dshash.c previously maintained flags to be able to assert that you
didn't hold any partition lock. These flags could get out of sync with
reality in error scenarios.
Get rid of all that, and make assertions about the locks themselves
instead. Since LWLockHeldByMe() loops internally, we don't want to put
that inside another loop over all partition locks. Introduce a new
debugging-only interface LWLockAnyHeldByMe() to avoid that.
This problem was noted by Tom and Andres while reviewing changes to
support the new shared memory stats system, and later showed up in
reality while working on commit 389869af.
Back-patch to 11, where dshash.c arrived.
Reported-by: Tom Lane <tgl@sss.pgh.pa.us>
Reported-by: Andres Freund <andres@anarazel.de>
Reviewed-by: Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>
Reviewed-by: Zhihong Yu <zyu@yugabyte.com>
Reviewed-by: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/20220311012712.botrpsikaufzteyt@alap3.anarazel.de
Discussion: https://postgr.es/m/CA%2BhUKGJ31Wce6HJ7xnVTKWjFUWQZPBngxfJVx4q0E98pDr3kAw%40mail.gmail.com
SPI_commit previously left it up to the caller to recover from any error
occurring during commit. Since that's complicated and requires use of
low-level xact.c facilities, it's not too surprising that no caller got
it right. Let's move the responsibility for cleanup into spi.c. Doing
that requires redefining SPI_commit as starting a new transaction, so
that it becomes equivalent to SPI_commit_and_chain except that you get
default transaction characteristics instead of preserving the prior
transaction's characteristics. We can make this pretty transparent
API-wise by redefining SPI_start_transaction() as a no-op. Callers
that expect to do something in between might be surprised, but
available evidence is that no callers do so.
Having made that API redefinition, we can fix this mess by having
SPI_commit[_and_chain] trap errors and start a new, clean transaction
before re-throwing the error. Likewise for SPI_rollback[_and_chain].
Some cleanup is also needed in AtEOXact_SPI, which was nowhere near
smart enough to deal with SPI contexts nested inside a committing
context.
While plperl and pltcl need no changes beyond removing their now-useless
SPI_start_transaction() calls, plpython needs some more work because it
hadn't gotten the memo about catching commit/rollback errors in the
first place. Such an error resulted in longjmp'ing out of the Python
interpreter, which leaks Python stack entries at present and is reported
to crash Python 3.11 altogether. Add the missing logic to catch such
errors and convert them into Python exceptions.
This is a back-patch of commit 2e517818f. That's now aged long enough
to reduce the concerns about whether it will break something, and we
do need to ensure that supported branches will work with Python 3.11.
Peter Eisentraut and Tom Lane
Discussion: https://postgr.es/m/3375ffd8-d71c-2565-e348-a597d6e739e3@enterprisedb.com
Discussion: https://postgr.es/m/17416-ed8fe5d7213d6c25@postgresql.org
We were not updating the partition map cache in the subscriber even when
the corresponding remote rel is changed. Due to this data was getting
incorrectly replicated for partition tables after the publisher has
changed the table schema.
Fix it by resetting the required entries in the partition map cache after
receiving a new relation mapping from the publisher.
Reported-by: Shi Yu
Author: Shi Yu, Hou Zhijie
Reviewed-by: Amit Langote, Amit Kapila
Backpatch-through: 13, where it was introduced
Discussion: https://postgr.es/m/OSZPR01MB6310F46CD425A967E4AEF736FDA49@OSZPR01MB6310.jpnprd01.prod.outlook.com
This is a slight, convenient semantics change from what commit
0f0cfb494004 ("Fix parallel operations that prevent oldest xmin from
advancing") introduced that lets us simplify the coding in the one place
where it is used.
Backpatch to 13. This is related to commit 6fea65508a1a ("Tighten
ComputeXidHorizons' handling of walsenders") rewriting the code site
where this is used, which has not yet been backpatched, but it may well
be in the future.
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Discussion: https://postgr.es/m/202204191637.eldwa2exvguw@alvherre.pgsql
The problem is that we don't send keep-alive messages for a long time
while processing large transactions during logical replication where we
don't send any data of such transactions. This can happen when the table
modified in the transaction is not published or because all the changes
got filtered. We do try to send the keep_alive if necessary at the end of
the transaction (via WalSndWriteData()) but by that time the
subscriber-side can timeout and exit.
To fix this we try to send the keepalive message if required after
processing certain threshold of changes.
Reported-by: Fabrice Chapuis
Author: Wang wei and Amit Kapila
Reviewed By: Masahiko Sawada, Euler Taveira, Hou Zhijie, Hayato Kuroda
Backpatch-through: 10
Discussion: https://postgr.es/m/CAA5-nLARN7-3SLU_QUxfy510pmrYK6JJb=bk3hcgemAM_pAv+w@mail.gmail.com
inline_cte() expected to find exactly as many references to the
target CTE as its cterefcount indicates. While that should be
accurate for the tree as emitted by the parser, there are some
optimizations that occur upstream of here that could falsify it,
notably removal of unused subquery output expressions.
Trying to make the accounting 100% accurate seems expensive and
doomed to future breakage. It's not really worth it, because
all this code is protecting is downstream assumptions that every
referenced CTE has a plan. Let's convert those assertions to
regular test-and-elog just in case there's some actual problem,
and then drop the failing assertion.
Per report from Tomas Vondra (thanks also to Richard Guo for
analysis). Back-patch to v12 where the faulty code came in.
Discussion: https://postgr.es/m/29196a1e-ed47-c7ca-9be2-b1c636816183@enterprisedb.com
The back-patch of commit bbace5697df12398e87ffd9879171c39d27f5b33 had
the unfortunate effect of changing the layout of PGPROC in the
back-branches, which could break extensions. This happened because it
changed the delayChkpt from type bool to type int. So, change it back,
and add a new bool delayChkptEnd field instead. The new field should
fall within what used to be padding space within the struct, and so
hopefully won't cause any extensions to break.
Per report from Markus Wanner and discussion with Tom Lane and others.
Patch originally by me, somewhat revised by Markus Wanner per a
suggestion from Michael Paquier. A very similar patch was developed
by Kyotaro Horiguchi, but I failed to see the email in which that was
posted before writing one of my own.
Discussion: http://postgr.es/m/CA+Tgmoao-kUD9c5nG5sub3F7tbo39+cdr8jKaOVEs_1aBWcJ3Q@mail.gmail.com
Discussion: http://postgr.es/m/20220406.164521.17171257901083417.horikyota.ntt@gmail.com
heap_fetch() used to have a "keep_buf" parameter that told it to return
ownership of the buffer pin to the caller after finding that the
requested tuple TID exists but is invisible to the specified snapshot.
This was thoughtlessly removed in commit 5db6df0c0, which broke
heapam_tuple_lock() (formerly EvalPlanQualFetch) because that function
needs to do more accesses to the tuple even if it's invisible. The net
effect is that we would continue to touch the page for a microsecond or
two after releasing pin on the buffer. Usually no harm would result;
but if a different session decided to defragment the page concurrently,
we could see garbage data and mistakenly conclude that there's no newer
tuple version to chain up to. (It's hard to say whether this has
happened in the field. The bug was actually found thanks to a later
change that allowed valgrind to detect accesses to non-pinned buffers.)
The most reasonable way to fix this is to reintroduce keep_buf,
although I made it behave slightly differently: buffer ownership
is passed back only if there is a valid tuple at the requested TID.
In HEAD, we can just add the parameter back to heap_fetch().
To avoid an API break in the back branches, introduce an additional
function heap_fetch_extended() in those branches.
In HEAD there is an additional, less obvious API change: tuple->t_data
will be set to NULL in all cases where buffer ownership is not returned,
in particular when the tuple exists but fails the time qual (and
!keep_buf). This is to defend against any other callers attempting to
access non-pinned buffers. We concluded that making that change in back
branches would be more likely to introduce problems than cure any.
In passing, remove a comment about heap_fetch that was obsoleted by
9a8ee1dc6.
Per bug #17462 from Daniil Anisimov. Back-patch to v12 where the bug
was introduced.
Discussion: https://postgr.es/m/17462-9c98a0f00df9bd36@postgresql.org
clang 13 with -Wextra warns that "performing pointer subtraction with
a null pointer has undefined behavior" in the places where freepage.c
tries to set a relptr variable to constant NULL. This appears to be
a compiler bug, but it's unlikely to get fixed instantly. Fortunately,
we can work around it by introducing an inline support function, which
seems like a good change anyway because it removes the macro's existing
double-evaluation hazard.
Backpatch to v10 where this code was introduced.
Patch by me, based on an idea of Andres Freund's.
Discussion: https://postgr.es/m/48826.1648310694@sss.pgh.pa.us
Crash recovery on standby may encounter missing directories when
replaying create database WAL records. Prior to this patch, the standby
would fail to recover in such a case. However, the directories could be
legitimately missing. Consider a sequence of WAL records as follows:
CREATE DATABASE
DROP DATABASE
DROP TABLESPACE
If, after replaying the last WAL record and removing the tablespace
directory, the standby crashes and has to replay the create database
record again, the crash recovery must be able to move on.
This patch adds a mechanism similar to invalid-page tracking, to keep a
tally of missing directories during crash recovery. If all the missing
directory references are matched with corresponding drop records at the
end of crash recovery, the standby can safely continue following the
primary.
Backpatch to 13, at least for now. The bug is older, but fixing it in
older branches requires more careful study of the interactions with
commit e6d8069522c8, which appeared in 13.
A new TAP test file is added to verify the condition. However, because
it depends on commit d6d317dbf615, it can only be added to branch
master. I (Álvaro) manually verified that the code behaves as expected
in branch 14. It's a bit nervous-making to leave the code uncovered by
tests in older branches, but leaving the bug unfixed is even worse.
Also, the main reason this fix took so long is precisely that we
couldn't agree on a good strategy to approach testing for the bug, so
perhaps this is the best we can do.
Diagnosed-by: Paul Guo <paulguo@gmail.com>
Author: Paul Guo <paulguo@gmail.com>
Author: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Author: Asim R Praveen <apraveen@pivotal.io>
Discussion: https://postgr.es/m/CAEET0ZGx9AvioViLf7nbR_8tH9-=27DN5xWJ2P9-ROH16e4JUA@mail.gmail.com
If TRUNCATE causes some buffers to be invalidated and thus the
checkpoint does not flush them, TRUNCATE must also ensure that the
corresponding files are truncated on disk. Otherwise, a replay
from the checkpoint might find that the buffers exist but have
the wrong contents, which may cause replay to fail.
Report by Teja Mupparti. Patch by Kyotaro Horiguchi, per a design
suggestion from Heikki Linnakangas, with some changes to the
comments by me. Review of this and a prior patch that approached
the issue differently by Heikki Linnakangas, Andres Freund, Álvaro
Herrera, Masahiko Sawada, and Tom Lane.
Discussion: http://postgr.es/m/BYAPR06MB6373BF50B469CA393C614257ABF00@BYAPR06MB6373.namprd06.prod.outlook.com
If we run out of space in the checkpointer sync request queue (which is
hopefully rare on real systems, but common with very small buffer pool),
we wait for it to drain. While waiting, we should report that as a wait
event so that users know what is going on, and also handle postmaster
death, since otherwise the loop might never terminate if the
checkpointer has exited.
Back-patch to 12. Although the problem exists in earlier releases too,
the code is structured differently before 12 so I haven't gone any
further for now, in the absence of field complaints.
Reported-by: Andres Freund <andres@anarazel.de>
Reviewed-by: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/20220226213942.nb7uvb2pamyu26dj%40alap3.anarazel.de
This macro cast the result to BlockNumber after shifting, not before,
which is the wrong thing. Per the C spec, the uint16 fields would
promote to int not unsigned int, so that (for 32-bit int) the shift
potentially shifts a nonzero bit into the sign position. I doubt
there are any production systems where this would actually end with
the wrong answer, but it is undefined behavior per the C spec, and
clang's -fsanitize=undefined option reputedly warns about it on some
platforms. (I can't reproduce that right now, but the code is
undeniably wrong per spec.) It's easy to fix by casting to
BlockNumber (uint32) in the proper places.
It's been wrong for ages, so back-patch to all supported branches.
Report and patch by Zhihong Yu (cosmetic tweaking by me)
Discussion: https://postgr.es/m/CALNJ-vT9r0DSsAOw9OXVJFxLENoVS_68kJ5x0p44atoYH+H4dg@mail.gmail.com
GCC 12 complains that set_stack_base is storing the address of
a local variable in a long-lived pointer. This is an entirely
reasonable warning (indeed, it just helped us find a bug);
but that behavior is intentional here. We can work around it
by using __builtin_frame_address(0) instead of a specific local
variable; that produces an address a dozen or so bytes different,
in my testing, but we don't care about such a small difference.
Maybe someday a compiler lacking that function will start to issue
a similar warning, but we'll worry about that when it happens.
Patch by me, per a suggestion from Andres Freund. Back-patch to
v12, which is as far back as the patch will go without some pain.
(Recently-established project policy would permit a back-patch as
far as 9.2, but I'm disinclined to expend the work until GCC 12
is much more widespread.)
Discussion: https://postgr.es/m/3773792.1645141467@sss.pgh.pa.us
Although select_common_type() has a failure-return convention, an
apparent successful return just provides a type OID that *might* work
as a common supertype; we've not validated that the required casts
actually exist. In the mainstream use-cases that doesn't matter,
because we'll proceed to invoke coerce_to_common_type() on each input,
which will fail appropriately if the proposed common type doesn't
actually work. However, a few callers didn't read the (nonexistent)
fine print, and thought that if they got back a nonzero OID then the
coercions were sure to work.
This affects in particular the recently-added "anycompatible"
polymorphic types; we might think that a function/operator using
such types matches cases it really doesn't. A likely end result
of that is unexpected "ambiguous operator" errors, as for example
in bug #17387 from James Inform. Another, much older, case is that
the parser might try to transform an "x IN (list)" construct to
a ScalarArrayOpExpr even when the list elements don't actually have
a common supertype.
It doesn't seem desirable to add more checking to select_common_type
itself, as that'd just slow down the mainstream use-cases. Instead,
write a separate function verify_common_type that performs the
missing checks, and add a call to that where necessary. Likewise add
verify_common_type_from_oids to go with select_common_type_from_oids.
Back-patch to v13 where the "anycompatible" types came in. (The
symptom complained of in bug #17387 doesn't appear till v14, but
that's just because we didn't get around to converting || to use
anycompatible till then.) In principle the "x IN (list)" fix could
go back all the way, but I'm not currently convinced that it makes
much difference in real-world cases, so I won't bother for now.
Discussion: https://postgr.es/m/17387-5dfe54b988444963@postgresql.org
Commit 8431e296ea reworked ProcArrayApplyRecoveryInfo to sort XIDs
before adding them to KnownAssignedXids. But the XIDs are sorted using
xidComparator, which compares the XIDs simply as uint32 values, not
logically. KnownAssignedXidsAdd() however expects XIDs in logical order,
and calls TransactionIdFollowsOrEquals() to enforce that. If there are
XIDs for which the two orderings disagree, an error is raised and the
recovery fails/restarts.
Hitting this issue is fairly easy - you just need two transactions, one
started before the 4B limit (e.g. XID 4294967290), the other sometime
after it (e.g. XID 1000). Logically (4294967290 <= 1000) but when
compared using xidComparator we try to add them in the opposite order.
Which makes KnownAssignedXidsAdd() fail with an error like this:
ERROR: out-of-order XID insertion in KnownAssignedXids
This only happens during replica startup, while processing RUNNING_XACTS
records to build the snapshot. Once we reach STANDBY_SNAPSHOT_READY, we
skip these records. So this does not affect already running replicas,
but if you restart (or create) a replica while there are transactions
with XIDs for which the two orderings disagree, you may hit this.
Long-running transactions and frequent replica restarts increase the
likelihood of hitting this issue. Once the replica gets into this state,
it can't be started (even if the old transactions are terminated).
Fixed by sorting the XIDs logically - this is fine because we're dealing
with normal XIDs (because it's XIDs assigned to backends) and from the
same wraparound epoch (otherwise the backends could not be running at
the same time on the primary node). So there are no problems with the
triangle inequality, which is why xidComparator compares raw values.
Investigation and root cause analysis by Abhijit Menon-Sen. Patch by me.
This issue is present in all releases since 9.4, however releases up to
9.6 are EOL already so backpatch to 10 only.
Reviewed-by: Abhijit Menon-Sen
Reviewed-by: Alvaro Herrera
Backpatch-through: 10
Discussion: https://postgr.es/m/36b8a501-5d73-277c-4972-f58a4dce088a%40enterprisedb.com
In logical replication mode, a WalSender is supposed to be able
to execute any regular SQL command, as well as the special
replication commands. Poor design of the replication-command
parser caused it to fail in various cases, notably:
* semicolons embedded in a command, or multiple SQL commands
sent in a single message;
* dollar-quoted literals containing odd numbers of single
or double quote marks;
* commands starting with a comment.
The basic problem here is that we're trying to run repl_scanner.l
across the entire input string even when it's not a replication
command. Since repl_scanner.l does not understand all of the
token types known to the core lexer, this is doomed to have
failure modes.
We certainly don't want to make repl_scanner.l as big as scan.l,
so instead rejigger stuff so that we only lex the first token of
a non-replication command. That will usually look like an IDENT
to repl_scanner.l, though a comment would end up getting reported
as a '-' or '/' single-character token. If the token is a replication
command keyword, we push it back and proceed normally with repl_gram.y
parsing. Otherwise, we can drop out of exec_replication_command()
without examining the rest of the string.
(It's still theoretically possible for repl_scanner.l to fail on
the first token; but that could only happen if it's an unterminated
single- or double-quoted string, in which case you'd have gotten
largely the same error from the core lexer too.)
In this way, repl_gram.y isn't involved at all in handling general
SQL commands, so we can get rid of the SQLCmd node type. (In
the back branches, we can't remove it because renumbering enum
NodeTag would be an ABI break; so just leave it sit there unused.)
I failed to resist the temptation to clean up some other sloppy
coding in repl_scanner.l while at it. The only externally-visible
behavior change from that is it now accepts \r and \f as whitespace,
same as the core lexer.
Per bug #17379 from Greg Rychlewski. Back-patch to all supported
branches.
Discussion: https://postgr.es/m/17379-6a5c6cfb3f1f5e77@postgresql.org
Commit 4ace45677 failed to fix the problem fully, because the
same issue of attempting to fetch a non-returnable index column
can occur when rechecking the indexqual after using a lossy index
operator. Moreover, it broke EXPLAIN for such indexquals (which
indicates a gap in our test cases :-().
Revert the code changes of 4ace45677 in favor of adding a new field
to struct IndexOnlyScan, containing a version of the indexqual that
can be executed against the index-returned tuple without using any
non-returnable columns. (The restrictions imposed by check_index_only
guarantee this is possible, although we may have to recompute indexed
expressions.) Support construction of that during setrefs.c
processing by marking IndexOnlyScan.indextlist entries as resjunk
if they can't be returned, rather than removing them entirely.
(We could alternatively require setrefs.c to look up the IndexOptInfo
again, but abusing resjunk this way seems like a reasonably safe way
to avoid needing to do that.)
This solution isn't great from an API-stability standpoint: if there
are any extensions out there that build IndexOnlyScan structs directly,
they'll be broken in the next minor releases. However, only a very
invasive extension would be likely to do such a thing. There's no
change in the Path representation, so typical planner extensions
shouldn't have a problem.
As before, back-patch to all supported branches.
Discussion: https://postgr.es/m/3179992.1641150853@sss.pgh.pa.us
Discussion: https://postgr.es/m/17350-b5bdcf476e5badbb@postgresql.org
If an index has both returnable and non-returnable columns, and one of
the non-returnable columns is an expression using a Var that is in a
returnable column, then a query returning that expression could result
in an index-only scan plan that attempts to read the non-returnable
column, instead of recomputing the expression from the returnable
column as intended.
To fix, redefine the "indextlist" list of an IndexOnlyScan plan node
as containing null Consts in place of any non-returnable columns.
This solves the problem by preventing setrefs.c from falsely matching
to such entries. The executor is happy since it only cares about the
exposed types of the entries, and ruleutils.c doesn't care because a
correct plan won't reference those entries. I considered some other
ways to prevent setrefs.c from doing the wrong thing, but this way
seems good since (a) it allows a very localized fix, (b) it makes
the indextlist structure more compact in many cases, and (c) the
indextlist is now a more faithful representation of what the index AM
will actually produce, viz. nulls for any non-returnable columns.
This is easier to hit since we introduced included columns, but it's
possible to construct failing examples without that, as per the
added regression test. Hence, back-patch to all supported branches.
Per bug #17350 from Louis Jachiet.
Discussion: https://postgr.es/m/17350-b5bdcf476e5badbb@postgresql.org
catalog/pg_class.h was stating that REPLICA_IDENTITY_INDEX with a
dropped index is equivalent to REPLICA_IDENTITY_DEFAULT. The code tells
a different story, as it is equivalent to REPLICA_IDENTITY_NOTHING.
The behavior exists since the introduction of replica identities, and
fe7fd4e even added tests for this case but I somewhat forgot to fix this
comment.
While on it, this commit reorganizes the documentation about replica
identities on the ALTER TABLE page, and a note is added about the case
of dropped indexes with REPLICA_IDENTITY_INDEX.
Author: Michael Paquier, Wei Wang
Reviewed-by: Euler Taveira
Discussion: https://postgr.es/m/OS3PR01MB6275464AD0A681A0793F56879E759@OS3PR01MB6275.jpnprd01.prod.outlook.com
Backpatch-through: 10
Surround the contents with a test that the feature is enabled by
configure, to silence header checking tools on systems without GSSAPI
installed.
Backpatch to 12, where the file appeared.
Discussion: https://postgr.es/m/202111161709.u3pbx5lxdimt@alvherre.pgsql
While determining xid horizons, we skip over backends that are running
Vacuum. We also ignore Create Index Concurrently, or Reindex Concurrently
for the purposes of computing Xmin for Vacuum. But we were not setting the
flags corresponding to these operations when they are performed in
parallel which was preventing Xid horizon from advancing.
The optimization related to skipping Create Index Concurrently, or Reindex
Concurrently operations was implemented in PG-14 but the fix is the same
for the Parallel Vacuum as well so back-patched till PG-13.
Author: Masahiko Sawada
Reviewed-by: Amit Kapila
Backpatch-through: 13
Discussion: https://postgr.es/m/CAD21AoCLQqgM1sXh9BrDFq0uzd3RBFKi=Vfo6cjjKODm0Onr5w@mail.gmail.com
The server collects up to a bufferload of data whenever it reads data
from the client socket. When SSL or GSS encryption is requested
during startup, any additional data received with the initial
request message remained in the buffer, and would be treated as
already-decrypted data once the encryption handshake completed.
Thus, a man-in-the-middle with the ability to inject data into the
TCP connection could stuff some cleartext data into the start of
a supposedly encryption-protected database session.
This could be abused to send faked SQL commands to the server,
although that would only work if the server did not demand any
authentication data. (However, a server relying on SSL certificate
authentication might well not do so.)
To fix, throw a protocol-violation error if the internal buffer
is not empty after the encryption handshake.
Our thanks to Jacob Champion for reporting this problem.
Security: CVE-2021-23214
As in commits 6301c3ada and e9d9ba2a4, avoid doing repetitive
list_delete_first() operations, since that would be expensive when
there are many files waiting to be unlinked. This is a slightly
larger change than in those cases. We have to keep the list state
valid for calls to AbsorbSyncRequests(), so it's necessary to invent a
"canceled" field instead of immediately deleting PendingUnlinkEntry
entries. Also, because we might not be able to process all the
entries, we need a new list primitive list_delete_first_n().
list_delete_first_n() is almost list_copy_tail(), but it modifies the
input List instead of making a new copy. I found a couple of existing
uses of the latter that could profitably use the new function. (There
might be more, but the other callers look like they probably shouldn't
overwrite the input List.)
As before, back-patch to v13.
Discussion: https://postgr.es/m/CD2F0E7F-9822-45EC-A411-AE56F14DEA9F@amazon.com
The purpose of commit 8a54e12a38d1545d249f1402f66c8cde2837d97c was to
fix this, and it sufficed when the PREPARE TRANSACTION completed before
the CIC looked for lock conflicts. Otherwise, things still broke. As
before, in a cluster having used CIC while having enabled prepared
transactions, queries that use the resulting index can silently fail to
find rows. It may be necessary to reindex to recover from past
occurrences; REINDEX CONCURRENTLY suffices. Fix this for future index
builds by making CIC wait for arbitrarily-recent prepared transactions
and for ordinary transactions that may yet PREPARE TRANSACTION. As part
of that, have PREPARE TRANSACTION transfer locks to its dummy PGPROC
before it calls ProcArrayClearTransaction(). Back-patch to 9.6 (all
supported versions).
Andrey Borodin, reviewed (in earlier versions) by Andres Freund.
Discussion: https://postgr.es/m/01824242-AA92-4FE9-9BA7-AEBAFFEA3D0C@yandex-team.ru
CIC and REINDEX CONCURRENTLY assume backends see their catalog changes
no later than each backend's next transaction start. That failed to
hold when a backend absorbed a relevant invalidation in the middle of
running RelationBuildDesc() on the CIC index. Queries that use the
resulting index can silently fail to find rows. Fix this for future
index builds by making RelationBuildDesc() loop until it finishes
without accepting a relevant invalidation. It may be necessary to
reindex to recover from past occurrences; REINDEX CONCURRENTLY suffices.
Back-patch to 9.6 (all supported versions).
Noah Misch and Andrey Borodin, reviewed (in earlier versions) by Andres
Freund.
Discussion: https://postgr.es/m/20210730022548.GA1940096@gust.leadboat.com
The code does not expect sh_error() to return, but the patch
that made this header usable in frontend didn't get that memo.
While here, plaster unlikely() on the tests that decide whether
to invoke sh_error(), and add our standard copyright notice.
Noted by Andres Freund. Back-patch to v13 where this frontend
support came in.
Discussion: https://postgr.es/m/0D54435C-1199-4361-9D74-2FBDCF8EA164@anarazel.de
During a replication slot creation, an ERROR generated in the same
transaction as the one creating a to-be-exported snapshot would have
left the backend in an inconsistent state, as the associated static
export snapshot state was not being reset on transaction abort, but only
on the follow-up command received by the WAL sender that created this
snapshot on replication slot creation. This would trigger inconsistency
failures if this session tried to export again a snapshot, like during
the creation of a replication slot.
Note that a snapshot export cannot happen in a transaction block, so
there is no need to worry resetting this state for subtransaction
aborts. Also, this inconsistent state would very unlikely show up to
users. For example, one case where this could happen is an
out-of-memory error when building the initial snapshot to-be-exported.
Dilip found this problem while poking at a different patch, that caused
an error in this code path for reasons unrelated to HEAD.
Author: Dilip Kumar
Reviewed-by: Michael Paquier, Zhihong Yu
Discussion: https://postgr.es/m/CAFiTN-s0zA1Kj0ozGHwkYkHwa5U0zUE94RSc_g81WrpcETB5=w@mail.gmail.com
Backpatch-through: 9.6
Commit 84f5c2908 forgot to consider the possibility that
EnsurePortalSnapshotExists could run inside a subtransaction with
lifespan shorter than the Portal's. In that case, the new active
snapshot would be popped at the end of the subtransaction, leaving
a dangling pointer in the Portal, with mayhem ensuing.
To fix, make sure the ActiveSnapshot stack entry is marked with
the same subtransaction nesting level as the associated Portal.
It's certainly safe to do so since we won't be here at all unless
the stack is empty; hence we can't create an out-of-order stack.
Let's also apply this logic in the case where PortalRunUtility
sets portalSnapshot, just to be sure that path can't cause similar
problems. It's slightly less clear that that path can't create
an out-of-order stack, so add an assertion guarding it.
Report and patch by Bertrand Drouvot (with kibitzing by me).
Back-patch to v11, like the previous commit.
Discussion: https://postgr.es/m/ff82b8c5-77f4-3fe7-6028-fcf3303e82dd@amazon.com
Physical replication always ships WAL segment files to replicas once
they are complete. This is a problem if one WAL record is split across
a segment boundary and the primary server crashes before writing down
the segment with the next portion of the WAL record: WAL writing after
crash recovery would happily resume at the point where the broken record
started, overwriting that record ... but any standby or backup may have
already received a copy of that segment, and they are not rewinding.
This causes standbys to stop following the primary after the latter
crashes:
LOG: invalid contrecord length 7262 at A8/D9FFFBC8
because the standby is still trying to read the continuation record
(contrecord) for the original long WAL record, but it is not there and
it will never be. A workaround is to stop the replica, delete the WAL
file, and restart it -- at which point a fresh copy is brought over from
the primary. But that's pretty labor intensive, and I bet many users
would just give up and re-clone the standby instead.
A fix for this problem was already attempted in commit 515e3d84a0b5, but
it only addressed the case for the scenario of WAL archiving, so
streaming replication would still be a problem (as well as other things
such as taking a filesystem-level backup while the server is down after
having crashed), and it had performance scalability problems too; so it
had to be reverted.
This commit fixes the problem using an approach suggested by Andres
Freund, whereby the initial portion(s) of the split-up WAL record are
kept, and a special type of WAL record is written where the contrecord
was lost, so that WAL replay in the replica knows to skip the broken
parts. With this approach, we can continue to stream/archive segment
files as soon as they are complete, and replay of the broken records
will proceed across the crash point without a hitch.
Because a new type of WAL record is added, users should be careful to
upgrade standbys first, primaries later. Otherwise they risk the standby
being unable to start if the primary happens to write such a record.
A new TAP test that exercises this is added, but the portability of it
is yet to be seen.
This has been wrong since the introduction of physical replication, so
backpatch all the way back. In stable branches, keep the new
XLogReaderState members at the end of the struct, to avoid an ABI
break.
Author: Álvaro Herrera <alvherre@alvh.no-ip.org>
Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Reviewed-by: Nathan Bossart <bossartn@amazon.com>
Discussion: https://postgr.es/m/202108232252.dh7uxf6oxwcy@alvherre.pgsql
That allows to include just visibilitymapdefs.h from file.c, and in turn,
remove include of postgres.h from relcache.h.
Reported-by: Andres Freund
Discussion: https://postgr.es/m/20210913232614.czafiubr435l6egi%40alap3.anarazel.de
Author: Alexander Korotkov
Reviewed-by: Andres Freund, Tom Lane, Alvaro Herrera
Backpatch-through: 13
Updates/Deletes on a partition were allowed even without replica identity
after the parent table was added to a publication. This would later lead
to an error on subscribers. The reason was that we were not invalidating
the partition's relcache and the publication information for partitions
was not getting rebuilt. Similarly, we were not invalidating the
partitions' relcache after dropping a partitioned table from a publication
which will prohibit Updates/Deletes on its partition without replica
identity even without any publication.
Reported-by: Haiying Tang
Author: Hou Zhijie and Vignesh C
Reviewed-by: Vignesh C and Amit Kapila
Backpatch-through: 13
Discussion: https://postgr.es/m/OS0PR01MB6113D77F583C922F1CEAA1C3FBD29@OS0PR01MB6113.jpnprd01.prod.outlook.com
Formerly, we sent signals for outgoing NOTIFY messages within
ProcessCompletedNotifies, which was also responsible for sending
relevant ones of those messages to our connected client. It therefore
had to run during the main-loop processing that occurs just before
going idle. This arrangement had two big disadvantages:
* Now that procedures allow intra-command COMMITs, it would be
useful to send NOTIFYs to other sessions immediately at COMMIT
(though, for reasons of wire-protocol stability, we still shouldn't
forward them to our client until end of command).
* Background processes such as replication workers would not send
NOTIFYs at all, since they never execute the client communication
loop. We've had requests to allow triggers running in replication
workers to send NOTIFYs, so that's a problem.
To fix these things, move transmission of outgoing NOTIFY signals
into AtCommit_Notify, where it will happen during CommitTransaction.
Also move the possible call of asyncQueueAdvanceTail there, to
ensure we don't bloat the async SLRU if a background worker sends
many NOTIFYs with no one listening.
We can also drop the call of asyncQueueReadAllNotifications,
allowing ProcessCompletedNotifies to go away entirely. That's
because commit 790026972 added a call of ProcessNotifyInterrupt
adjacent to PostgresMain's call of ProcessCompletedNotifies,
and that does its own call of asyncQueueReadAllNotifications,
meaning that we were uselessly doing two such calls (inside two
separate transactions) whenever inbound notify signals coincided
with an outbound notify. We need only set notifyInterruptPending
to ensure that ProcessNotifyInterrupt runs, and we're done.
The existing documentation suggests that custom background workers
should call ProcessCompletedNotifies if they want to send NOTIFY
messages. To avoid an ABI break in the back branches, reduce it
to an empty routine rather than removing it entirely. Removal
will occur in v15.
Although the problems mentioned above have existed for awhile,
I don't feel comfortable back-patching this any further than v13.
There was quite a bit of churn in adjacent code between 12 and 13.
At minimum we'd have to also backpatch 51004c717, and a good deal
of other adjustment would also be needed, so the benefit-to-risk
ratio doesn't look attractive.
Per bug #15293 from Michael Powers (and similar gripes from others).
Artur Zakirov and Tom Lane
Discussion: https://postgr.es/m/153243441449.1404.2274116228506175596@wrigleys.postgresql.org
If an allocation failed within LLVM it is not safe to call back into LLVM as
LLVM is not generally safe against exceptions / stack-unwinding. Thus errors
while in LLVM code are promoted to FATAL. However llvm_shutdown() did call
back into LLVM even in such cases, while llvm_release_context() was careful
not to do so.
We cannot generally skip shutting down LLVM, as that can break profiling. But
it's OK to do so if there was an error from within LLVM.
Reported-By: Jelte Fennema <Jelte.Fennema@microsoft.com>
Author: Andres Freund <andres@anarazel.de>
Author: Justin Pryzby <pryzby@telsasoft.com>
Discussion: https://postgr.es/m/AM5PR83MB0178C52CCA0A8DEA0207DC14F7FF9@AM5PR83MB0178.EURPRD83.prod.outlook.com
Backpatch: 11-, where jit was introduced
This reverts commit 515e3d84a0b5 and equivalent commits in back
branches. This solution to the problem has a number of problems, so
we'll try again with a different approach.
Per note from Andres Freund
Discussion: https://postgr.es/m/20210831042949.52eqp5xwbxgrfank@alap3.anarazel.de
Commit 325f2ec555 introduced pg_class.relwrite to skip operations on
tables created as part of a heap rewrite during DDL. It links such
transient heaps to the original relation OID via this new field in
pg_class but forgot to do anything about toast tables. So, logical
decoding was not able to skip operations on internally created toast
tables. This leads to an error when we tried to decode the WAL for the
next operation for which it appeared that there is a toast data where
actually it didn't have any toast data.
To fix this, we set pg_class.relwrite for internally created toast tables
as well which allowed skipping operations on them during logical decoding.
Author: Bertrand Drouvot
Reviewed-by: David Zhang, Amit Kapila
Backpatch-through: 11, where it was introduced
Discussion: https://postgr.es/m/b5146fb1-ad9e-7d6e-f980-98ed68744a7c@amazon.com
WAL records may span multiple segments, but XLogWrite() does not
wait for the entire record to be written out to disk before
creating archive status files. Instead, as soon as the last WAL page of
the segment is written, the archive status file is created, and the
archiver may process it. If PostgreSQL crashes before it is able to
write and flush the rest of the record (in the next WAL segment), the
wrong version of the first segment file lingers in the archive, which
causes operations such as point-in-time restores to fail.
To fix this, keep track of records that span across segments and ensure
that segments are only marked ready-for-archival once such records have
been completely written to disk.
This has always been wrong, so backpatch all the way back.
Author: Nathan Bossart <bossartn@amazon.com>
Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Reviewed-by: Ryo Matsumura <matsumura.ryo@fujitsu.com>
Reviewed-by: Andrey Borodin <x4mmm@yandex-team.ru>
Discussion: https://postgr.es/m/CBDDFA01-6E40-46BB-9F98-9340F4379505@amazon.com
transformLockingClause neglected to exclude the pseudo-RTEs for
OLD/NEW when processing a rule's query. This led to odd errors
or even crashes later on. This bug is very ancient, but it's
not terribly surprising that nobody noticed, since the use-case
for SELECT FOR UPDATE in a non-view rule is somewhere between
thin and non-existent. Still, crashing is not OK.
Per bug #17151 from Zhiyong Wu. Thanks to Masahiko Sawada
for analysis of the problem.
Discussion: https://postgr.es/m/17151-c03a3e6e4ec9aadb@postgresql.org
If recordDependencyOnCurrentExtension is invoked on a pre-existing,
free-standing object during an extension update script, that object
will become owned by the extension. In our current code this is
possible in three cases:
* Replacing a "shell" type or operator.
* CREATE OR REPLACE overwriting an existing object.
* ALTER TYPE SET, ALTER DOMAIN SET, and ALTER OPERATOR SET.
The first of these cases is intentional behavior, as noted by the
existing comments for GenerateTypeDependencies. It seems like
appropriate behavior for CREATE OR REPLACE too; at least, the obvious
alternatives are not better. However, the fact that it happens during
ALTER is an artifact of trying to share code (GenerateTypeDependencies
and makeOperatorDependencies) between the CREATE and ALTER cases.
Since an extension script would be unlikely to ALTER an object that
didn't already belong to the extension, this behavior is not very
troubling for the direct target object ... but ALTER TYPE SET will
recurse to dependent domains, and it is very uncool for those to
become owned by the extension if they were not already.
Let's fix this by redefining the ALTER cases to never change extension
membership, full stop. We could minimize the behavioral change by
only changing the behavior when ALTER TYPE SET is recursing to a
domain, but that would complicate the code and it does not seem like
a better definition.
Per bug #17144 from Alex Kozhemyakin. Back-patch to v13 where ALTER
TYPE SET was added. (The other cases are older, but since they only
affect the directly-named object, there's not enough of a problem to
justify changing the behavior further back.)
Discussion: https://postgr.es/m/17144-e67d7a8f049de9af@postgresql.org
Like the ARM case, just use gcc's __sync_lock_test_and_set();
that will compile into AMOSWAP.W.AQ which does what we need.
At some point it might be worth doing some work on atomic ops
for RISC-V, but this should be enough for a creditable port.
Back-patch to all supported branches, just in case somebody
wants to try them on RISC-V.
Marek Szuba
Discussion: https://postgr.es/m/dea97b6d-f55f-1f6d-9109-504aa7dfa421@gentoo.org
This fixes a bug in simplehash.h which caused an incorrect size mask to be
used when the hash table grew to SH_MAX_SIZE (2^32). The code was
incorrectly setting the size mask to 0 when the hash tables reached the
maximum possible number of buckets. This would result always trying to
use the 0th bucket causing an infinite loop of trying to grow the hash
table due to there being too many collisions.
Seemingly it's not that common for simplehash tables to ever grow this big
as this bug dates back to v10 and nobody seems to have noticed it before.
However, probably the most likely place that people would notice it would
be doing a large in-memory Hash Aggregate with something close to at least
2^31 groups.
After this fix, the code now works correctly with up to within 98% of 2^32
groups and will fail with the following error when trying to insert any
more items into the hash table:
ERROR: hash table size exceeded
However, the work_mem (or hash_mem_multiplier in newer versions) settings
will generally cause Hash Aggregates to spill to disk long before reaching
that many groups. The minimal test case I did took a work_mem setting of
over 192GB to hit the bug.
simplehash hash tables are used in a few other places such as Bitmap Index
Scans, however, again the size that the hash table can become there is
also limited to work_mem and it would take a relation of around 16TB
(2^31) pages and a very large work_mem setting to hit this. With smaller
work_mem values the table would become lossy and never grow large enough
to hit the problem.
Author: Yura Sokolov
Reviewed-by: David Rowley, Ranier Vilela
Discussion: https://postgr.es/m/b1f7f32737c3438136f64b26f4852b96@postgrespro.ru
Backpatch-through: 10, where simplehash.h was added
The point of introducing the hash_mem_multiplier GUC was to let users
reproduce the old behavior of hash aggregation, i.e. that it could use
more than work_mem at need. However, the implementation failed to get
the job done on Win64, where work_mem is clamped to 2GB to protect
various places that calculate memory sizes using "long int". As
written, the same clamp was applied to hash_mem. This resulted in
severe performance regressions for queries requiring a bit more than
2GB for hash aggregation, as they now spill to disk and there's no
way to stop that.
Getting rid of the work_mem restriction seems like a good idea, but
it's a big job and could not conceivably be back-patched. However,
there's only a fairly small number of places that are concerned with
the hash_mem value, and it turns out to be possible to remove the
restriction there without too much code churn or any ABI breaks.
So, let's do that for now to fix the regression, and leave the
larger task for another day.
This patch does introduce a bit more infrastructure that should help
with the larger task, namely pg_bitutils.h support for working with
size_t values.
Per gripe from Laurent Hasson. Back-patch to v13 where the
behavior change came in.
Discussion: https://postgr.es/m/997817.1627074924@sss.pgh.pa.us
Discussion: https://postgr.es/m/MN2PR15MB25601E80A9B6D1BA6F592B1985E39@MN2PR15MB2560.namprd15.prod.outlook.com
When triggers are cloned from partitioned tables to their partitions,
the 'tgenabled' flag (origin/replica/always/disable) was not propagated.
Make it so that the flag on the trigger on partition is initially set to
the same value as on the partitioned table.
Add a test case to verify the behavior.
Backpatch to 11, where this appeared in commit 86f575948c77.
Author: Álvaro Herrera <alvherre@alvh.no-ip.org>
Reported-by: Justin Pryzby <pryzby@telsasoft.com>
Discussion: https://postgr.es/m/20200930223450.GA14848@telsasoft.com
When some slots are invalidated due to the max_slot_wal_keep_size limit,
the old segment horizon should move forward to stay within the limit.
However, in commit c6550776394e we forgot to call KeepLogSeg again to
recompute the horizon after invalidating replication slots. In cases
where other slots remained, the limits would be recomputed eventually
for other reasons, but if all slots were invalidated, the limits would
not move at all afterwards. Repair.
Backpatch to 13 where the feature was introduced.
Author: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Reported-by: Marcin Krupowicz <mk@071.ovh>
Discussion: https://postgr.es/m/17103-004130e8f27782c9@postgresql.org
The separate libldap_r is gone and libldap itself is now always
thread-safe. Unfortunately there seems no easy way to tell by
inspection whether libldap is thread-safe, so we have to take
it on faith that libldap is thread-safe if there's no libldap_r.
That should be okay, as it appears that libldap_r was a standard
part of the installation going back at least 20 years.
Report and patch by Adrian Ho. Back-patch to all supported
branches, since people might try to build any of them with
a newer OpenLDAP.
Discussion: https://postgr.es/m/17083-a19190d9591946a7@postgresql.org
The existing code tried to do syscache lookups in an already-failed
transaction, which is problematic to say the least. After some
consideration of alternatives, the best fix seems to be to just drop
type names from the error message altogether. The table and column
names seem like sufficient localization. If the user is unsure what
types are involved, she can check the local and remote table
definitions.
Having done that, we can also discard the LogicalRepTypMap hash
table, which had no other use. Arguably, LOGICAL_REP_MSG_TYPE
replication messages are now obsolete as well; but we should
probably keep them in case some other use emerges. (The complexity
of removing something from the replication protocol would likely
outweigh any savings anyhow.)
Masahiko Sawada and Bharath Rupireddy, per complaint from Andres
Freund. Back-patch to v10 where this code originated.
Discussion: https://postgr.es/m/20210106020229.ne5xnuu6wlondjpe@alap3.anarazel.de
Move the newly defined enum value REORDER_BUFFER_CHANGE_INTERNAL_SPEC_ABORT
at the end to avoid ABI break in the back branches. We need to back-patch
this till v11 because before that it is already at the end.
Reported-by: Tomas Vondra
Backpatch-through: 11
Discussion: https://postgr.es/m/CAExHW5sPKF-Oovx_qZe4p5oM6Dvof7_P+XgsNAViug15Fm99jA@mail.gmail.com
During decoding for speculative inserts, we were relying for cleaning
toast hash on confirmation records or next change records. But that
could lead to multiple problems (a) memory leak if there is neither a
confirmation record nor any other record after toast insertion for a
speculative insert in the transaction, (b) error and assertion failures
if the next operation is not an insert/update on the same table.
The fix is to start queuing spec abort change and clean up toast hash
and change record during its processing. Currently, we are queuing the
spec aborts for both toast and main table even though we perform cleanup
while processing the main table's spec abort record. Later, if we have a
way to distinguish between the spec abort record of toast and the main
table, we can avoid queuing the change for spec aborts of toast tables.
Reported-by: Ashutosh Bapat
Author: Dilip Kumar
Reviewed-by: Amit Kapila
Backpatch-through: 9.6, where it was introduced
Discussion: https://postgr.es/m/CAExHW5sPKF-Oovx_qZe4p5oM6Dvof7_P+XgsNAViug15Fm99jA@mail.gmail.com
