, 20 Dec 2018 05:24:42 +0000 (14:24 +0900)]
Add more tab completion for CREATE TABLE in psql
The following completion patterns are added:
- CREATE TABLE <name> with '(', OF or PARTITION OF.
- CREATE TABLE <name> OF with list of composite types.
- CREATE TABLE name (...) with PARTITION OF, WITH, TABLESPACE, ON
COMMIT (depending on the presence of a temporary table).
- CREATE TABLE ON COMMIT with actions (only for temporary tables).
Author: Dagfinn Ilmari Mannsåker
Discussion: https://postgr.es/m/d8j1s77kdbb.fsf@dalvik.ping.uio.no
Greg Stark [Wed, 19 Dec 2018 23:28:35 +0000 (18:28 -0500)]
Fix ADD IF NOT EXISTS used in conjunction with ALTER TABLE ONLY
The flag for IF NOT EXISTS was only being passed down in the normal
recursing case. It's been this way since originally added in 9.6 in
commit 2cd40adb85 so backpatch back to 9.6.
Tom Lane [Wed, 19 Dec 2018 22:46:07 +0000 (17:46 -0500)]
Add text-vs-name cross-type operators, and unify name_ops with text_ops.
Now that name comparison has effectively the same behavior as text
comparison, we might as well merge the name_ops opfamily into text_ops,
allowing cross-type comparisons to be processed without forcing a
datatype coercion first. We need do little more than add cross-type
operators to make the opfamily complete, and fix one or two places
in the planner that assumed text_ops was a single-datatype opfamily.
I chose to unify hash name_ops into hash text_ops as well, since the
types have compatible hashing semantics. This allows marking the
new cross-type equality operators as oprcanhash.
(Note: this doesn't remove the name_ops opclasses, so there's no
breakage of index definitions. Those opclasses are just reparented
into the text_ops opfamily.)
Tom Lane [Wed, 19 Dec 2018 22:35:12 +0000 (17:35 -0500)]
Make type "name" collation-aware.
The "name" comparison operators now all support collations, making them
functionally equivalent to "text" comparisons, except for the different
physical representation of the datatype. They do, in fact, mostly share
the varstr_cmp and varstr_sortsupport infrastructure, which has been
slightly enlarged to handle the case.
To avoid changes in the default behavior of the datatype, set name's
typcollation to C_COLLATION_OID not DEFAULT_COLLATION_OID, so that
by default comparisons to a name value will continue to use strcmp
semantics. (This would have been the case for system catalog columns
anyway, because of commit 6b0faf723, but doing this makes it true for
user-created name columns as well. In particular, this avoids
locale-dependent changes in our regression test results.)
In consequence, tweak a couple of places that made assumptions about
collatable base types always having typcollation DEFAULT_COLLATION_OID.
I have not, however, attempted to relax the restriction that user-
defined collatable types must have that. Hence, "name" doesn't
behave quite like a user-defined type; it acts more like a domain
with COLLATE "C". (Conceivably, if we ever get rid of the need for
catalog name columns to be fixed-length, "name" could actually become
such a domain over text. But that'd be a pretty massive undertaking,
and I'm not volunteering.)
Tom Lane [Wed, 19 Dec 2018 16:41:36 +0000 (11:41 -0500)]
Small improvements for allocation logic in ginHeapTupleFastCollect().
Avoid repetitive calls to repalloc() when the required size of the
collector array grows more than 2x in one call. Also ensure that the
array size is a power of 2 (since palloc will probably consume a power
of 2 anyway) and doesn't start out very small (which'd likely just lead
to extra repallocs).
Peter Geoghegan [Wed, 19 Dec 2018 05:40:38 +0000 (21:40 -0800)]
Remove obsolete nbtree duplicate entries comment.
Remove a comment from the Berkeley days claiming that nbtree must
disambiguate duplicate keys within _bt_moveright(). There is no special
care taken around duplicates within _bt_moveright(), at least since
commit 9e85183bfc3 removed inscrutable _bt_moveright() code to handle
pages full of duplicates.
Peter Geoghegan [Wed, 19 Dec 2018 00:59:50 +0000 (16:59 -0800)]
Correct obsolete nbtree recovery comments.
Commit 40dae7ec537, which made the handling of interrupted nbtree page
splits more robust, removed an nbtree-specific end-of-recovery cleanup
step. This meant that it was no longer possible to complete an
interrupted page split during recovery. However, a reference to
recovery as a reason for using a NULL stack while inserting into a
parent page was missed. Remove the reference.
Remove a similar obsolete reference to recovery that was introduced much
more recently, as part of the btree fastpath optimization enhancement
that made it into Postgres 11 (commit 2b272734, and follow-up commits).
Backpatch: 11-, where the fastpath optimization was introduced.
Tom Lane [Tue, 18 Dec 2018 17:48:15 +0000 (12:48 -0500)]
Make collation-aware system catalog columns use "C" collation.
Up to now we allowed text columns in system catalogs to use collation
"default", but that isn't really safe because it might mean something
different in template0 than it means in a database cloned from template0.
In particular, this could mean that cloned pg_statistic entries for such
columns weren't entirely valid, possibly leading to bogus planner
estimates, though (probably) not any outright failures.
In the wake of commit 5e0928005, a better solution is available: if we
label such columns with "C" collation, then their pg_statistic entries
will also use that collation and hence will be valid independently of
the database collation.
This also provides a cleaner solution for indexes on such columns than
the hack added by commit 0b28ea79c: the indexes will naturally inherit
"C" collation and don't have to be forced to use text_pattern_ops.
Also, with the planned improvement of type "name" to be collation-aware,
this policy will apply cleanly to both text and name columns.
Because of the pg_statistic angle, we should also apply this policy
to the tables in information_schema. This patch does that by adjusting
information_schema's textual domain types to specify "C" collation.
That has the user-visible effect that order-sensitive comparisons to
textual information_schema view columns will now use "C" collation
by default. The SQL standard says that the collation of those view
columns is implementation-defined, so I think this is legal per spec.
At some point this might allow for translation of such comparisons
into indexable conditions on the underlying "name" columns, although
additional work will be needed before that can happen.
Tom Lane [Tue, 18 Dec 2018 16:19:38 +0000 (11:19 -0500)]
Fix ancient thinko in mergejoin cost estimation.
"rescanratio" was computed as 1 + rescanned-tuples / total-inner-tuples,
which is sensible if it's to be multiplied by total-inner-tuples or a cost
value corresponding to scanning all the inner tuples. But in reality it
was (mostly) multiplied by inner_rows or a related cost, numbers that take
into account the possibility of stopping short of scanning the whole inner
relation thanks to a limited key range in the outer relation. This'd
still make sense if we could expect that stopping short would result in a
proportional decrease in the number of tuples that have to be rescanned.
It does not, however. The argument that establishes the validity of our
estimate for that number is independent of whether we scan all of the inner
relation or stop short, and experimentation also shows that stopping short
doesn't reduce the number of rescanned tuples. So the correct calculation
is 1 + rescanned-tuples / inner_rows, and we should be sure to multiply
that by inner_rows or a corresponding cost value.
Most of the time this doesn't make much difference, but if we have
both a high rescan rate (due to lots of duplicate values) and an outer
key range much smaller than the inner key range, then the error can
be significant, leading to a large underestimate of the cost associated
with rescanning.
Per report from Vijaykumar Jain. This thinko appears to go all the way
back to the introduction of the rescan estimation logic in commit 70fba7043, so back-patch to all supported branches.
Michael Paquier [Tue, 18 Dec 2018 07:37:51 +0000 (16:37 +0900)]
Include partitioned indexes to system view pg_indexes
pg_tables already includes partitioned tables, so for consistency
pg_indexes should show partitioned indexes.
Author: Suraj Kharage Reviewed-by: Amit Langote, Michael Paquier
Discussion: https://postgr.es/m/CAF1DzPVrYo4XNTEnc=PqVp6aLJc7LFYpYR4rX=_5pV=wJ2KdZg@mail.gmail.com
Michael Paquier [Tue, 18 Dec 2018 00:28:16 +0000 (09:28 +0900)]
Include ALTER INDEX SET STATISTICS in pg_dump
The new grammar pattern of ALTER INDEX SET STATISTICS able to use column
numbers on top of the existing column names introduced by commit 5b6d13e
forgot to add support for the feature in pg_dump, so defining statistics
on index columns was missing from the dumps, potentially causing silent
planning problems with a subsequent restore.
pg_dump ought to not use column names in what it generates as these are
automatically generated by the server and could conflict with real
relation attributes with matching patterns. "expr" and "exprN", N
incremented automatically after the creation of the first one, are used
as default attribute names for index expressions, and that could easily
match what is defined in other relations, causing the dumps to fail if
some of those attributes are renamed at some point. So to avoid any
problems, the new grammar with column numbers gets used.
Reported-by: Ronan Dunklau
Author: Michael Paquier Reviewed-by: Tom Lane, Adrien Nayrat, Amul Sul
Discussion: https://postgr.es/m/CAARsnT3UQ4V=yDNW468w8RqHfYiY9mpn2r_c5UkBJ97NAApUEw@mail.gmail.com
Backpatch-through: 11, where the new syntax has been introduced.
Tom Lane [Mon, 17 Dec 2018 18:50:07 +0000 (13:50 -0500)]
Drop support for getting signal descriptions from sys_siglist[].
It appears that all platforms that have sys_siglist[] also have
strsignal(), making that fallback case in pg_strsignal() dead code.
Getting rid of it allows dropping a configure test, which seems worth
more than providing textual signal descriptions on whatever platforms
might still hypothetically have use for the fallback case.
Alvaro Herrera [Mon, 17 Dec 2018 18:37:40 +0000 (15:37 -0300)]
Fix tablespace handling for partitioned tables
When partitioned tables were introduced, we failed to realize that by
copying the tablespace handling for other relation kinds with no
physical storage we were causing the secondary effect that their
partitions would not automatically inherit the tablespace setting. This
is surprising and unhelpful, so change it to adopt the behavior
introduced in pg11 (commit 33e6c34c3267) for partitioned indexes: the
parent relation remembers the tablespace specification, which is then
used for any new partitions that don't declare one.
Because this commit changes behavior of the TABLESPACE clause for
partitioned tables (it's no longer a no-op), it is not backpatched.
Author: David Rowley, Álvaro Herrera Reviewed-by: Michael Paquier
Discussion: https://postgr.es/m/CAKJS1f9SxVzqDrGD1teosFd6jBMM0UEaa14_8mRvcWE19Tu0hA@mail.gmail.com
Amit Kapila [Mon, 17 Dec 2018 08:37:27 +0000 (14:07 +0530)]
Remove extra semicolons.
Reported-by: David Rowley
Author: David Rowley Reviewed-by: Amit Kapila
Backpatch-through: 10
Discussion: https://postgr.es/m/CAKJS1f8EneeYyzzvdjahVZ6gbAHFkHbSFB5m_C0Y6TUJs9Dgdg@mail.gmail.com
Michael Paquier [Mon, 17 Dec 2018 01:34:44 +0000 (10:34 +0900)]
Make constraint rename issue relcache invalidation on target relation
When a constraint gets renamed, it may have associated with it a target
relation (for example domain constraints don't have one). Not
invalidating the target relation cache when issuing the renaming can
result in issues with subsequent commands that refer to the old
constraint name using the relation cache, causing various failures. One
pattern spotted was using CREATE TABLE LIKE after a constraint
renaming.
Reported-by: Stuart <sfbarbee@gmail.com>
Author: Amit Langote Reviewed-by: Michael Paquier
Discussion: https://postgr.es/m/2047094.V130LYfLq4@station53.ousa.org
Tom Lane [Mon, 17 Dec 2018 00:38:57 +0000 (19:38 -0500)]
Modernize our code for looking up descriptive strings for Unix signals.
At least as far back as the 2008 spec, POSIX has defined strsignal(3)
for looking up descriptive strings for signal numbers. We hadn't gotten
the word though, and were still using the crufty old sys_siglist array,
which is in no standard even though most Unixen provide it.
Aside from not being formally standards-compliant, this was just plain
ugly because it involved #ifdef's at every place using the code.
To eliminate the #ifdef's, create a portability function pg_strsignal,
which wraps strsignal(3) if available and otherwise falls back to
sys_siglist[] if available. The set of Unixen with neither API is
probably empty these days, but on any platform with neither, you'll
just get "unrecognized signal". All extant callers print the numeric
signal number too, so no need to work harder than that.
Along the way, upgrade pg_basebackup's child-error-exit reporting
to match the rest of the system.
Tom Lane [Sun, 16 Dec 2018 19:51:47 +0000 (14:51 -0500)]
Make error handling in parallel pg_upgrade less bogus.
reap_child() basically ignored the possibility of either an error in
waitpid() itself or a child process failure on signal. We don't really
need to do more than report and crash hard, but proceeding as though
nothing is wrong is definitely Not Acceptable. The error report for
nonzero child exit status was pretty off-point, as well.
Noted while fooling around with child-process failure detection
logic elsewhere. It's been like this a long time, so back-patch to
all supported branches.
Tom Lane [Sun, 16 Dec 2018 19:32:14 +0000 (14:32 -0500)]
Improve detection of child-process SIGPIPE failures.
Commit ffa4cbd62 added logic to detect SIGPIPE failure of a COPY child
process, but it only worked correctly if the SIGPIPE occurred in the
immediate child process. Depending on the shell in use and the
complexity of the shell command string, we might instead get back
an exit code of 128 + SIGPIPE, representing a shell error exit
reporting SIGPIPE in the child process.
We could just hack up ClosePipeToProgram() to add the extra case,
but it seems like this is a fairly general issue deserving a more
general and better-documented solution. I chose to add a couple
of functions in src/common/wait_error.c, which is a natural place
to know about wait-result encodings, that will test for either a
specific child-process signal type or any child-process signal failure.
Then, adjust other places that were doing ad-hoc tests of this type
to use the common functions.
In RestoreArchivedFile, this fixes a race condition affecting whether
the process will report an error or just silently proc_exit(1): before,
that depended on whether the intermediate shell got SIGTERM'd itself
or reported a child process failing on SIGTERM.
Like the previous patch, back-patch to v10; we could go further
but there seems no real need to.
Tom Lane [Fri, 14 Dec 2018 17:52:49 +0000 (12:52 -0500)]
Make pg_statistic and related code account more honestly for collations.
When we first put in collations support, we basically punted on teaching
pg_statistic, ANALYZE, and the planner selectivity functions about that.
They've just used DEFAULT_COLLATION_OID independently of the actual
collation of the data. It's time to improve that, so:
* Add columns to pg_statistic that record the specific collation associated
with each statistics slot.
* Teach ANALYZE to use the column's actual collation when comparing values
for statistical purposes, and record this in the appropriate slot. (Note
that type-specific typanalyze functions are now expected to fill
stats->stacoll with the appropriate collation, too.)
* Teach assorted selectivity functions to use the actual collation of
the stats they are looking at, instead of just assuming it's
DEFAULT_COLLATION_OID.
This should give noticeably better results in selectivity estimates for
columns with nondefault collations, at least for query clauses that use
that same collation (which would be the default behavior in most cases).
It's still true that comparisons with explicit COLLATE clauses different
from the stored data's collation won't be well-estimated, but that's no
worse than before. Also, this patch does make the first step towards
doing better with that, which is that it's now theoretically possible to
collect stats for a collation other than the column's own collation.
Patch by me; thanks to Peter Eisentraut for review.
Michael Paquier [Thu, 13 Dec 2018 23:59:35 +0000 (08:59 +0900)]
Introduce new extended routines for FDW and foreign server lookups
The cache lookup routines for foreign-data wrappers and foreign servers
are extended with an extra argument to handle a set of flags. The only
value which can be used now is to indicate if a missing object should
result in an error or not, and are designed to be extensible on need.
Those new routines are added into the existing set of user-visible
FDW APIs and documented in consequence. They will be used for future
patches to improve the SQL interface for object addresses.
Author: Michael Paquier Reviewed-by: Álvaro Herrera
Discussion: https://postgr.es/m/CAB7nPqSZxrSmdHK-rny7z8mi=EAFXJ5J-0RbzDw6aus=wB5azQ@mail.gmail.com
Andres Freund [Thu, 13 Dec 2018 22:50:57 +0000 (14:50 -0800)]
Create a separate oid range for oids assigned by genbki.pl.
The changes I made in 578b229718e assigned oids below
FirstBootstrapObjectId to objects in include/catalog/*.dat files that
did not have an oid assigned, starting at the max oid explicitly
assigned. Tom criticized that for mainly two reasons:
1) It's not clear which values are manually and which explicitly
assigned.
2) The space below FirstBootstrapObjectId gets pretty crowded, and
some PostgreSQL forks have used oids >= 9000 for their own objects,
to avoid conflicting.
Thus create a new range for objects not assigned explicit oids, but
assigned by genbki.pl. For now 1-9999 is for explicitly assigned oids,
FirstGenbkiObjectId (10000) to FirstBootstrapObjectId (1200) -1 is for
genbki.pl assigned oids, and < FirstNormalObjectId (16384) is for oids
assigned during bootstrap. It's possible that we'll have to adjust
these boundaries, but there's some headroom for now.
Add a note suggesting that oids in forks should be assigned in the
9000-9999 range.
Catversion bump for obvious reasons.
Per complaint from Tom Lane.
Author: Andres Freund
Discussion: https://postgr.es/m/16845.1544393682@sss.pgh.pa.us
Tom Lane [Thu, 13 Dec 2018 20:11:09 +0000 (15:11 -0500)]
Fix bogus logic for skipping unnecessary partcollation dependencies.
The idea here is to not call recordDependencyOn for the default collation,
since we know that's pinned. But what the code actually did was to record
the partition key's dependency on the opclass twice, instead.
Evidently introduced by sloppy coding in commit 2186b608b. Back-patch
to v10 where that came in.
Tom Lane [Thu, 13 Dec 2018 18:24:43 +0000 (13:24 -0500)]
Drop no-op CoerceToDomain nodes from expressions at planning time.
If a domain has no constraints, then CoerceToDomain doesn't really do
anything and can be simplified to a RelabelType. This not only
eliminates cycles at execution, but allows the planner to optimize better
(for instance, match the coerced expression to an index on the underlying
column). However, we do have to support invalidating the plan later if
a constraint gets added to the domain. That's comparable to the case of
a change to a SQL function that had been inlined into a plan, so all the
necessary logic already exists for plans depending on functions. We
need only duplicate or share that logic for domains.
ALTER DOMAIN ADD/DROP CONSTRAINT need to be taught to send out sinval
messages for the domain's pg_type entry, since those operations don't
update that row. (ALTER DOMAIN SET/DROP NOT NULL do update that row,
so no code change is needed for them.)
Testing this revealed what's really a pre-existing bug in plpgsql:
it caches the SQL-expression-tree expansion of type coercions and
had no provision for invalidating entries in that cache. Up to now
that was only a problem if such an expression had inlined a SQL
function that got changed, which is unlikely though not impossible.
But failing to track changes of domain constraints breaks an existing
regression test case and would likely cause practical problems too.
We could fix that locally in plpgsql, but what seems like a better
idea is to build some generic infrastructure in plancache.c to store
standalone expressions and track invalidation events for them.
(It's tempting to wonder whether plpgsql's "simple expression" stuff
could use this code with lower overhead than its current use of the
heavyweight plancache APIs. But I've left that idea for later.)
Other stuff fixed in passing:
* Allow estimate_expression_value() to drop CoerceToDomain
unconditionally, effectively assuming that the coercion will succeed.
This will improve planner selectivity estimates for cases involving
estimatable expressions that are coerced to domains. We could have
done this independently of everything else here, but there wasn't
previously any need for eval_const_expressions_mutator to know about
CoerceToDomain at all.
* Use a dlist for plancache.c's list of cached plans, rather than a
manually threaded singly-linked list. That eliminates a potential
performance problem in DropCachedPlan.
* Fix a couple of inconsistencies in typecmds.c about whether
operations on domains drop RowExclusiveLock on pg_type. Our common
practice is that DDL operations do drop catalog locks, so standardize
on that choice.
Prevent GIN deleted pages from being reclaimed too early
When GIN vacuum deletes a posting tree page, it assumes that no concurrent
searchers can access it, thanks to ginStepRight() locking two pages at once.
However, since 9.4 searches can skip parts of posting trees descending from the
root. That leads to the risk that page is deleted and reclaimed before
concurrent search can access it.
This commit prevents the risk of above by waiting for every transaction, which
might wait to reference this page, to finish. Due to binary compatibility
we can't change GinPageOpaqueData to store corresponding transaction id.
Instead we reuse page header pd_prune_xid field, which is unused in index pages.
Discussion: https://postgr.es/m/31a702a.14dd.166c1366ac1.Coremail.chjischj%40163.com
Author: Andrey Borodin, Alexander Korotkov Reviewed-by: Alexander Korotkov
Backpatch-through: 9.4
On standby ginRedoDeletePage() can work concurrently with read-only queries.
Those queries can traverse posting tree in two ways.
1) Using rightlinks by ginStepRight(), which locks the next page before
unlocking its left sibling.
2) Using downlinks by ginFindLeafPage(), which locks at most one page at time.
Original lock order was: page, parent, left sibling. That lock order can
deadlock with ginStepRight(). In order to prevent deadlock this commit changes
lock order to: left sibling, page, parent. Note, that position of parent in
locking order seems insignificant, because we only lock one page at time while
traversing downlinks.
Reported-by: Chen Huajun Diagnosed-by: Chen Huajun, Peter Geoghegan, Andrey Borodin
Discussion: https://postgr.es/m/31a702a.14dd.166c1366ac1.Coremail.chjischj%40163.com
Author: Alexander Korotkov
Backpatch-through: 9.4
Fix deadlock in GIN vacuum introduced by 218f51584d5
Before 218f51584d5 if posting tree page is about to be deleted, then the whole
posting tree is locked by LockBufferForCleanup() on root preventing all the
concurrent inserts. 218f51584d5 reduced locking to the subtree containing
page to be deleted. However, due to concurrent parent split, inserter doesn't
always holds pins on all the pages constituting path from root to the target
leaf page. That could cause a deadlock between GIN vacuum process and GIN
inserter. And we didn't find non-invasive way to fix this.
This commit reverts VACUUM behavior to lock the whole posting tree before
delete any page. However, we keep another useful change by 218f51584d5: the
tree is locked only if there are pages to be deleted.
Reported-by: Chen Huajun Diagnosed-by: Chen Huajun, Andrey Borodin, Peter Geoghegan
Discussion: https://postgr.es/m/31a702a.14dd.166c1366ac1.Coremail.chjischj%40163.com
Author: Alexander Korotkov, based on ideas from Andrey Borodin and Peter Geoghegan Reviewed-by: Andrey Borodin
Backpatch-through: 10
Tom Lane [Wed, 12 Dec 2018 21:08:30 +0000 (16:08 -0500)]
Repair bogus EPQ plans generated for postgres_fdw foreign joins.
postgres_fdw's postgresGetForeignPlan() assumes without checking that the
outer_plan it's given for a join relation must have a NestLoop, MergeJoin,
or HashJoin node at the top. That's been wrong at least since commit 4bbf6edfb (which could cause insertion of a Sort node on top) and it seems
like a pretty unsafe thing to Just Assume even without that.
Through blind good fortune, this doesn't seem to have any worse
consequences today than strange EXPLAIN output, but it's clearly trouble
waiting to happen.
To fix, test the node type explicitly before touching Join-specific
fields, and avoid jamming the new tlist into a node type that can't
do projection. Export a new support function from createplan.c
to avoid building low-level knowledge about the latter into FDWs.
Back-patch to 9.6 where the faulty coding was added. Note that the
associated regression test cases don't show any changes before v11,
apparently because the tests back-patched with 4bbf6edfb don't actually
exercise the problem case before then (there's no top-level Sort
in those plans).
Tom Lane [Wed, 12 Dec 2018 18:49:41 +0000 (13:49 -0500)]
Repair bogus handling of multi-assignment Params in upper plan levels.
Our support for multiple-set-clauses in UPDATE assumes that the Params
referencing a MULTIEXPR_SUBLINK SubPlan will appear before that SubPlan
in the targetlist of the plan node that calculates the updated row.
(Yeah, it's a hack...) In some PG branches it's possible that a Result
node gets inserted between the primary calculation of the update tlist
and the ModifyTable node. setrefs.c did the wrong thing in this case
and left the upper-level Params as Params, causing a crash at runtime.
What it should do is replace them with "outer" Vars referencing the child
plan node's output. That's a result of careless ordering of operations
in fix_upper_expr_mutator, so we can fix it just by reordering the code.
Fix fix_join_expr_mutator similarly for consistency, even though join
nodes could never appear in such a context. (In general, it seems
likely to be a bit cheaper to use Vars than Params in such situations
anyway, so this patch might offer a tiny performance improvement.)
The hazard extends back to 9.5 where the MULTIEXPR_SUBLINK stuff
was introduced, so back-patch that far. However, this may be a live
bug only in 9.6.x and 10.x, as the other branches don't seem to want
to calculate the final tlist below the Result node. (That plan shape
change between branches might be a mini-bug in itself, but I'm not
really interested in digging into the reasons for that right now.
Still, add a regression test memorializing what we expect there,
so we'll notice if it changes again.)
Michael Paquier [Wed, 12 Dec 2018 00:49:39 +0000 (09:49 +0900)]
Tweak pg_partition_tree for undefined relations and unsupported relkinds
This fixes a crash which happened when calling the function directly
with a relation OID referring to a non-existing object, and changes the
behavior so as NULL is returned for unsupported relkinds instead of
generating an error. This puts the new function in line with many other
system functions, and eases actions like full scans of pg_class.
Author: Michael Paquier Reviewed-by: Amit Langote, Stephen Frost
Discussion: https://postgr.es/m/20181207010406.GO2407@paquier.xyz
Tom Lane [Tue, 11 Dec 2018 16:48:00 +0000 (11:48 -0500)]
Fix test_rls_hooks to assign expression collations properly.
This module overlooked this necessary fixup step on the results of
transformWhereClause(). It accidentally worked anyway, because the
constructed expression involved type "name" which is not collatable,
but it fell over while I was experimenting with changing "name" to
be collatable.
Back-patch, not because there's any live bug here in back branches,
but because somebody might use this code as a model for some real
application and then not understand why it doesn't work.
Tom Lane [Tue, 11 Dec 2018 16:21:36 +0000 (11:21 -0500)]
Doc: improve documentation about ALTER LARGE OBJECT requirements.
Unlike other ALTER ref pages, this one neglected to mention that
ALTER OWNER requires being a member of the new owning role.
Per bug #15546 from Stefan Kadow.
Noah Misch [Tue, 11 Dec 2018 04:15:42 +0000 (20:15 -0800)]
Raise some timeouts to 180s, in test code.
Slow runs of buildfarm members chipmunk, hornet and mandrill saw the
shorter timeouts expire. The 180s timeout in poll_query_until has been
trouble-free since 2a0f89cd717ce6d49cdc47850577823682167e87 introduced
it two years ago, so use 180s more widely. Back-patch to 9.6, where the
first of these timeouts was introduced.
Stephen Frost [Mon, 10 Dec 2018 14:31:38 +0000 (09:31 -0500)]
Remove dead code in toast_fetch_datum_slice
In toast_fetch_datum_slice(), we Assert() that what is passed in isn't
compressed, but we then later had a check to see what the length of if
what was passed in is compressed. That later check is rather confusing
since toast_fetch_datum_slice() is only ever called with non-compressed
datums and the Assert() earlier makes it clear that one shouldn't be
passing in compressed datums.
Add a comment to make it clear that toast_fetch_datum_slice() is just
for non-compressed datums, and remove the dead code.
Michael Paquier [Mon, 10 Dec 2018 06:00:59 +0000 (15:00 +0900)]
Ensure cleanup of orphan archive status files
When a WAL segment is recycled, its ".ready" and ".done" status files
get also automatically removed, however this is not done in a durable
manner. Hence, in a subsequent crash, it could be possible that a
".ready" status file is still around with its corresponding segment
already gone.
If the backend reaches such a state, the archive command would most
likely complain about a segment non-existing and would keep retrying,
causing WAL segments to bloat pg_wal/, potentially making Postgres crash
hard when running out of space.
As status files are removed after each individual segment, using
durable_unlink() does not completely close the window either, as a crash
could happen between the moment the WAL segment is recycled and the
moment its status files are removed. This has also some performance
impact with the additional fsync() calls needed to make the removal in a
durable manner. Doing the cleanup at recovery is not cost-free either
as this makes crash recovery potentially take longer than necessary.
So, instead, as per an idea of Stephen Frost, make the archiver aware of
orphan status files and remove them on-the-fly if the corresponding
segment goes missing. Removal failures follow a model close to what
happens for WAL segments, where multiple attempts are done before giving
up temporarily, and where a successful orphan removal makes the archiver
move immediately to the next WAL segment thought as ready to be
archived.
Author: Michael Paquier Reviewed-by: Nathan Bossart, Andres Freund, Stephen Frost, Kyotaro
Horiguchi
Discussion: https://postgr.es/m/20180928032827.GF1500@paquier.xyz
Michael Paquier [Sun, 9 Dec 2018 07:35:06 +0000 (16:35 +0900)]
Add timestamp of last received message from standby to pg_stat_replication
The timestamp generated by the standby at message transmission has been
included in the protocol since its introduction for both the status
update message and hot standby feedback message, but it has never
appeared in pg_stat_replication. Seeing this timestamp does not matter
much with a cluster which has a lot of activity, but on a mostly-idle
cluster, this makes monitoring able to react faster than the configured
timeouts.
Author: MyungKyu LIM Reviewed-by: Michael Paquier, Masahiko Sawada
Discussion: https://postgr.es/m/1657809367.407321.1533027417725.JavaMail.jboss@ep2ml404
Tom Lane [Fri, 7 Dec 2018 21:40:58 +0000 (16:40 -0500)]
Doc: document that we expect CHECK constraint conditions to be immutable.
This restriction is implicit in the check-only-once implementation we use
for table and domain constraints, but it wasn't spelled out anywhere, nor
was there any advice about how to alter a constraint's behavior safely.
Improve that.
I was also dissatisfied with the documentation of ALTER DOMAIN VALIDATE
CONSTRAINT, which entirely failed to explain the use of that feature; and
thence decided that ALTER TABLE VALIDATE CONSTRAINT could be documented
better as well.
Perhaps we should back-patch this, along with the related commit 36d442a25,
but for now I refrained.
Tom Lane [Fri, 7 Dec 2018 18:11:30 +0000 (13:11 -0500)]
In PQprint(), write HTML table trailer before closing the output pipe.
This is an astonishingly ancient bit of silliness, dating AFAICS to
commit edb519b14 of 27-Jul-1996 which added the pipe close stanza in
the wrong place. It happens to be harmless given that the code above
this won't enable the pager if html3 output mode is selected. Still,
somebody might try to relax that restriction someday, and in any case
it could confuse readers and static analysis tools, so let's fix it in
HEAD.
Tom Lane [Fri, 7 Dec 2018 17:11:59 +0000 (12:11 -0500)]
Fix misapplication of pgstat_count_truncate to wrong relation.
The stanza of ExecuteTruncate[Guts] that truncates a target table's toast
relation re-used the loop local variable "rel" to reference the toast rel.
This was safe enough when written, but commit d42358efb added code below
that that supposed "rel" still pointed to the parent table. Therefore,
the stats counter update was applied to the wrong relcache entry (the
toast rel not the user rel); and if we were unlucky and that relcache
entry had been flushed during reindex_relation, very bad things could
ensue.
(I'm surprised that CLOBBER_CACHE_ALWAYS testing hasn't found this.
I'm even more surprised that the problem wasn't detected during the
development of d42358efb; it must not have been tested in any case
with a toast table, as the incorrect stats counts are very obvious.)
To fix, replace use of "rel" in that code branch with a more local
variable. Adjust test cases added by d42358efb so that some of them
use tables with toast tables.
Per bug #15540 from Pan Bian. Back-patch to 9.5 where d42358efb came in.
Tom Lane [Fri, 7 Dec 2018 16:02:39 +0000 (11:02 -0500)]
Clean up sloppy coding in publicationcmds.c's OpenTableList().
Remove dead code (which would be incorrect if it weren't dead),
per report from Pan Bian. Add a CHECK_FOR_INTERRUPTS in the
inner loop over child relations, because there's little point
in having one in the outer loop if there's not one here too.
Minor stylistic adjustments and comment improvements.
Seems to be aboriginal to this code (cf commit 665d1fad9).
Back-patch to v10 where that came in, not because any of this
is significant, but just to keep the branches looking similar.
Tom Lane [Thu, 6 Dec 2018 20:08:44 +0000 (15:08 -0500)]
Improve our response to invalid format strings, and detect more cases.
Places that are testing for *printf failure ought to include the format
string in their error reports, since bad-format-string is one of the
more likely causes of such failure. This both makes it easier to find
and repair the mistake, and provides at least some useful info to the
user who stumbles across such a problem.
Also, tighten snprintf.c to report EINVAL for an invalid flag or
final character in a format %-spec (including the case where the
%-spec is missing a final character altogether). This seems like
better project policy, and it also allows removing an instruction
or two from the hot code path.
Back-patch the error reporting change in pvsnprintf, since it should be
harmless and may be helpful; but not the snprintf.c change.
Per discussion of bug #15511 from Ertuğrul Kahveci, which reported an
invalid translated format string. These changes don't fix that error,
but they should improve matters next time we make such a mistake.
Stephen Frost [Thu, 6 Dec 2018 16:39:09 +0000 (11:39 -0500)]
Improve planner stats documentation
It was pointed out that in the planner stats documentation under
Extended Statistics, one of the sentences was a bit awkward. Improve
that by rewording it slightly.
Stephen Frost [Thu, 6 Dec 2018 16:11:21 +0000 (11:11 -0500)]
Cleanup minor pg_dump memory leaks
In dumputils, we may have successfully parsed the acls when we discover
that we can't parse the reverse ACLs and then return- check and free
aclitems if that happens.
In dumpTableSchema, move ftoptions and srvname under the relkind !=
RELKIND_VIEW branch (since they're only used there) and then check if
they've been allocated and, if so, free them at the end of that block.
Pointed out by Pavel Raiskup, though I didn't use those patches.
Stephen Frost [Thu, 6 Dec 2018 16:05:39 +0000 (11:05 -0500)]
Cleanup comments in xlog compression
Skipping over the "hole" in full page images in the XLOG code was
described as being a form of compression, but this got a bit confusing
since we now have PGLZ-based compression happening, so adjust the
wording to discuss "removing" the "hole" and keeping the talk about
compression to where we're talking about using PGLZ-based compression of
the full page images.
Tatsuo Ishii [Thu, 6 Dec 2018 03:15:15 +0000 (12:15 +0900)]
Change true/false to on/off.
We prefer to use on/off than true/false for boolean configuration
parameters in the documentation, but there were a few places where
true/false were still used.
Alvaro Herrera [Wed, 5 Dec 2018 16:31:51 +0000 (13:31 -0300)]
Don't mark partitioned indexes invalid unnecessarily
When an indexes is created on a partitioned table using ONLY (don't
recurse to partitions), it gets marked invalid until index partitions
are attached for each table partition. But there's no reason to do this
if there are no partitions ... and moreover, there's no way to get the
index to become valid afterwards, because all partitions that get
created/attached get their own index partition already attached to the
parent index, so there's no chance to do ALTER INDEX ... ATTACH PARTITION
that would make the parent index valid.
Fix by not marking the index as invalid to begin with.
This is very similar to 9139aa19423b, but the pg_dump aspect does not
appear to be relevant until we add FKs that can point to PKs on
partitioned tables. (I tried to cause the pg_upgrade test to break by
leaving some of these bogus tables around, but wasn't able to.)
Making this change means that an index that was supposed to be invalid
in the insert_conflict regression test is no longer invalid; reorder the
DDL so that the test continues to verify the behavior we want it to.
Author: Álvaro Herrera Reviewed-by: Amit Langote
Discussion: https://postgr.es/m/20181203225019.2vvdef2ybnkxt364@alvherre.pgsql
Etsuro Fujita [Tue, 4 Dec 2018 08:18:58 +0000 (17:18 +0900)]
postgres_fdw: Improve cost and size estimation for aggregate pushdown.
In commit 7012b132d07c2b4ea15b0b3cb1ea9f3278801d98, which added aggregate
pushdown to postgres_fdw, we didn't account for the evaluation cost and the
selectivity of HAVING quals attached to ForeignPaths performing aggregate
pushdown, as core had never accounted for that for AggPaths and GroupPaths.
And we didn't set these values of the locally-checked quals (ie, fpinfo's
local_conds_cost and local_conds_sel), which were initialized to zeros, but
since estimate_path_cost_size factors in these to estimate the result size
and the evaluation cost of such a ForeignPath when the use_remote_estimate
option is enabled, this caused it to produce underestimated results in that
case.
By commit 7b6c07547190f056b0464098bb5a2247129d7aa2 core was changed so that
it accounts for the evaluation cost and the selectivity of HAVING quals in
aggregation paths, so change the postgres_fdw's aggregate pushdown code as
well as such. This not only fixes the underestimation issue mentioned
above, but improves the estimation using local statistics in that function
when that option is disabled.
This would be a bug fix rather than an improvement, but apply it to HEAD
only to avoid destabilizing existing plan choices.
Tom Lane [Mon, 3 Dec 2018 16:40:49 +0000 (11:40 -0500)]
Refactor documentation about privileges to centralize the info.
Expand section 5.6 "Privileges" to include the full definition of
each privilege type, and an explanation of aclitem privilege displays,
along with some helpful summary tables. Most of this material came
out of the GRANT reference page, although some of it is new.
Adjust a bunch of links that were pointing to GRANT to point to 5.6.
Fabien Coelho and Tom Lane, reviewed by Bradley DeJong
Michael Paquier [Mon, 3 Dec 2018 00:27:35 +0000 (09:27 +0900)]
Add PGXS options to control TAP and isolation tests, take two
The following options are added for extensions:
- TAP_TESTS, to allow an extention to run TAP tests which are the ones
present in t/*.pl. A subset of tests can always be run with the
existing PROVE_TESTS for developers.
- ISOLATION, to define a list of isolation tests.
- ISOLATION_OPTS, to pass custom options to isolation_tester.
A couple of custom Makefile rules have been accumulated across the tree
to cover the lack of facility in PGXS for a couple of releases when
using those test suites, which are all now replaced with the new flags,
without reducing the test coverage. Note that tests of contrib/bloom/
are not enabled yet, as those are proving unstable in the buildfarm.
Author: Michael Paquier Reviewed-by: Adam Berlin, Álvaro Herrera, Tom Lane, Nikolay Shaplov,
Arthur Zakirov
Discussion: https://postgr.es/m/20180906014849.GG2726@paquier.xyz
Tom Lane [Sat, 1 Dec 2018 22:19:51 +0000 (17:19 -0500)]
Eliminate parallel-make hazard in ecpg/preproc.
Re-making ecpglib's typename.o is dangerous because another make thread
could be doing that at the same time. While we've not heard field
complaints traceable to this, it seems inevitable that it'd bite someone
eventually. Instead, symlink typename.c into the preproc directory and
recompile it there. That file is small enough that compiling it twice
isn't much of a penalty. Furthermore, this way we get a .o file that's
made without shlib CFLAGS, which seems cleaner.
This requires adding more stuff to the module's -I list. The MSVC
aspect of that is untested, but I'm sure the buildfarm will tell me
if I got it wrong.
Per a suggestion from Peter Eisentraut. Although this is theoretically
a bug fix, the lack of field reports makes me feel we needn't back-patch.
Tom Lane [Sat, 1 Dec 2018 21:34:00 +0000 (16:34 -0500)]
Rename ecpg's various "extern.h" files to have distinct names.
This should reduce confusion, and in particular make it safe to
copy typename.c into preproc/ and compile it there.
This doesn't affect anything outside ecpg, and particularly not
end users, because these files don't get installed; they just
exist to share declarations among the .c files of each subdirectory.
Tom Lane [Sat, 1 Dec 2018 20:45:11 +0000 (15:45 -0500)]
Add a --socketdir option to pg_upgrade.
This allows control of the directory in which the postmaster sockets
are created for the temporary postmasters started by pg_upgrade.
The default location remains the current working directory, which is
typically fine, but if it is deeply nested then its pathname might
be too long to be a socket name.
In passing, clean up some messiness in pg_upgrade's option handling,
particularly the confusing and undocumented way that configuration-only
datadirs were handled. And fix check_required_directory's substantially
under-baked cleanup of directory pathnames.
Daniel Gustafsson, reviewed by Hironobu Suzuki, some code cleanup by me
Michael Paquier [Fri, 30 Nov 2018 22:53:18 +0000 (07:53 +0900)]
Fix tablespace path TAP test of pg_verify_checksums for msys
TAP tests on msys need to run with the DTK perl, which understands msys
virtualized paths. Postgres, however, does not understand such paths,
so before a path can be used safely with CREATE TABLESPACE, it needs to
be translated into a path on the underlying file system.
Per report from buildfarm member jacana. Suggested fix is from Andrew
Dunstan.
Michael Paquier [Fri, 30 Nov 2018 01:34:45 +0000 (10:34 +0900)]
Fix various checksum check problems for pg_verify_checksums and base backups
Three issues are fixed in this patch:
- Base backups forgot to ignore files specific to EXEC_BACKEND, leading
to spurious warnings when checksums are enabled, per analysis from me.
- pg_verify_checksums forgot about files specific to EXEC_BACKEND,
leading to failures of the tool on any such build, particularly Windows.
This error was originally found by newly-introduced TAP tests in various
buildfarm members using EXEC_BACKEND.
- pg_verify_checksums forgot to count for temporary files and temporary
paths, which could be valid relation files, without checksums, per
report from Andres Freund. More tests are added to cover this case.
A new test case which emulates corruption for a file in a different
tablespace is added, coming from from Michael Banck, while I have coded
the main code and refactored the test code.
Author: Michael Banck, Michael Paquier Reviewed-by: Stephen Frost, David Steele
Discussion: https://postgr.es/m/20181021134206.GA14282@paquier.xyz
Michael Paquier [Fri, 30 Nov 2018 01:14:58 +0000 (10:14 +0900)]
Switch pg_verify_checksums back to a blacklist
This basically reverts commit d55241af705667d4503638e3f77d3689fd6be31,
leaving around a portion of the regression tests still adapted with
empty relation files, and corrupted cases. This is also proving to be
failing to check properly relation files located in a non-default
tablespace path.
Per discussion with various folks, including Stephen Frost, David
Steele, Andres Freund, Michael Banck and myself.
Reported-by: Michael Banck
Discussion: https://postgr.es/m/20181021134206.GA14282@paquier.xyz
Backpatch-through: 11
Tom Lane [Thu, 29 Nov 2018 23:28:10 +0000 (18:28 -0500)]
Document handling of invalid/ambiguous timestamp input near DST boundaries.
The source code comments documented this, but the user-facing docs, not
so much. Add a section to Appendix B that discusses it.
In passing, improve a couple other things in Appendix B --- notably,
a long-obsolete claim that time zone abbreviations are looked up in
a fixed table.
Alvaro Herrera [Thu, 29 Nov 2018 21:42:53 +0000 (18:42 -0300)]
Add log_statement_sample_rate parameter
This allows to set a lower log_min_duration_statement value without
incurring excessive log traffic (which reduces performance). This can
be useful to analyze workloads with lots of short queries.
Author: Adrien Nayrat Reviewed-by: David Rowley, Vik Fearing
Discussion: https://postgr.es/m/c30ee535-ee1e-db9f-fa97-146b9f62caed@anayrat.info
Tom Lane [Thu, 29 Nov 2018 20:53:44 +0000 (15:53 -0500)]
Ensure static libraries have correct mod time even if ranlib messes it up.
In at least Apple's version of ranlib, the output file is updated to have
a mod time equal to the max of the timestamps of its components, and that
data only has seconds precision. On a filesystem with sub-second file
timestamp precision --- say, APFS --- this can result in the finished
static library appearing older than its input files, which causes useless
rebuilds and possible outright failures in parallel makes.
We've only seen this reported in the field from people using Apple's
ranlib with a non-Apple make, because Apple's make doesn't know about
sub-second timestamps either so it doesn't decide rebuilds are needed.
But Apple's ranlib presumably shares code with at least some BSDen,
so it's not that unlikely that the same problem could arise elsewhere.
To fix, just "touch" the output file after ranlib finishes.
We seem to need this in only one place. There are other calls of
ranlib in our makefiles, but they are working on intermediate files
whose timestamps are not actually important, or else on an installed
static library for which sub-second timestamp precision is unlikely
to matter either. (Also, so far as I can tell, Apple's ranlib doesn't
mess up the file timestamp in the latter usage anyhow.)
In passing, change "ranlib" to "$(RANLIB)" in one place that was
bypassing the make macro for no good reason.
Per bug #15525 from Jack Kelly (via Alyssa Ross).
Back-patch to all supported branches.
Peter Eisentraut [Thu, 29 Nov 2018 12:58:28 +0000 (13:58 +0100)]
doc: Add appendix detailing some limits of PostgreSQL
This used to be on the web site but was removed. The documentation is
a better place for it anyway.
Author: David Rowley <david.rowley@2ndquadrant.com> Reviewed-by: John Naylor <jcnaylor@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/CAKJS1f_dKdejdKB94nKZC9S5NzB-UZRcAKkE84e=JEEecDuotg@mail.gmail.com/
Michael Paquier [Thu, 29 Nov 2018 01:31:12 +0000 (10:31 +0900)]
Add support for NO_INSTALLCHECK in MSVC scripts
When fetching a list of tests for a given extension in contrib/ or
src/test/modules/, NO_INSTALLCHECK now gets checked first. If present,
an empty list of tests is returned to let the caller know that tests
for this module need to be bypassed.
This actually fixes a set of issues with MSVC with modules using
REGRESS_OPTS, as an incorrect parsing caused the launched command
to eat the first test listed. The actual effect on the tree is that
several modules listed a single test, so regressions have been running
with no actual tests. pg_stat_statements, test_rls_hooks and commit_ts
were impacted by that. Some other modules like test_decoding (or
snapshot_too_old) don't use yet PGXS rules, but their makefiles will
soon be refactored with an upcoming patch.
Author: Michael Paquier Reviewed-by: Andrew Dunstan
Discussion: https://postgr.es/m/20181126054302.GI1776@paquier.xyz
Michael Paquier [Thu, 29 Nov 2018 00:39:07 +0000 (09:39 +0900)]
Add missing NO_INSTALLCHECK in commit_ts and test_rls_hooks
This bypasses installcheck if specified, which makes sense for those
modules as they require non-default configuration, something which
typical users don't have. Those have been missing from the start, still
no back-patch is done.
This will be used by an upcoming patch for MSVC scripts adding support
for NO_INSTALLCHECK as installcheck is the default mode for contrib and
modules for performance reasons in the buildfarm.
Author: Michael Paquier Reviewed-by: Andrew Dunstan
Discussion: https://postgr.es/m/20181126054302.GI1776@paquier.xyz
Michael Paquier [Thu, 29 Nov 2018 00:12:19 +0000 (09:12 +0900)]
Fix handling of synchronous replication for stopping WAL senders
This fixes an oversight from c6c3334 which forgot that if a subset of
WAL senders are stopping and in a sync state, other WAL senders could
still be waiting for a WAL position to be synced while committing a
transaction. However the subset of stopping senders would not release
waiters, potentially breaking synchronous replication guarantees. This
commit makes sure that even WAL senders stopping are able to release
waiters and are tracked properly.
On 9.4, this can also trigger an assertion failure when setting for
example max_wal_senders to 1 where a WAL sender is not able to find
itself as in synchronous state when the instance stops.
Reported-by: Paul Guo
Author: Paul Guo, Michael Paquier
Discussion: https://postgr.es/m/CAEET0ZEv8VFqT3C-cQm6byOB4r4VYWcef1J21dOX-gcVhCSpmA@mail.gmail.com
Backpatch-through: 9.4
Peter Geoghegan [Wed, 28 Nov 2018 22:42:54 +0000 (14:42 -0800)]
Have BufFileSize() ereport() on FileSize() failure.
Move the responsibility for checking for and reporting a failure from
the only current BufFileSize() caller, logtape.c, to BufFileSize()
itself. Code within buffile.c is generally responsible for interfacing
with fd.c to report irrecoverable failures. This seems like a
convention that's worth sticking to.
Reorganizing things this way makes it easy to make the error message
raised in the event of BufFileSize() failure descriptive of the
underlying problem. We're now clear on the distinction between
temporary file name and BufFile name, and can show errno, confident that
its value actually relates to the error being reported. In passing, an
existing, similar buffile.c ereport() + errcode_for_file_access() site
is changed to follow the same conventions.
The API of the function BufFileSize() is changed by this commit, despite
already being in a stable release (Postgres 11). This seems acceptable,
since the BufFileSize() ABI was changed by commit aa551830421, which
hasn't made it into a point release yet. Besides, it's difficult to
imagine a third party BufFileSize() caller not just raising an error
anyway, since BufFile state should be considered corrupt when
BufFileSize() fails.
Per complaint from Tom Lane.
Discussion: https://postgr.es/m/26974.1540826748@sss.pgh.pa.us
Backpatch: 11-, where shared BufFiles were introduced.
Peter Eisentraut [Wed, 28 Nov 2018 11:36:49 +0000 (12:36 +0100)]
Only allow one recovery target setting
The previous recovery.conf regime accepted multiple recovery_target*
settings and used the last one. This does not translate well to the
general GUC system. Specifically, under EXEC_BACKEND, the settings
are written out not in any particular order, so the order in which
they were originally set is not available to new processes.
Rather than redesign the GUC system, it was decided to abandon the old
behavior and only allow one recovery target setting. A second setting
will cause an error. However, it is allowed to set the same parameter
multiple times or unset a parameter and set a different one.
Thomas Munro [Wed, 28 Nov 2018 01:00:57 +0000 (14:00 +1300)]
Don't set PAM_RHOST for Unix sockets.
Since commit 2f1d2b7a we have set PAM_RHOST to "[local]" for Unix
sockets. This caused Linux PAM's libaudit integration to make DNS
requests for that name. It's not exactly clear what value PAM_RHOST
should have in that case, but it seems clear that we shouldn't set it
to an unresolvable name, so don't do that.
Back-patch to 9.6. Bug #15520.
Author: Thomas Munro Reviewed-by: Peter Eisentraut Reported-by: Albert Schabhuetl
Discussion: https://postgr.es/m/15520-4c266f986998e1c5%40postgresql.org