We must tell init about each role name we plan to connect as,
else SSPI auth fails. Similar to previous patches such as
14793f471, 973542866.
Oversight in 208927e65, per buildfarm member drongo.
(Although that was back-patched to v13, the test script
only exists in v16 and up.)
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
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
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
Commit 4b754d6c1 introduced the concept of an excludeOnly scan key,
which cannot select matching index entries but can reject
non-matching tuples, for example a tsquery such as '!term'. There are
poorly-documented assumptions that such scan keys do not appear as the
first scan key. ginNewScanKey did nothing to ensure that, however,
with the result that certain GIN index searches could go into an
infinite loop while apparently-equivalent queries with the clauses in
a different order were fine.
Fix by teaching ginNewScanKey to place all excludeOnly scan keys
after all not-excludeOnly ones. So far as we know at present,
it might be sufficient to avoid the case where the very first
scan key is excludeOnly; but I'm not very convinced that there
aren't other dependencies on the ordering.
Bug: #19031
Reported-by: Tim Wood <washwithcare@gmail.com>
Author: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/19031-0638148643d25548@postgresql.org
Backpatch-through: 13
This commit adds CHECK_FOR_INTERRUPTS to loops iterating over shared
buffers in several pg_buffercache functions, allowing them to be
interrupted during long-running operations.
Backpatch to all supported versions. Add CHECK_FOR_INTERRUPTS to the
loop in pg_buffercache_pages() in all supported branches, and to
pg_buffercache_summary() and pg_buffercache_usage_counts() in version
16 and newer.
Author: SATYANARAYANA NARLAPURAM <satyanarlapuram@gmail.com>
Discussion: https://postgr.es/m/CAHg+QDcejeLx7WunFT3DX6XKh1KshvGKa8F5au8xVhqVvvQPRw@mail.gmail.com
Backpatch-through: 13
The tests fixed in this commit were changing the sampling setting of a
foreign server, but then were analyzing a local table instead of a
foreign table, meaning that the test was not running for its original
purpose.
This commit changes the ANALYZE commands to analyze the foreign table,
and changes the foreign table definition to point to a valid remote
table. Attempting to analyze the foreign table "analyze_ftable" would
have failed before this commit, because "analyze_rtable1" is not defined
on the remote side.
Issue introduced by 8ad51b5f44.
Author: Corey Huinker <corey.huinker@gmail.com>
Discussion: https://postgr.es/m/CADkLM=cpUiJ3QF7aUthTvaVMmgQcm7QqZBRMDLhBRTR+gJX-Og@mail.gmail.com
Backpatch-through: 16
The code used
return (Selectivity) 0.0;
where
PG_RETURN_FLOAT8(0.0);
would be correct.
On 64-bit systems, these are pretty much equivalent, but on 32-bit
systems, PG_RETURN_FLOAT8() correctly produces a pointer, but the old
wrong code would return a null pointer, possibly leading to a crash
elsewhere.
We think this code is actually not reachable because bqarr_in won't
accept an empty query, and there is no other function that will
create query_int values. But better be safe and not let such
incorrect code lie around.
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://www.postgresql.org/message-id/flat/8246d7ff-f4b7-4363-913e-827dadfeb145%40eisentraut.org
Commit 9e6104c66 disallowed transition tables on foreign tables, but
failed to account for cases where a foreign table is a child table of a
partitioned/inherited table on which transition tables exist, leading to
incorrect transition tuples collected from such foreign tables for
queries on the parent table triggering transition capture. This
occurred not only for inherited UPDATE/DELETE but for partitioned INSERT
later supported by commit 3d956d956, which should have handled it at
least for the INSERT case, but didn't.
To fix, modify ExecAR*Triggers to throw an error if the given relation
is a foreign table requesting transition capture. Also, this commit
fixes make_modifytable so that in case of an inherited UPDATE/DELETE
triggering transition capture, FDWs choose normal operations to modify
child foreign tables, not DirectModify; which is needed because they
would otherwise skip the calls to ExecAR*Triggers at execution, causing
unexpected behavior.
Author: Etsuro Fujita <etsuro.fujita@gmail.com>
Reviewed-by: Amit Langote <amitlangote09@gmail.com>
Discussion: https://postgr.es/m/CAPmGK14QJYikKzBDCe3jMbpGENnQ7popFmbEgm-XTNuk55oyHg%40mail.gmail.com
Backpatch-through: 13
Currently, ALTER DATABASE/ROLE/SYSTEM RESET [ALL] with an unknown
custom GUC with a prefix reserved by MarkGUCPrefixReserved() errors
(unless a superuser runs a RESET ALL variant). This is problematic
for cases such as an extension library upgrade that removes a GUC.
To fix, simply make sure the relevant code paths explicitly allow
it. Note that we require superuser or privileges on the parameter
to reset it. This is perhaps a bit more restrictive than is
necessary, but it's not clear whether further relaxing the
requirements is safe.
Oversight in commit 88103567cb. The ALTER SYSTEM fix is dependent
on commit 2d870b4aef, which first appeared in v17. Unfortunately,
back-patching that commit would introduce ABI breakage, and while
that breakage seems unlikely to bother anyone, it doesn't seem
worth the risk. Hence, the ALTER SYSTEM part of this commit is
omitted on v15 and v16.
Reported-by: Mert Alev <mert@futo.org>
Reviewed-by: Laurenz Albe <laurenz.albe@cybertec.at>
Discussion: https://postgr.es/m/18964-ba09dea8c98fccd6%40postgresql.org
Backpatch-through: 15
libxml2 has deprecated the members of xmlBuffer, and it is recommended
to access them with dedicated routines. We have only one case in the
tree where this shows an impact: xml2/xpath.c where "content" was
getting directly accessed. The rest of the code looked fine, checking
the PostgreSQL code with libxml2 close to the top of its "2.14" branch.
xmlBufferContent() exists since year 2000 based on a check of the
upstream libxml2 tree, so let's switch to it.
Like 400928b83b, backpatch all the way down as this can have an impact
on all the branches already released once newer versions of libxml2 get
more popular.
Reported-by: Walid Ibrahim <walidib@amazon.com>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/aGdSdcR4QTjEHX6s@paquier.xyz
Backpatch-through: 13
Commit 4909b38af0 introduced logic to distribute invalidation messages
from catalog-modifying transactions to all concurrent in-progress
transactions. However, since each transaction distributes not only its
original invalidation messages but also previously distributed
messages to other transactions, this leads to an exponential increase
in allocation request size for invalidation messages, ultimately
causing memory allocation failure.
This commit fixes this issue by tracking distributed invalidation
messages separately per decoded transaction and not redistributing
these messages to other in-progress transactions. The maximum size of
distributed invalidation messages that one transaction can store is
limited to MAX_DISTR_INVAL_MSG_PER_TXN (8MB). Once the size of the
distributed invalidation messages exceeds this threshold, we
invalidate all caches in locations where distributed invalidation
messages need to be executed.
Back-patch to all supported versions where we introduced the fix by
commit 4909b38af0.
Note that this commit adds two new fields to ReorderBufferTXN to store
the distributed transactions. This change breaks ABI compatibility in
back branches, affecting third-party extensions that depend on the
size of the ReorderBufferTXN struct, though this scenario seems
unlikely.
Additionally, it adds a new flag to the txn_flags field of
ReorderBufferTXN to indicate distributed invalidation message
overflow. This should not affect existing implementations, as it is
unlikely that third-party extensions use unused bits in the txn_flags
field.
Bug: #18938#18942
Author: vignesh C <vignesh21@gmail.com>
Reported-by: Duncan Sands <duncan.sands@deepbluecap.com>
Reported-by: John Hutchins <john.hutchins@wicourts.gov>
Reported-by: Laurence Parry <greenreaper@hotmail.com>
Reported-by: Max Madden <maxmmadden@gmail.com>
Reported-by: Braulio Fdo Gonzalez <brauliofg@gmail.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Reviewed-by: Hayato Kuroda <kuroda.hayato@fujitsu.com>
Discussion: https://postgr.es/m/680bdaf6-f7d1-4536-b580-05c2760c67c6@deepbluecap.com
Discussion: https://postgr.es/m/18942-0ab1e5ae156613ad@postgresql.org
Discussion: https://postgr.es/m/18938-57c9a1c463b68ce0@postgresql.org
Discussion: https://postgr.es/m/CAD1FGCT2sYrP_70RTuo56QTizyc+J3wJdtn2gtO3VttQFpdMZg@mail.gmail.com
Discussion: https://postgr.es/m/CANO2=B=2BT1hSYCE=nuuTnVTnjidMg0+-FfnRnqM6kd23qoygg@mail.gmail.com
Backpatch-through: 13
We store values for these options as array elements with the syntax
"name=value", hence a name containing "=" confuses matters when
it's time to read the array back in. Since validation of the
options is often done (long) after this conversion to array format,
that leads to confusing and off-point error messages. We can
improve matters by rejecting names containing "=" up-front.
(Probably a better design would have involved pairs of array
elements, but it's too late now --- and anyway, there's no
evident use-case for option names like this. We already
reject such names in some other contexts such as GUCs.)
Reported-by: Chapman Flack <jcflack@acm.org>
Author: Tom Lane <tgl@sss.pgh.pa.us>
Reviewed-by: Chapman Flack <jcflack@acm.org>
Discussion: https://postgr.es/m/6830EB30.8090904@acm.org
Backpatch-through: 13
postgres_fdw tries to use PG_TRY blocks to ensure that it will
eventually free the PGresult created by the remote modify command.
However, it's fundamentally impossible for this scheme to work
reliably when there's RETURNING data, because the query could fail
in between invocations of postgres_fdw's DirectModify methods.
There is at least one instance of exactly this situation in the
regression tests, and the ensuing session-lifespan leak is visible
under Valgrind.
We can improve matters by using a memory context reset callback
attached to the ExecutorState context. That ensures that the
PGresult will be freed when the ExecutorState context is torn
down, even if control never reaches postgresEndDirectModify.
I have little faith that there aren't other potential PGresult
leakages in the backend modules that use libpq. So I think it'd
be a good idea to apply this concept universally by creating
infrastructure that attaches a reset callback to every PGresult
generated in the backend. However, that seems too invasive for
v18 at this point, let alone the back branches. So for the
moment, apply this narrow fix that just makes DirectModify safe.
I have a patch in the queue for the more general idea, but it
will have to wait for v19.
Author: Tom Lane <tgl@sss.pgh.pa.us>
Reviewed-by: Matheus Alcantara <matheusssilv97@gmail.com>
Discussion: https://postgr.es/m/2976982.1748049023@sss.pgh.pa.us
Backpatch-through: 13
If we hit out-of-memory between creating the PGconn and inserting
it into dblink's hashtable, we'd lose track of the PGconn, which
is quite bad since it represents a live connection to a remote DB.
Fix by rearranging things so that we create the hashtable entry
first.
Also reduce the number of states we have to deal with by getting rid
of the separately-allocated remoteConn object, instead allocating it
in-line in the hashtable entries. (That incidentally removes a
session-lifespan memory leak observed in the regression tests.)
There is an apparently-irreducible remaining OOM hazard, which
is that if the connection fails at the libpq level (ie it's
CONNECTION_BAD) then we have to pstrdup the PGconn's error message
before we can release it, and theoretically that could fail. However,
in such cases we're only leaking memory not a live remote connection,
so I'm not convinced that it's worth sweating over.
This is a pretty low-probability failure mode of course, but losing
a live connection seems bad enough to justify back-patching.
Author: Tom Lane <tgl@sss.pgh.pa.us>
Reviewed-by: Matheus Alcantara <matheusssilv97@gmail.com>
Discussion: https://postgr.es/m/1346940.1748381911@sss.pgh.pa.us
Backpatch-through: 13
An assertion test added in commit 049ef33 could fail when pg_prewarm()
was called on objects without storage, such as partitioned tables.
This resulted in the following failure in assert-enabled builds:
Failed Assert("RelFileNumberIsValid(rlocator.relNumber)")
Note that, in non-assert builds, pg_prewarm() just failed with an error
in that case, so there was no ill effect in practice.
This commit fixes the issue by having pg_prewarm() raise an error early
if the specified object has no storage. This approach is similar to
the fix in commit 4623d7144 for pg_freespacemap.
Back-patched to v17, where the issue was introduced.
Author: Masahiro Ikeda <ikedamsh@oss.nttdata.com>
Reviewed-by: Dilip Kumar <dilipbalaut@gmail.com>
Reviewed-by: Richard Guo <guofenglinux@gmail.com>
Reviewed-by: Fujii Masao <masao.fujii@gmail.com>
Discussion: https://postgr.es/m/e082e6027610fd0a4091ae6d033aa117@oss.nttdata.com
Backpatch-through: 17
pg_stat_statements anticipates that certain constant locations may be
recorded multiple times and attempts to avoid calculating a length for
these locations in fill_in_constant_lengths().
However, during generate_normalized_query() where normalized query
strings are generated, these locations are not excluded from
consideration. This could increment the parameter number counter for
every recorded occurrence at such a location, leading to an incorrect
normalization in certain cases with gaps in the numbers reported.
For example, take this query:
SELECT WHERE '1' IN ('2'::int, '3'::int::text)
Before this commit, it would be normalized like that, with gaps in the
parameter numbers:
SELECT WHERE $1 IN ($3::int, $4::int::text)
However the correct, less confusing one should be like that:
SELECT WHERE $1 IN ($2::int, $3::int::text)
This commit fixes the computation of the parameter numbers to track the
number of constants replaced with an $n by a separate counter instead of
the iterator used to loop through the list of locations.
The underlying query IDs are not changed, neither are the normalized
strings for existing PGSS hash entries. New entries with fresh
normalized queries would automatically get reshaped based on the new
parameter numbering.
Issue discovered while discussing a separate problem for HEAD, but this
affects all the stable branches.
Author: Sami Imseih <samimseih@gmail.com>
Discussion: https://postgr.es/m/CAA5RZ0tzxvWXsacGyxrixdhy3tTTDfJQqxyFBRFh31nNHBQ5qA@mail.gmail.com
Backpatch-through: 13
During logical decoding, we advance catalog_xmin of logical too early in
fast_forward mode, resulting in required catalog data being removed by
vacuum. This mode is normally used to advance the slot without processing
the changes, but we still can't let the slot's xmin to advance to an
incorrect value.
Commit f49a80c481 fixed a similar issue where the logical slot's
catalog_xmin was getting advanced prematurely during non-fast-forward
mode. During xl_running_xacts processing, instead of directly advancing
the slot's xmin to the oldest running xid in the record, it allowed the
xmin to be held back for snapshots that can be used for
not-yet-replayed transactions, as those might consider older txns as
running too. However, it missed the fact that the same problem can happen
during fast_forward mode decoding, as we won't build a base snapshot in
that mode, and the future call to get_changes from the same slot can miss
seeing the required catalog changes leading to incorrect reslts.
This commit allows building the base snapshot even in fast_forward mode to
prevent the early advancement of xmin.
Reported-by: Amit Kapila <amit.kapila16@gmail.com>
Author: Zhijie Hou <houzj.fnst@fujitsu.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Reviewed-by: shveta malik <shveta.malik@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Backpatch-through: 13
Discussion: https://postgr.es/m/CAA4eK1LqWncUOqKijiafe+Ypt1gQAQRjctKLMY953J79xDBgAg@mail.gmail.com
Discussion: https://postgr.es/m/OS0PR01MB57163087F86621D44D9A72BF94BB2@OS0PR01MB5716.jpnprd01.prod.outlook.com
The original intent in heap_page_items() was to return nulls, not
throw an error or crash, if an item was sufficiently corrupt that
we couldn't safely extract data from it. However, commit d6061f83a
utterly missed that memo, and not only put in an un-length-checked
copy of the tuple's data section, but also managed to break the check
on sane nulls-bitmap length. Either mistake could possibly lead to
a SIGSEGV crash if the tuple is corrupt.
Bug: #18896
Reported-by: Dmitry Kovalenko <d.kovalenko@postgrespro.ru>
Author: Dmitry Kovalenko <d.kovalenko@postgrespro.ru>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/18896-add267b8e06663e3@postgresql.org
Backpatch-through: 13
Commit 0f21db36d made an assumption that GIN triConsistentFns
would not modify their input entryRes[] arrays. But in fact,
the "shim" triConsistentFn that we use for opclasses that don't
supply their own did exactly that, potentially leading to wrong
answers from a GIN index search. Through bad luck, none of the
test cases that we have for such opclasses exposed the bug.
One response to this could be that the assumption of consistency check
functions not modifying entryRes[] arrays is a bad one, but it still
seems reasonable to me. Notably, shimTriConsistentFn is itself
assuming that with respect to the underlying boolean consistentFn,
so it's sure being self-centered in supposing that it gets to do so.
Fortunately, it's quite simple to fix shimTriConsistentFn to restore
the entry-time state of entryRes[], so let's do that instead.
This issue doesn't affect any core GIN opclasses, since they all
supply their own triConsistentFns. It does affect contrib modules
btree_gin, hstore, and intarray.
Along the way, I (tgl) noticed that shimTriConsistentFn failed to
pick up on a "recheck" flag returned by its first call to the boolean
consistentFn. This may be only a latent problem, since it would be
unlikely for a consistentFn to set recheck for the all-false case
and not any other cases. (Indeed, none of our contrib modules do
that.) Nonetheless, it's formally wrong.
Reported-by: Vinod Sridharan <vsridh90@gmail.com>
Author: Vinod Sridharan <vsridh90@gmail.com>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/CAFMdLD7XzsXfi1+DpTqTgrD8XU0i2C99KuF=5VHLWjx4C1pkcg@mail.gmail.com
Backpatch-through: 13
Data loss can happen when the DDLs like ALTER PUBLICATION ... ADD TABLE ...
or ALTER TYPE ... that don't take a strong lock on table happens
concurrently to DMLs on the tables involved in the DDL. This happens
because logical decoding doesn't distribute invalidations to concurrent
transactions and those transactions use stale cache data to decode the
changes. The problem becomes bigger because we keep using the stale cache
even after those in-progress transactions are finished and skip the
changes required to be sent to the client.
This commit fixes the issue by distributing invalidation messages from
catalog-modifying transactions to all concurrent in-progress transactions.
This allows the necessary rebuild of the catalog cache when decoding new
changes after concurrent DDL.
We observed performance regression primarily during frequent execution of
*publication DDL* statements that modify the published tables. The
regression is minor or nearly nonexistent for DDLs that do not affect the
published tables or occur infrequently, making this a worthwhile cost to
resolve a longstanding data loss issue.
An alternative approach considered was to take a strong lock on each
affected table during publication modification. However, this would only
address issues related to publication DDLs (but not the ALTER TYPE ...)
and require locking every relation in the database for publications
created as FOR ALL TABLES, which is impractical.
The bug exists in all supported branches, but we are backpatching till 14.
The fix for 13 requires somewhat bigger changes than this fix, so the fix
for that branch is still under discussion.
Reported-by: hubert depesz lubaczewski <depesz@depesz.com>
Reported-by: Tomas Vondra <tomas.vondra@enterprisedb.com>
Author: Shlok Kyal <shlok.kyal.oss@gmail.com>
Author: Hayato Kuroda <kuroda.hayato@fujitsu.com>
Reviewed-by: Zhijie Hou <houzj.fnst@fujitsu.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Tested-by: Benoit Lobréau <benoit.lobreau@dalibo.com>
Backpatch-through: 14
Discussion: https://postgr.es/m/de52b282-1166-1180-45a2-8d8917ca74c6@enterprisedb.com
Discussion: https://postgr.es/m/CAD21AoAenVqiMjpN-PvGHL1N9DWnHSq673bfgr6phmBUzx=kLQ@mail.gmail.com
If a column is omitted in an INSERT, and there's no column default,
the code in preptlist.c generates a NULL Const to be inserted.
Furthermore, if the column is of a domain type, we wrap the Const
in CoerceToDomain, so as to throw a run-time error if the domain
has a NOT NULL constraint. That's fine as far as it goes, but
there are two problems:
1. We're being sloppy about the type/typmod that the Const is
labeled with. It really should have the domain's base type/typmod,
since it's the input to CoerceToDomain not the output. This can
result in coerce_to_domain inserting a useless length-coercion
function (useless because it's being applied to a null). The
coercion would typically get const-folded away later, but it'd
be better not to create it in the first place.
2. We're not applying expression preprocessing (specifically,
eval_const_expressions) to the resulting expression tree.
The planner's primary expression-preprocessing pass already happened,
so that means the length coercion step and CoerceToDomain node miss
preprocessing altogether.
This is at the least inefficient, since it means the length coercion
and CoerceToDomain will actually be executed for each inserted row,
though they could be const-folded away in most cases. Worse, it
seems possible that missing preprocessing for the length coercion
could result in an invalid plan (for example, due to failing to
perform default-function-argument insertion). I'm not aware of
any live bug of that sort with core datatypes, and it might be
unreachable for extension types as well because of restrictions of
CREATE CAST, but I'm not entirely convinced that it's unreachable.
Hence, it seems worth back-patching the fix (although I only went
back to v14, as the patch doesn't apply cleanly at all in v13).
There are several places in the rewriter that are building null
domain constants the same way as preptlist.c. While those are
before the planner and hence don't have any reachable bug, they're
still applying a length coercion that will be const-folded away
later, uselessly wasting cycles. Hence, make a utility routine
that all of these places can call to do it right.
Making this code more careful about the typmod assigned to the
generated NULL constant has visible but cosmetic effects on some
of the plans shown in contrib/postgres_fdw's regression tests.
Discussion: https://postgr.es/m/1865579.1738113656@sss.pgh.pa.us
Backpatch-through: 14
This patch fixes two distinct errors that both ultimately trace
to commit 71d60e2aa, which added the ats_modifiedcols field.
The more severe error is that ats_modifiedcols wasn't accounted for
in afterTriggerAddEvent's scanning loop that looks for a pre-existing
duplicate AfterTriggerSharedData. Thus, a new event could be
incorrectly matched to an AfterTriggerSharedData that has a different
value of ats_modifiedcols, resulting in the wrong tg_updatedcols
bitmap getting passed to the trigger whenever it finally gets fired.
We'd not noticed because (a) few triggers consult tg_updatedcols,
and (b) we had no tests exercising a case where such a trigger was
called as an AFTER trigger. In the test case added by this commit,
contrib/lo's trigger fails to remove a large object when expected
because (without this fix) it thinks the LO OID column hasn't changed.
The other problem was introduced by commit ce5aaea8c, which copied the
modified-columns bitmap into trigger-related storage. It made a copy
for every trigger event, whereas what we really want is to make a new
copy only when we make a new AfterTriggerSharedData entry. (We could
imagine adding extra logic to reduce the number of bitmap copies still
more, but it doesn't look worthwhile at the moment.) In a simple test
of an UPDATE of 10000000 rows with a single AFTER trigger, this thinko
roughly tripled the amount of memory consumed by the pending-triggers
data structures, from 160446744 to 480443440 bytes.
Fixing the first problem requires introducing a bms_equal() call into
afterTriggerAddEvent's scanning loop, which is slightly annoying from
a speed perspective. However, getting rid of the excessive bms_copy()
calls from the second problem balances that out; overall speed of
trigger operations is the same or slightly better, in my tests.
Discussion: https://postgr.es/m/3496294.1737501591@sss.pgh.pa.us
Backpatch-through: 13
It's possible that external code is calling smgrtruncate(). Any
external callers might like to consider the recent changes to
RelationTruncate(), but commit 38c579b0 should not have changed the
function prototype in the back-branches, per ABI stability policy.
Restore smgrtruncate()'s traditional argument list in the back-branches,
but make it a wrapper for a new function smgrtruncate2(). The three
callers in core can use smgrtruncate2() directly. In master (18-to-be),
smgrtruncate2() is effectively renamed to smgrtruncate(), so this wart
is cleaned up.
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Discussion: https://postgr.es/m/CA%2BhUKG%2BThae6x6%2BjmQiuALQBT2Ae1ChjMh1%3DkMvJ8y_SBJZrvA%40mail.gmail.com
Despite the best efforts of commit 0e5c82380, we're still seeing
occasional failures of postgres_fdw's query_cancel test in the
buildfarm. Investigation suggests that its 100ms timeout is
still not enough to reliably ensure that the remote side starts
the query before receiving the cancel request --- and if it
hasn't, it will just discard the request because it's idle.
We discussed allowing a cancel request to kill the next-received
query, but that would have wide and perhaps unpleasant side-effects.
What seems safer is to make postgres_fdw do what a human user would
likely do, which is issue another cancel request if the first one
didn't seem to do anything. We'll keep the same overall 30 second
grace period before concluding things are broken, but issue additional
cancel requests after 1 second, then 2 more seconds, then 4, then 8.
(The next one in series is 16 seconds, but we'll hit the 30 second
timeout before that.)
Having done that, revert the timeout in query_cancel.sql to 10 ms.
That will still be enough on most machines, most of the time, for
the remote query to start; but now we're intentionally risking the
race condition occurring sometimes in the buildfarm, so that the
repeat-cancel code path will get some testing.
As before, back-patch to v17. We might eventually contemplate
back-patching this further, and/or adding similar logic to dblink.
But given the lack of field complaints to date, this feels like
mostly an exercise in test case stabilization, so v17 is enough.
Discussion: https://postgr.es/m/colnv3lzzmc53iu5qoawynr6qq7etn47lmggqr65ddtpjliq5d@glkveb4m6nop
RelationTruncate() does three things, while holding an
AccessExclusiveLock and preventing checkpoints:
1. Logs the truncation.
2. Drops buffers, even if they're dirty.
3. Truncates some number of files.
Step 2 could previously be canceled if it had to wait for I/O, and step
3 could and still can fail in file APIs. All orderings of these
operations have data corruption hazards if interrupted, so we can't give
up until the whole operation is done. When dirty pages were discarded
but the corresponding blocks were left on disk due to ERROR, old page
versions could come back from disk, reviving deleted data (see
pgsql-bugs #18146 and several like it). When primary and standby were
allowed to disagree on relation size, standbys could panic (see
pgsql-bugs #18426) or revive data unknown to visibility management on
the primary (theorized).
Changes:
* WAL is now unconditionally flushed first
* smgrtruncate() is now called in a critical section, preventing
interrupts and causing PANIC on file API failure
* smgrtruncate() has a new parameter for existing fork sizes,
because it can't call smgrnblocks() itself inside a critical section
The changes apply to RelationTruncate(), smgr_redo() and
pg_truncate_visibility_map(). That last is also brought up to date with
other evolutions of the truncation protocol.
The VACUUM FileTruncate() failure mode had been discussed in older
reports than the ones referenced below, with independent analysis from
many people, but earlier theories on how to fix it were too complicated
to back-patch. The more recently invented cancellation bug was
diagnosed by Alexander Lakhin. Other corruption scenarios were spotted
by me while iterating on this patch and earlier commit 75818b3a.
Back-patch to all supported releases.
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Reviewed-by: Robert Haas <robertmhaas@gmail.com>
Reported-by: rootcause000@gmail.com
Reported-by: Alexander Lakhin <exclusion@gmail.com>
Discussion: https://postgr.es/m/18146-04e908c662113ad5%40postgresql.org
Discussion: https://postgr.es/m/18426-2d18da6586f152d6%40postgresql.org
Commit dae761a87e modified brin_page_items() to return the new "empty"
flag for each BRIN range. But the new output parameter was added in the
middle, which may cause crashes when using the new binary with old
function definition.
The ideal solution would be to introduce API versioning similar to what
pg_stat_statements does, but it's too late for that as PG17 was already
released (so we can't introduce a new extension version). We could do
something similar in brin_page_items() by checking the number of output
columns (and ignoring the new flag), but it doesn't seem very nice.
Instead, simply error out and suggest updating the extension to the
latest version. pageinspect is a superuser-only extension, and there's
not much reason to run an older version. Moreover, there's a precedent
for this approach in 691e8b2e18.
Reported by Ľuboslav Špilák, investigation and patch by me. Backpatch to
17, same as dae761a87e.
Reported-by: Ľuboslav Špilák
Reviewed-by: Michael Paquier, Hayato Kuroda, Peter Geoghegan
Backpatch-through: 17
Discussion: https://postgr.es/m/VI1PR02MB63331C3D90E2104FD12399D38A5D2@VI1PR02MB6333.eurprd02.prod.outlook.com
Discussion: https://postgr.es/m/flat/3385a58f-5484-49d0-b790-9a198a0bf236@vondra.me
The @extschema:name@ feature added by 72a5b1fc8 allows us to
make earthdistance's references to the cube extension fully
search-path-secure, so long as all those references are
resolved at extension installation time not runtime.
To do that, we must convert earthdistance's SQL functions to
the new SQL-standard style; but we wanted to do that anyway.
The functions can be updated in our customary style by running
CREATE OR REPLACE FUNCTION in an extension update script.
However, there's still problems in the "CREATE DOMAIN earth"
command: its references to cube functions could be captured
by hostile objects in earthdistance's installation schema,
if that's not where the cube extension is. Worse, the reference
to the cube type itself as the domain's base could be captured,
and that's not something we could fix after-the-fact in the
update script.
What I've done about that is to change the "CREATE DOMAIN earth"
command in the base script earthdistance--1.1.sql. Ordinarily,
changing a released extension script is forbidden; but I think
it's okay here since the results of successful (non-trojaned)
script execution will be identical to before.
A good deal of care is still needed to make the extension's scripts
proof against search-path-based attacks. We have to make sure that
all the function and operator invocations have exact argument-type
matches, to forestall attacks based on supplying a better match.
Fortunately earthdistance isn't very big, so I've just gone through
it and inspected each call to be sure of that. The only actual code
changes needed were to spell all floating-point constants in the style
'-1'::float8, rather than depending on runtime type conversions and/or
negations. (I'm not sure that the shortcuts previously used were
attackable, but removing run-time effort is a good thing anyway.)
I believe that this fixes earthdistance enough that we could
mark it trusted and remove the warnings about it that were
added by 7eeb1d986; but I've not done that here.
The primary reason for dealing with this now is that we've
received reports of pg_upgrade failing for databases that use
earthdistance functions in contexts like generated columns.
That's a consequence of 2af07e2f7 having restricted the search_path
used while evaluating such expressions. The only way to fix that
is to make the earthdistance functions independent of run-time
search_path. This patch is very much nicer than the alternative of
attaching "SET search_path" clauses to earthdistance's functions:
it is more secure and doesn't create a run-time penalty. Therefore,
I've chosen to back-patch this to v16 where @extschema:name@
was added. It won't help unless users update to 16.7 and issue
"ALTER EXTENSION earthdistance UPDATE" before upgrading to 17,
but at least there's now a way to deal with the problem without
manual intervention in the dump/restore process.
Tom Lane and Ronan Dunklau
Discussion: https://postgr.es/m/3316564.aeNJFYEL58@aivenlaptop
Discussion: https://postgr.es/m/6a6439f1-8039-44e2-8fb9-59028f7f2014@mailbox.org
818119afcc has introduced the "generation" concept in pgstats entries,
incremented a counter when a pgstats entry is reinitialized, but it did
not count on the fact that backends still holding local references to
such entries need to be refreshed if the cache age is outdated. The
previous logic only updated local references when an entry was dropped,
but it needs also to consider entries that are reinitialized.
This matters for replication slot stats (as well as custom pgstats kinds
in 18~), where concurrent drops and creates of a slot could cause
incorrect stats to be locally referenced. This would lead to an
assertion failure at shutdown when writing out the stats file, as the
backend holding an outdated local reference would not be able to drop
during its shutdown sequence the stats entry that should be dropped, as
the last process holding a reference to the stats entry. The
checkpointer was then complaining about such an entry late in the
shutdown sequence, after the shutdown checkpoint is finished with the
control file updated, causing the stats file to not be generated. In
non-assert builds, the entry would just be skipped with the stats file
written.
Note that only logical replication slots use statistics.
A test case based on TAP is added to test_decoding, where a persistent
connection peeking at a slot's data is kept with concurrent drops and
creates of the same slot. This is based on the isolation test case that
Anton has sent. As it requires a node shutdown with a check to make
sure that the stats file is written with this specific sequence of
events, TAP is used instead.
Reported-by: Anton A. Melnikov
Reviewed-by: Bertrand Drouvot
Discussion: https://postgr.es/m/56bf8ff9-dd8c-47b2-872a-748ede82af99@postgrespro.ru
Backpatch-through: 15
Most came in during the 17 cycle, so backpatch there. Some
(particularly reorderbuffer.h) are very old, but backpatching doesn't
seem useful.
Like commits c9d2977519, c4f113e8fe.
The current code calls array_eq() and does not provide FmgrInfo. This commit
provides initialization of FmgrInfo and uses C collation as the safe option
for text comparison because we don't know anything about the semantics of
opclass options.
Backpatch to 13, where opclass options were introduced.
Reported-by: Nicolas Maus
Discussion: https://postgr.es/m/18692-72ea398df3ec6712%40postgresql.org
Backpatch-through: 13
The test was unstable in branches 14 and 15 as we were relying on the
number of changes in the table having a toast column to start streaming.
On branches >= 16, we have a GUC debug_logical_replication_streaming which
can stream each change, so the test was stable in those branches.
Change the test to use PREPARE TRANSACTION as that should make the result
consistent and test the code changed in 022564f60c.
Reported-by: Daniel Gustafsson as per buildfarm
Author: Hou Zhijie, Amit Kapila
Backpatch-through: 14
Discussion: https://postgr.es/m/8C2F86AA-981E-4803-B14D-E264C0255330@yesql.se
Several places treat MyStartTime as a "long", which is only 32 bits
wide on some platforms. In reality, MyStartTime is a pg_time_t,
i.e., a signed 64-bit integer. This will lead to interesting bugs
on the aforementioned systems in 2038 when signed 32-bit integers
are no longer sufficient to store Unix time (e.g., "pg_ctl start"
hanging). To fix, ensure that MyStartTime is handled as a 64-bit
value everywhere. (Of course, users will need to ensure that
time_t is 64 bits wide on their system, too.)
Co-authored-by: Max Johnson
Discussion: https://postgr.es/m/CO1PR07MB905262E8AC270FAAACED66008D682%40CO1PR07MB9052.namprd07.prod.outlook.com
Backpatch-through: 12
During logical decoding of in-progress transactions, we perform the toast
table scan while fetching the default toast value for an attribute. We
forgot to initialize the flag during this scan to indicate that the system
table scan is in progress. We need this flag to ensure that during logical
decoding we never directly access the tableam or heap APIs because we check
for concurrent aborts only in systable_* APIs.
Reported-by: Alexander Lakhin
Author: Takeshi Ideriha, Hou Zhijie
Reviewed-by: Amit Kapila, Hou Zhijie
Backpatch-through: 14
Discussion: https://postgr.es/m/18641-6687273b7f15269d@postgresql.org
When instantiating an existing partitioned index for a new child
partition, we use generateClonedIndexStmt to build a suitable
IndexStmt to pass to DefineIndex. However, when DefineIndex needs
to recurse to instantiate a newly created partitioned index on an
existing child partition, it was doing copyObject on the given
IndexStmt and then applying a bunch of ad-hoc fixups. This has
a number of problems, primarily that it implies fresh lookups of
referenced objects such as opclasses and collations. Since commit
2af07e2f7 caused DefineIndex to restrict search_path internally, those
lookups could fail or deliver different results than the original one.
We can avoid those problems and save a few dozen lines of code by
using generateClonedIndexStmt in this code path too.
Another thing this fixes is incorrect propagation of parent-index
comments to child indexes (because the copyObject approach copies
the idxcomment field while generateClonedIndexStmt doesn't). I had
noticed this in connection with commit c01eb619a, but not run the
problem to ground.
I'm tempted to back-patch this further than v17, but the only thing
it's known to fix in older branches is the comment issue, which is
pretty minor and doesn't seem worth the risk of introducing new
issues in stable branches. (If anyone does care about that,
clearing idxcomment in the copied IndexStmt would be a safer fix.)
Per bug #18637 from usamoi. Back-patch to v17 where the search_path
change came in.
Discussion: https://postgr.es/m/18637-f51e314546e3ba2a@postgresql.org
This commit adds query ID reports for two code paths when processing
extended query protocol messages:
- When receiving a bind message, setting it to the first Query retrieved
from a cached cache.
- When receiving an execute message, setting it to the first PlannedStmt
stored in a portal.
An advantage of this method is that this is able to cover all the types
of portals handled in the extended query protocol, particularly these
two when the report done in ExecutorStart() is not enough (neither is an
addition in ExecutorRun(), actually, for the second point):
- Multiple execute messages, with multiple ExecutorRun().
- Portal with execute/fetch messages, like a query with a RETURNING
clause and a fetch size that stores the tuples in a first execute
message going though ExecutorStart() and ExecuteRun(), followed by one
or more execute messages doing only fetches from the tuplestore created
in the first message. This corresponds to the case where
execute_is_fetch is set, for example.
Note that the query ID reporting done in ExecutorStart() is still
necessary, as an EXECUTE requires it. Query ID reporting is optimistic
and more calls to pgstat_report_query_id() don't matter as the first
report takes priority except if the report is forced. The comment in
ExecutorStart() is adjusted to reflect better the reality with the
extended query protocol.
The test added in pg_stat_statements is a courtesy of Robert Haas. This
uses psql's \bind metacommand, hence this part is backpatched down to
v16.
Reported-by: Kaido Vaikla, Erik Wienhold
Author: Sami Imseih
Reviewed-by: Jian He, Andrei Lepikhov, Michael Paquier
Discussion: https://postgr.es/m/CA+427g8DiW3aZ6pOpVgkPbqK97ouBdf18VLiHFesea2jUk3XoQ@mail.gmail.com
Discussion: https://postgr.es/m/CA+TgmoZxtnf_jZ=VqBSyaU8hfUkkwoJCJ6ufy4LGpXaunKrjrg@mail.gmail.com
Discussion: https://postgr.es/m/1391613709.939460.1684777418070@office.mailbox.org
Backpatch-through: 14
In existing releases of libxml2, xmlXPathCompile can be driven
to stack overflow because it fails to protect itself against
too-deeply-nested input. While there is an upstream fix as of
yesterday, it will take years for that to propagate into all
shipping versions. In the meantime, we can protect our own
usages basically for free by calling xmlXPathCtxtCompile instead.
(The actual bug is that libxml2 keeps its nesting counter in the
xmlXPathContext, and its parsing code was willing to just skip
counting nesting levels if it didn't have a context. So if we supply
a context, all is well. It seems odd actually that it works at all
to not supply a context, because this means that XPath parsing does
not have access to XML namespace info. Apparently libxml2 never
checks namespaces until runtime? Anyway, this seems like good
future-proofing even if its only immediate effect is to dodge a bug.)
Sadly, this hack only offers protection with libxml2 2.9.11 and newer.
Before that there are multiple similar problems, so if you are
processing untrusted XML it behooves you to get a newer version.
But we have some pretty old libxml2 in the buildfarm, so it seems
impractical to add a regression test to verify this fix.
Per bug #18617 from Jingzhou Fu. Back-patch to all supported
versions.
Discussion: https://postgr.es/m/18617-1cee4d2ed1f4e7ae@postgresql.org
Discussion: https://gitlab.gnome.org/GNOME/libxml2/-/issues/799
I managed to break this test in two different ways in commit
05036a3155.
First, the output of the new call to tuple_data_split() on the test
sequence is dependent on endianness. This is fixed by setting a
special start value for the test sequence that produces the same
output regardless of the endianness of the machine.
Second, on versions older than v15, the new test case fails under
"force_parallel_mode = regress" with the following error:
ERROR: cannot access temporary tables during a parallel operation
This is because pageinspect's disk-accessing functions are
incorrectly marked PARALLEL SAFE on versions older than v15 (see
commit aeaaf520f4 for details). This one is fixed by changing the
test sequence to be permanent. The only reason it was previously
marked temporary was to avoid needing a DROP SEQUENCE command at
the end of the test. Unlike some other tests in this file, the use
of a permanent sequence here shouldn't result in any test
instability like what was fixed by commit e2933a6e11.
Reviewed-by: Tom Lane
Discussion: https://postgr.es/m/ZuOKOut5hhDlf_bP%40nathan
Backpatch-through: 12
Commit 4b82664156 restricted a number of functions provided by
contrib modules to only relations that use the "heap" table access
method. Sequences always use this table access method, but they do
not advertise as such in the pg_class system catalog, so the
aforementioned commit also (presumably unintentionally) removed
support for sequences from some of these functions. This commit
reintroduces said support for sequences to these functions and adds
a couple of relevant tests.
Co-authored-by: Ayush Vatsa
Reviewed-by: Robert Haas, Michael Paquier, Matthias van de Meent
Discussion: https://postgr.es/m/CACX%2BKaP3i%2Bi9tdPLjF5JCHVv93xobEdcd_eB%2B638VDvZ3i%3DcQA%40mail.gmail.com
Backpatch-through: 12
This test occasionally shows
+WARNING: could not get result of cancel request due to timeout
which appears to be because the cancel request is sometimes unluckily
sent to the remote session between queries, and then it's ignored.
This patch tries to make that less probable in three ways:
1. Use a test query that does not involve remote estimates, so that
no EXPLAINs are sent.
2. Make sure that the remote session is ready-to-go (transaction
started, SET commands sent) before we start the timer.
3. Increase the statement_timeout to 100ms, to give the local
session enough time to plan and issue the query.
We might have to go higher than 100ms to make this adequately
stable in the buildfarm, but let's see how it goes.
Back-patch to v17 where this test was introduced.
Jelte Fennema-Nio and Tom Lane
Discussion: https://postgr.es/m/578934.1725045685@sss.pgh.pa.us
