Several functions could overflow their size calculations, when presented
with very large inputs from remote and/or untrusted locations, and then
allocate buffers that were too small to hold the intended contents.
Switch from int to size_t where appropriate, and check for overflow
conditions when the inputs could have plausibly originated outside of
the libpq trust boundary. (Overflows from within the trust boundary are
still possible, but these will be fixed separately.) A version of
add_size() is ported from the backend to assist with code that performs
more complicated concatenation.
Reported-by: Aleksey Solovev (Positive Technologies)
Reviewed-by: Noah Misch <noah@leadboat.com>
Reviewed-by: Álvaro Herrera <alvherre@kurilemu.de>
Security: CVE-2025-12818
Backpatch-through: 13
generic-gcc.h maps our read and write barriers to C11 acquire and
release fences using compiler builtins, for platforms where we don't
have our own hand-rolled assembler. This is apparently enough for GCC,
but the C11 memory model is only defined in terms of atomic accesses,
and our barriers for non-atomic, non-volatile accesses were not always
respected under Clang's stricter interpretation of the standard.
This explains the occasional breakage observed on new RISC-V + Clang
animal greenfly in lock-free PgAioHandle manipulation code containing a
repeating pattern of loads and read barriers. The problem can also be
observed in code generated for MIPS and LoongAarch, though we aren't
currently testing those with Clang, and on x86, though we use our own
assembler there. The scariest aspect is that we use the generic version
on very common ARM systems, but it doesn't seem to reorder the relevant
code there (or we'd have debugged this long ago).
Fix by inserting an explicit compiler barrier. It expands to an empty
assembler block declared to have memory side-effects, so registers are
flushed and reordering is prevented. In those respects this is like the
architecture-specific assembler versions, but the compiler is still in
charge of generating the appropriate fence instruction. Done for write
barriers on principle, though concrete problems have only been observed
with read barriers.
Reported-by: Alexander Lakhin <exclusion@gmail.com>
Tested-by: Alexander Lakhin <exclusion@gmail.com>
Discussion: https://postgr.es/m/d79691be-22bd-457d-9d90-18033b78c40a%40gmail.com
Backpatch-through: 13
The following parameters can only be set at server start because
their context is PGC_POSTMASTER, but this information was missing
or incorrectly documented. This commit adds or corrects
that information for the following parameters:
* debug_io_direct
* dynamic_shared_memory_type
* event_source
* huge_pages
* io_max_combine_limit
* max_notify_queue_pages
* shared_memory_type
* track_commit_timestamp
* wal_decode_buffer_size
Backpatched to all supported branches.
Author: Karina Litskevich <litskevichkarina@gmail.com>
Reviewed-by: Chao Li <lic@highgo.com>
Reviewed-by: Fujii Masao <masao.fujii@gmail.com>
Discussion: https://postgr.es/m/CAHGQGwGfPzcin-_6XwPgVbWTOUFVZgHF5g9ROrwLUdCTfjy=0A@mail.gmail.com
Backpatch-through: 13
XLogRecPtrIsInvalid() is inconsistent with the affirmative form of
macros used for other datatypes, and leads to awkward double negatives
in a few places. This commit introduces XLogRecPtrIsValid(), which
allows code to be written more naturally.
This patch only adds the new macro. XLogRecPtrIsInvalid() is left in
place, and all existing callers remain untouched. This means all
supported branches can accept hypothetical bug fixes that use the new
macro, and at the same time any code that compiled with the original
formulation will continue to silently compile just fine.
Author: Bertrand Drouvot <bertranddrouvot.pg@gmail.com>
Backpatch-through: 13
Discussion: https://postgr.es/m/aQB7EvGqrbZXrMlg@ip-10-97-1-34.eu-west-3.compute.internal
postgres_fdw supports EvalPlanQual testing by using the infrastructure
provided by the core with the RecheckForeignScan callback routine (cf.
commits 5fc4c26db and 385f337c9), but there has been no test coverage
for that, except that recent commit 12609fbac, which fixed an issue in
commit 385f337c9, added a test case to exercise only a code path added
by that commit to the core infrastructure. So let's add test cases to
exercise other code paths as well at this time.
Like commit 12609fbac, back-patch to all supported branches.
Reported-by: Masahiko Sawada <sawada.mshk@gmail.com>
Author: Etsuro Fujita <etsuro.fujita@gmail.com>
Discussion: https://postgr.es/m/CAPmGK15%2B6H%3DkDA%3D-y3Y28OAPY7fbAdyMosVofZZ%2BNc769epVTQ%40mail.gmail.com
Backpatch-through: 13
The test introduced by 17b2d5ec75 verifies that a WAL receiver
survives across a timeline jump by searching the server logs for
termination messages. However, it called restart() before the timeline
switch, which kills the WAL receiver and may log the exact message being
checked, hence failing the test. As TAP tests reuse the same log file
across restarts, a rotate_logfile() is used before the restart so as the
log matching check is not impacted by log entries generated by a
previous shutdown.
Recent changes to file handle inheritance altered I/O timing enough to
make this fail consistently while testing another patch.
While on it, this adds an extra check based on a PID comparison. This
test may lead to false positives as it could be possible that the WAL
receiver has processed a timeline jump before the initial PID is
grabbed, but it should be good enough in most cases.
Like 17b2d5ec75, backpatch down to v13.
Author: Bryan Green <dbryan.green@gmail.com>
Co-authored-by: Xuneng Zhou <xunengzhou@gmail.com>
Discussion: https://postgr.es/m/9d00b597-d64a-4f1e-802e-90f9dc394c70@gmail.com
Backpatch-through: 13
In 2a0faed9d7, which added JIT compilation support for expressions, I
accidentally used sizeof(LLVMBasicBlockRef *) instead of
sizeof(LLVMBasicBlockRef) as part of computing the size of an allocation. That
turns out to have no real negative consequences due to LLVMBasicBlockRef being
a pointer itself (and thus having the same size). It still is wrong and
confusing, so fix it.
Reported by coverity.
Backpatch-through: 13
Commit a95e3d84c0 added ActiveSnapshot push+pop when processing
work-items (BRIN autosummarization), but forgot to handle the case of
a transaction failing during the run, which drops the snapshot untimely.
Fix by making the pop conditional on an element being actually there.
Author: Álvaro Herrera <alvherre@kurilemu.de>
Backpatch-through: 13
Discussion: https://postgr.es/m/202511041648.nofajnuddmwk@alvherre.pgsql
Backpatch commit 7bc9a8bdd2 to 13-17. The motivation for backpatching is that
we want to update CI to Debian Trixie. Trixie contains a newer mingw
installation, which would trigger the warning addressed by 7bc9a8bdd2. The
risk of backpatching seems fairly low, given that it did not cause issues in
the branches the commit is already present.
While CI is not present in 13-14, it seems better to be consistent across
branches.
Author: Thomas Munro <tmunro@postgresql.org>
Discussion: https://postgr.es/m/o5yadhhmyjo53svzwvaocww6zkrp63i4f32cw3treuh46pxtza@hyqio5b2tkt6
Backpatch-through: 13
A generated column may end up being part of the partition key
expression, if it's specified as an expression e.g. "(<generated
column name>)" or if the partition key expression contains a whole-row
reference, even though we do not allow a generated column to be part
of partition key expression. Fix this hole.
Co-authored-by: jian he <jian.universality@gmail.com>
Co-authored-by: Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>
Reviewed-by: Fujii Masao <masao.fujii@oss.nttdata.com>
Discussion: https://www.postgresql.org/message-id/flat/CACJufxF%3DWDGthXSAQr9thYUsfx_1_t9E6N8tE3B8EqXcVoVfQw%40mail.gmail.com
It's possible to define BRIN indexes on functions that require a
snapshot to run, but the autosummarization feature introduced by commit
7526e10224 fails to provide one. This causes autovacuum to leave a
BRIN placeholder tuple behind after a failed work-item execution, making
such indexes less efficient. Repair by obtaining a snapshot prior to
running the task, and add a test to verify this behavior.
Author: Álvaro Herrera <alvherre@kurilemu.de>
Reported-by: Giovanni Fabris <giovanni.fabris@icon.it>
Reported-by: Arthur Nascimento <tureba@gmail.com>
Backpatch-through: 13
Discussion: https://postgr.es/m/202511031106.h4fwyuyui6fz@alvherre.pgsql
Commit b4f584f9d2 (affecting v15~, later backpatched down to 13 as of
3635a0a35a) introduced an unconditional WAL receiver shutdown when
switching from streaming to archive WAL sources. This causes problems
during a timeline switch, when a WAL receiver enters WALRCV_WAITING
state but remains alive, waiting for instructions.
The unconditional shutdown can break some monitoring scenarios as the
WAL receiver gets repeatedly terminated and re-spawned, causing
pg_stat_wal_receiver.status to show a "streaming" instead of "waiting"
status, masking the fact that the WAL receiver is waiting for a new TLI
and a new LSN to be able to continue streaming.
This commit changes the WAL receiver behavior so as the shutdown becomes
conditional, with InstallXLogFileSegmentActive being always reset to
prevent the regression fixed by b4f584f9d2a1: only terminate the WAL
receiver when it is actively streaming (WALRCV_STREAMING,
WALRCV_STARTING, or WALRCV_RESTARTING). When in WALRCV_WAITING state,
just reset InstallXLogFileSegmentActive flag to allow archive
restoration without killing the process. WALRCV_STOPPED and
WALRCV_STOPPING are not reachable states in this code path. For the
latter, the startup process is the one in charge of setting
WALRCV_STOPPING via ShutdownWalRcv(), waiting for the WAL receiver to
reach a WALRCV_STOPPED state after switching walRcvState, so
WaitForWALToBecomeAvailable() cannot be reached while a WAL receiver is
in a WALRCV_STOPPING state.
A regression test is added to check that a WAL receiver is not stopped
on timeline jump, that fails when the fix of this commit is reverted.
Reported-by: Ryan Bird <ryanzxg@gmail.com>
Author: Xuneng Zhou <xunengzhou@gmail.com>
Reviewed-by: Noah Misch <noah@leadboat.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Discussion: https://postgr.es/m/19093-c4fff49a608f82a0@postgresql.org
Backpatch-through: 13
The C standard says that the second and third arguments of a
conditional operator shall be both void type or both not-void
type. The Windows version of INTERRUPTS_PENDING_CONDITION()
got this wrong. It's pretty harmless because the result of
the operator is ignored anyway, but apparently recent versions
of MSVC have started issuing a warning about it. Silence the
warning by casting the dummy zero to void.
Reported-by: Christian Ullrich <chris@chrullrich.net>
Author: Bryan Green <dbryan.green@gmail.com>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/cc4ef8db-f8dc-4347-8a22-e7ebf44c0308@chrullrich.net
Backpatch-through: 13
Because long is 32-bit on 64-bit Windows, it isn't a good datatype to
store the difference between 2 pointers. The under-sized type could
overflow and lead to scary warnings in MEMORY_CONTEXT_CHECKING builds,
such as:
WARNING: problem in alloc set ExecutorState: bad single-chunk %p in block %p
However, the problem lies only in the code running the check, not from
an actual memory accounting bug.
Fix by using "Size" instead of "long". This means using an unsigned
type rather than the previous signed type. If the block's freeptr was
corrupted, we'd still catch that if the unsigned type wrapped. Unsigned
allows us to avoid further needless complexities around comparing signed
and unsigned types.
Author: David Rowley <dgrowleyml@gmail.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Backpatch-through: 13
Discussion: https://postgr.es/m/CAApHDvo-RmiT4s33J=aC9C_-wPZjOXQ232V-EZFgKftSsNRi4w@mail.gmail.com
The project Git server hasn't supported cloning with the Git protocol
in a very long time, but the documentation never got the memo. Remove
the mention of using the Git protocol, and while there wrap a mention
of Git in <productname> tags.
Backpatch down to all supported versions.
Author: Daniel Gustafsson <daniel@yesql.se>
Reported-by: Gurjeet Singh <gurjeet@singh.im>
Reviewed-by: Nathan Bossart <nathandbossart@gmail.com>
Reviewed-by: Jacob Champion <jacob.champion@enterprisedb.com>
Reviewed-by: Gurjeet Singh <gurjeet@singh.im>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/CABwTF4WMiMb-KT2NRcib5W0C8TQF6URMb+HK9a_=rnZnY8Q42w@mail.gmail.com
Backpatch-through: 13
These two functions expect there to be room to insert another item
in the FreePageBtree's array, but their assertions were too weak
to guarantee that. This has little practical effect granting that
the callers are not buggy, but it seems to be misleading late-model
Coverity into complaining about possible array overrun.
Author: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/799984.1761150474@sss.pgh.pa.us
Backpatch-through: 13
Commit c6f7f11d8 intended to prevent leaking any PyObject reference
counts in edge cases (such as out-of-memory during string
construction), but actually it introduced a leak in the normal case.
Repeating an error-trapping operation often enough would lead to
session-lifespan memory bloat. The problem is that I failed to
think about the fact that PyObject_GetAttrString() increments the
refcount of the returned PyObject, so that simply walking down the
list of error frame objects causes all but the first one to have
their refcount incremented.
I experimented with several more-or-less-complex ways around that,
and eventually concluded that the right fix is simply to drop the
newly-obtained refcount as soon as we walk to the next frame
object in PLy_traceback. This sounds unsafe, but it's perfectly
okay because the caller holds a refcount on the first frame object
and each frame object holds a refcount on the next one; so the
current frame object can't disappear underneath us.
By the same token, we can simplify the caller's cleanup back to
simply dropping its refcount on the first object. Cleanup of
each frame object will lead in turn to the refcount of the next
one going to zero.
I also added a couple of comments explaining why PLy_elog_impl()
doesn't try to free the strings acquired from PLy_get_spi_error_data()
or PLy_get_error_data(). That's because I got here by looking at a
Coverity complaint about how those strings might get leaked. They
are not leaked, but in testing that I discovered this other leak.
Back-patch, as c6f7f11d8 was. It's a bit nervous-making to be
putting such a fix into v13, which is only a couple weeks from its
final release; but I can't see that leaving a recently-introduced
leak in place is a better idea.
Author: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/1203918.1761184159@sss.pgh.pa.us
Backpatch-through: 13
Commit 883a95646a introduced overflow entries in the replication lag tracker
to fix an issue where lag columns in pg_stat_replication could stall when
the replay LSN stopped advancing.
This commit adds comments clarifying the purpose and behavior of overflow
entries to improve code readability and understanding.
Since commit 883a95646a was recently applied and backpatched to all
supported branches, this follow-up commit is also backpatched accordingly.
Author: Xuneng Zhou <xunengzhou@gmail.com>
Reviewed-by: Fujii Masao <masao.fujii@gmail.com>
Discussion: https://postgr.es/m/CABPTF7VxqQA_DePxyZ7Y8V+ErYyXkmwJ1P6NC+YC+cvxMipWKw@mail.gmail.com
Backpatch-through: 13
When JIT deformed tuples (controlled via the jit_tuple_deforming GUC),
types narrower than sizeof(Datum) would be zero-extended up to Datum
width. This wasn't the same as what fetch_att() does in the standard
tuple deforming code. Logically the values are the same when fetching
via the DatumGet*() marcos, but negative numbers are not the same in
binary form.
In the report, the problem was manifesting itself with:
ERROR: could not find memoization table entry
in a query which had a "Cache Mode: binary" Memoize node. However, it's
currently unclear what else is affected. Anything that uses
datum_image_eq() or datum_image_hash() on a Datum from a tuple deformed by
JIT could be affected, but it may not be limited to that.
The fix for this is simple: use signed extension instead of zero
extension.
Many thanks to Emmanuel Touzery for reporting this issue and providing
steps and backup which allowed the problem to easily be recreated.
Reported-by: Emmanuel Touzery <emmanuel.touzery@plandela.si>
Author: David Rowley <dgrowleyml@gmail.com>
Discussion: https://postgr.es/m/DB8P194MB08532256D5BAF894F241C06393F3A@DB8P194MB0853.EURP194.PROD.OUTLOOK.COM
Backpatch-through: 13
Previously, if primary_slot_name was set to an invalid slot name and
the configuration file was reloaded, both the postmaster and all other
backend processes reported a WARNING. With many processes running,
this could produce a flood of duplicate messages. The problem was that
the GUC check hook for primary_slot_name reported errors at WARNING
level via ereport().
This commit changes the check hook to use GUC_check_errdetail() and
GUC_check_errhint() for error reporting. As with other GUC parameters,
this causes non-postmaster processes to log the message at DEBUG3,
so by default, only the postmaster's message appears in the log file.
Backpatch to all supported versions.
Author: Fujii Masao <masao.fujii@gmail.com>
Reviewed-by: Chao Li <lic@highgo.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Reviewed-by: Álvaro Herrera <alvherre@kurilemu.de>
Reviewed-by: Hayato Kuroda <kuroda.hayato@fujitsu.com>
Discussion: https://postgr.es/m/CAHGQGwFud-cvthCTfusBfKHBS6Jj6kdAPTdLWKvP2qjUX6L_wA@mail.gmail.com
Backpatch-through: 13
Previously, when the replay LSN reported in feedback messages from a standby
stopped advancing, for example, due to a recovery conflict, the write_lag and
flush_lag columns in pg_stat_replication would initially update but then stop
progressing. This prevented users from correctly monitoring replication lag.
The problem occurred because when any LSN stopped updating, the lag tracker's
cyclic buffer became full (the write head reached the slowest read head).
In that state, the lag tracker could no longer compute round-trip lag values
correctly.
This commit fixes the issue by handling the slowest read entry (the one
causing the buffer to fill) as a separate overflow entry and freeing space
so the write and other read heads can continue advancing in the buffer.
As a result, write_lag and flush_lag now continue updating even if the reported
replay LSN remains stalled.
Backpatch to all supported versions.
Author: Fujii Masao <masao.fujii@gmail.com>
Reviewed-by: Chao Li <lic@highgo.com>
Reviewed-by: Shinya Kato <shinya11.kato@gmail.com>
Reviewed-by: Xuneng Zhou <xunengzhou@gmail.com>
Discussion: https://postgr.es/m/CAHGQGwGdGQ=1-X-71Caee-LREBUXSzyohkoQJd4yZZCMt24C0g@mail.gmail.com
Backpatch-through: 13
This file was previously added to v18 by commits a72f7d97be and
93fb76ca4e. Unlike the v18 version of the file, the back-branch
versions set the original baseline point to the most recent ABI
break documented in the git commit history. While we'd ordinarily
set it to something just before the .0 release, we're unlikely to
act upon ABI breaks in released minor versions, so it doesn't seem
worth the trouble to construct a comprehensive history.
Discussion: https://postgr.es/m/aPfDOD6F4FaJJd7M%40nathan
Backpatch-through: 13-17
A BlockNumber (32-bit) might not be large enough to add bo_pagesPerRange
to when the table contains close to 2^32 pages. At worst, this could
result in a cancellable infinite loop during the BRIN index scan with
power-of-2 pagesPerRange, and slow (inefficient) BRIN index scans and
scanning of unneeded heap blocks for non power-of-2 pagesPerRange.
Backpatch to all supported versions.
Author: sunil s <sunilfeb26@gmail.com>
Reviewed-by: David Rowley <dgrowleyml@gmail.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Discussion: https://postgr.es/m/CAOG6S4-tGksTQhVzJM19NzLYAHusXsK2HmADPZzGQcfZABsvpA@mail.gmail.com
Backpatch-through: 13
pgwin32_unsetenv() (compatibility routine of unsetenv() on Windows)
lacks the input validation that its sibling pgwin32_setenv() has.
Without these checks, calling unsetenv() with incorrect names crashes on
WIN32. However, invalid names should be handled, failing on EINVAL.
This commit adds the same checks as setenv() to fail with EINVAL for a
"name" set to NULL, an empty string, or if '=' is included in the value,
per POSIX requirements.
Like 7ca37fb040, backpatch down to v14. pgwin32_unsetenv() is defined
on REL_13_STABLE, but with the branch going EOL soon and the lack of
setenv() there for WIN32, nothing is done for v13.
Author: Bryan Green <dbryan.green@gmail.com>
Discussion: https://postgr.es/m/b6a1e52b-d808-4df7-87f7-2ff48d15003e@gmail.com
Backpatch-through: 14
It emerges that zlib's configuration logic is not robust enough
to guarantee that the macro will have the same ideas about struct
field layout as the library itself does, leading to corruption of
zlib's state struct followed by unintelligible failure messages.
This hazard has existed for a long time, but we'd not noticed
for several reasons:
(1) We only use gzgetc() when trying to read a manually-compressed
TOC file within a directory-format dump, which is a rarely-used
scenario that we weren't even testing before 20ec99589.
(2) No corruption actually occurs unless sizeof(long) is different
from sizeof(off_t) and the platform is big-endian.
(3) Some platforms have already fixed the configuration instability,
at least sufficiently for their environments.
Despite (3), it seems foolish to assume that the problem isn't
going to be present in some environments for a long time to come.
Hence, avoid relying on this macro. We can just #undef it and
fall back on the underlying function of the same name.
Author: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/2122679.1760846783@sss.pgh.pa.us
Backpatch-through: 13
pg_prewarm() currently checks for SELECT privileges on the target
relation. However, indexes do not have access rights of their own,
so a role may be denied permission to prewarm an index despite
having the SELECT privilege on its parent table. This commit fixes
this by locking the parent table before the index (to avoid
deadlocks) and checking for SELECT on the parent table. Note that
the code is largely borrowed from
amcheck_lock_relation_and_check().
An obvious downside of this change is the extra AccessShareLock on
the parent table during prewarming, but that isn't expected to
cause too much trouble in practice.
Author: Ayush Vatsa <ayushvatsa1810@gmail.com>
Co-authored-by: Nathan Bossart <nathandbossart@gmail.com>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Reviewed-by: Jeff Davis <pgsql@j-davis.com>
Discussion: https://postgr.es/m/CACX%2BKaMz2ZoOojh0nQ6QNBYx8Ak1Dkoko%3DD4FSb80BYW%2Bo8CHQ%40mail.gmail.com
Backpatch-through: 13
The original formulation failed to take into account the fact that for
the PGXS case, the source dir is not $(top_srcdir), so it ended up not
doing anything. Handle it explicitly.
Author: Ryo Matsumura <matsumura.ryo@fujitsu.com>
Reviewed-by: Bryan Green <dbryan.green@gmail.com>
Backpatch-through: 13
Discussion: https://postgr.es/m/TYCPR01MB113164770FB0B0BE6ED21E68EE8DCA@TYCPR01MB11316.jpnprd01.prod.outlook.com
If inside an EPQ recheck, ExecScanFetch would run the recheck method
function for foreign/custom joins even if they aren't descendant nodes
in the EPQ recheck plan tree, which is problematic at least in the
foreign-join case, because such a foreign join isn't guaranteed to have
an alternative local-join plan required for running the recheck method
function; in the postgres_fdw case this could lead to a segmentation
fault or an assert failure in an assert-enabled build when running the
recheck method function.
Even if inside an EPQ recheck, any scan nodes that aren't descendant
ones in the EPQ recheck plan tree should be normally processed by using
the access method function; fix by modifying ExecScanFetch so that if
inside an EPQ recheck, it runs the recheck method function for
foreign/custom joins that are descendant nodes in the EPQ recheck plan
tree as before and runs the access method function for foreign/custom
joins that aren't.
This fix also adds to postgres_fdw an isolation test for an EPQ recheck
that caused issues stated above.
Oversight in commit 385f337c9.
Reported-by: Kristian Lejao <kristianlejao@gmail.com>
Author: Masahiko Sawada <sawada.mshk@gmail.com>
Co-authored-by: Etsuro Fujita <etsuro.fujita@gmail.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Reviewed-by: Etsuro Fujita <etsuro.fujita@gmail.com>
Discussion: https://postgr.es/m/CAD21AoBpo6Gx55FBOW+9s5X=nUw3Xpq64v35fpDEKsTERnc4TQ@mail.gmail.com
Backpatch-through: 13
log_error() would probably fail completely if used, and would
certainly print garbage for anything that needed to be interpolated
into the message, because it was failing to use the correct printing
subroutine for a va_list argument.
This bug likely went undetected because the error cases this code
is used for are rarely exercised - they only occur when Windows
security API calls fail catastrophically (out of memory, security
subsystem corruption, etc).
The FRONTEND variant can be fixed just by calling vfprintf()
instead of fprintf(). However, there was no va_list variant
of write_stderr(), so create one by refactoring that function.
Following the usual naming convention for such things, call
it vwrite_stderr().
Author: Bryan Green <dbryan.green@gmail.com>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/CAF+pBj8goe4fRmZ0V3Cs6eyWzYLvK+HvFLYEYWG=TzaM+tWPnw@mail.gmail.com
Backpatch-through: 13
There was some confusion around how to adjust the n_distinct estimates
for partitioned tables. Here we try and clarify that
n_distinct_inherited needs to be adjusted rather than n_distinct.
Also fix some slightly misleading text which was talking about table
size rather than table rows, fix a grammatical error, and adjust some
text which indicated that ANALYZE was performing calculations based on
the n_distinct settings. Really it's the query planner that does this
and ANALYZE only stores the overridden n_distinct estimate value in
pg_statistic.
Author: David Rowley <dgrowleyml@gmail.com>
Reviewed-by: David G. Johnston <david.g.johnston@gmail.com>
Reviewed-by: Chao Li <li.evan.chao@gmail.com>
Backpatch-through: 13
Discussion: https://postgr.es/m/CAApHDvrL7a-ZytM1SP8Uk9nEw9bR2CPzVb+uP+bcNj=_q-ZmVw@mail.gmail.com
Commit 71f4c8c6f7 (which implemented DETACH CONCURRENTLY) added code
to create a separate table constraint when a table is detached
concurrently, identical to the partition constraint, on the theory that
such a constraint was needed in case the optimizer had constructed any
query plans that depended on the constraint being there. However, that
theory was apparently bogus because any such plans would be invalidated.
For hash partitioning, those constraints are problematic, because their
expressions reference the OID of the parent partitioned table, to which
the detached table is no longer related; this causes all sorts of
problems (such as inability of restoring a pg_dump of that table, and
the table no longer working properly if the partitioned table is later
dropped).
We'd like to get rid of all those constraints. In fact, for branch
master, do that -- no longer create any substitute constraints.
However, out of fear that some users might somehow depend on these
constraints for other partitioning strategies, for stable branches
(back to 14, which added DETACH CONCURRENTLY), only do it for hash
partitioning.
(If you repeatedly DETACH CONCURRENTLY and then ATTACH a partition, then
with this constraint addition you don't need to scan the table in the
ATTACH step, which presumably is good. But if users really valued this
feature, they would have requested that it worked for non-concurrent
DETACH also.)
Author: Haiyang Li <mohen.lhy@alibaba-inc.com>
Reported-by: Fei Changhong <feichanghong@qq.com>
Reported-by: Haiyang Li <mohen.lhy@alibaba-inc.com>
Backpatch-through: 14
Discussion: https://postgr.es/m/18371-7fef49f63de13f02@postgresql.org
Discussion: https://postgr.es/m/19070-781326347ade7c57@postgresql.org
An assertion in _bt_killitems expected the scan's currPos state to
contain a valid LSN, saved from when currPos's page was initially read.
The assertion failed to account for the fact that even logged relations
can have leaf pages with an invalid LSN when built with wal_level set to
"minimal". Remove the faulty assertion.
Oversight in commit e6eed40e (though note that the assertion was
backpatched to stable branches before 18 by commit 7c319f54).
Author: Peter Geoghegan <pg@bowt.ie>
Reported-By: Matthijs van der Vleuten <postgresql@zr40.nl>
Bug: #19082
Discussion: https://postgr.es/m/19082-628e62160dbbc1c1@postgresql.org
Backpatch-through: 13
An error happening while a slot data is saved on disk in
SaveSlotToPath() could cause a state.tmp file (temporary file holding
the slot state data, renamed to its permanent name at the end of the
function) to remain around after it has been created. This temporary
file is created with O_EXCL, meaning that if an existing state.tmp is
found, its creation would fail. This would prevent the slot data to be
saved, requiring a manual intervention to remove state.tmp before being
able to save again a slot. Possible scenarios where this temporary file
could remain on disk is for example a ENOSPC case (no disk space) while
writing, syncing or renaming it. The bug reports point to a write
failure as the principal cause of the problems.
Using O_TRUNC has been argued back in 2019 as a potential solution to
discard any temporary file that could exist. This solution was rejected
as O_EXCL can also act as a safety measure when saving the slot state,
crash recovery offering cleanup guarantees post-crash. This commit uses
the alternative approach that has been suggested by Andres Freund back
in 2019. When the temporary state file cannot be written, synced,
closed or renamed (note: not when created!), an unlink() is used to
remove the temporary state file while holding the in-progress I/O
LWLock, so as any follow-up attempts to save a slot's data would not
choke on an existing file that remained around because of a previous
failure.
This problem has been reported a few times across the years, going back
to 2019, but for some reason I have never come back to do something
about it and it has been forgotten. A recent report has reminded me
that this was still a problem.
Reported-by: Kevin K Biju <kevinkbiju@gmail.com>
Reported-by: Sergei Kornilov <sk@zsrv.org>
Reported-by: Grigory Smolkin <g.smolkin@postgrespro.ru>
Discussion: https://postgr.es/m/CAM45KeHa32soKL_G8Vk38CWvTBeOOXcsxAPAs7Jt7yPRf2mbVA@mail.gmail.com
Discussion: https://postgr.es/m/3559061693910326@qy4q4a6esb2lebnz.sas.yp-c.yandex.net
Discussion: https://postgr.es/m/08bbfab1-a61d-3750-fc18-4ab2c1aa7f09@postgrespro.ru
Backpatch-through: 13
On Windows, this code did not handle error conditions correctly at
all, since it looked at "errno" which is not used for socket-related
errors on that platform. This resulted, for example, in failure
to connect to a PostgreSQL server with GSSAPI enabled.
We have a convention for dealing with this within libpq, which is to
use SOCK_ERRNO and SOCK_ERRNO_SET rather than touching errno directly;
but the GSSAPI code is a relative latecomer and did not get that memo.
(The equivalent backend code continues to use errno, because the
backend does this differently. Maybe libpq's approach should be
rethought someday.)
Apparently nobody tries to build libpq with GSSAPI support on Windows,
or we'd have heard about this before, because it's been broken all
along. Back-patch to all supported branches.
Author: Ning Wu <ning94803@gmail.com>
Co-authored-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/CAFGqpvg-pRw=cdsUpKYfwY6D3d-m9tw8WMcAEE7HHWfm-oYWvw@mail.gmail.com
Backpatch-through: 13
Currently, pgbench aborts when a COPY response is received in
readCommandResponse(). However, as PQgetResult() returns an empty
result when there is no asynchronous result, through getCopyResult(),
the logic done at the end of readCommandResponse() for the error path
leads to an infinite loop.
This commit forcefully exits the COPY state with PQendcopy() before
moving to the error handler when fiding a COPY state, avoiding the
infinite loop. The COPY protocol is not supported by pgbench anyway, as
an error is assumed in this case, so giving up is better than having the
tool be stuck forever. pgbench was interruptible in this state.
A TAP test is added to check that an error happens if trying to use
COPY.
Author: Anthonin Bonnefoy <anthonin.bonnefoy@datadoghq.com>
Discussion: https://postgr.es/m/CAO6_XqpHyF2m73ifV5a=5jhXxH2chk=XrgefY+eWWPe2Eft3=A@mail.gmail.com
Backpatch-through: 13
pgstattuple checks the state of the pages retrieved for gist and hash
using some check functions from each index AM, respectively
gistcheckpage() and _hash_checkpage(). When these are called, they
would fail when bumping on data that is found as incorrect (like opaque
area size not matching, or empty pages), contrary to btree that simply
discards these cases and continues to aggregate data.
Zero pages can happen after a crash, with these AMs being able to do an
internal cleanup when these are seen. Also, sporadic failures are
annoying when doing for example a large-scale diagnostic query based on
pgstattuple with a join of pg_class, as it forces one to use tricks like
quals to discard hash or gist indexes, or use a PL wrapper able to catch
errors.
This commit changes the reports generated for btree, gist and hash to
be more user-friendly;
- When seeing an empty page, report it as free space. This new rule
applies to gist and hash, and already applied to btree.
- For btree, a check based on the size of BTPageOpaqueData is added.
- For gist indexes, gistcheckpage() is not called anymore, replaced by a
check based on the size of GISTPageOpaqueData.
- For hash indexes, instead of _hash_getbuf_with_strategy(), use a
direct call to ReadBufferExtended(), coupled with a check based on
HashPageOpaqueData. The opaque area size check was already used.
- Pages that do not match these criterias are discarded from the stats
reports generated.
There have been a couple of bug reports over the years that complained
about the current behavior for hash and gist, as being not that useful,
with nothing being done about it. Hence this change is backpatched down
to v13.
Reported-by: Noah Misch <noah@leadboat.com>
Author: Nitin Motiani <nitinmotiani@google.com>
Reviewed-by: Dilip Kumar <dilipbalaut@gmail.com>
Discussion: https://postgr.es/m/CAH5HC95gT1J3dRYK4qEnaywG8RqjbwDdt04wuj8p39R=HukayA@mail.gmail.com
Backpatch-through: 13
pgbench uses readCommandResponse() to process server responses.
When readCommandResponse() encounters an error during a call to
PQgetResult() to fetch the current result, it attempts to report it
with an additional error message from PQerrorMessage(). However,
previously, this extra error message could be lost or become incorrect.
The cause was that after fetching the current result (and detecting
an error), readCommandResponse() called PQgetResult() again to
peek at the next result. This second call could overwrite the libpq
connection's error message before the original error was reported,
causing the error message retrieved from PQerrorMessage() to be
lost or overwritten.
This commit fixes the issue by updating readCommandResponse()
to use PQresultErrorMessage() instead of PQerrorMessage()
to retrieve the error message generated when the PQgetResult()
for the current result causes an error, ensuring the correct message
is reported.
Backpatch to all supported versions.
Author: Yugo Nagata <nagata@sraoss.co.jp>
Reviewed-by: Chao Li <lic@highgo.com>
Reviewed-by: Fujii Masao <masao.fujii@gmail.com>
Discussion: https://postgr.es/m/20250925110940.ebacc31725758ec47d5432c6@sraoss.co.jp
Backpatch-through: 13
Neighbor get_statistics_object_oid() ignores objects in pg_temp, as has
been the standard for non-relation, non-type namespace searches since
CVE-2007-2138. Hence, most operations that name a statistics object
correctly decline to map an unqualified name to a statistics object in
pg_temp. StatisticsObjIsVisibleExt() did not. Consequently,
pg_statistics_obj_is_visible() wrongly returned true for such objects,
psql \dX wrongly listed them, and getObjectDescription()-based ereport()
and pg_describe_object() wrongly omitted namespace qualification. Any
malfunction beyond that would depend on how a human or application acts
on those wrong indications. Commit
d99d58cdc8 introduced this. Back-patch to
v13 (all supported versions).
Reviewed-by: Nathan Bossart <nathandbossart@gmail.com>
Discussion: https://postgr.es/m/20250920162116.2e.nmisch@google.com
Backpatch-through: 13
Because we failed to do this, DISTINCT in GROUP BY DISTINCT would be
ignored in PL/pgSQL assignment statements. It's not surprising that
no one noticed, since such statements will throw an error if the query
produces more than one row. That eliminates most scenarios where
advanced forms of GROUP BY could be useful, and indeed makes it hard
even to find a simple test case. Nonetheless it's wrong.
This is directly the fault of be45be9c3 which added the groupDistinct
field, but I think much of the blame has to fall on c9d529848, in
which I incautiously supposed that we'd manage to keep two copies of
a big chunk of parse-analysis logic in sync. As a follow-up, I plan
to refactor so that there's only one copy. But that seems useful
only in master, so let's use this one-line fix for the back branches.
Author: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/31027.1758919078@sss.pgh.pa.us
Backpatch-through: 14
Jacob Champion