Commits f92944137 et al. made IsInTransactionBlock() set the
XACT_FLAGS_NEEDIMMEDIATECOMMIT flag before returning "false",
on the grounds that that kept its API promises equivalent to those of
PreventInTransactionBlock(). This turns out to be a bad idea though,
because it allows an ANALYZE in a pipelined series of commands to
cause an immediate commit, which is unexpected.
Furthermore, if we return "false" then we have another issue,
which is that ANALYZE will decide it's allowed to do internal
commit-and-start-transaction sequences, thus possibly unexpectedly
committing the effects of previous commands in the pipeline.
To fix the latter situation, invent another transaction state flag
XACT_FLAGS_PIPELINING, which explicitly records the fact that we
have executed some extended-protocol command and not yet seen a
commit for it. Then, require that flag to not be set before allowing
InTransactionBlock() to return "false".
Having done that, we can remove its setting of NEEDIMMEDIATECOMMIT
without fear of causing problems. This means that the API guarantees
of IsInTransactionBlock now diverge from PreventInTransactionBlock,
which is mildly annoying, but it seems OK given the very limited usage
of IsInTransactionBlock. (In any case, a caller preferring the old
behavior could always set NEEDIMMEDIATECOMMIT for itself.)
For consistency also require XACT_FLAGS_PIPELINING to not be set
in PreventInTransactionBlock. This too is meant to prevent commands
such as CREATE DATABASE from silently committing previous commands
in a pipeline.
Per report from Peter Eisentraut. As before, back-patch to all
supported branches (which sadly no longer includes v10).
Discussion: https://postgr.es/m/65a899dd-aebc-f667-1d0a-abb89ff3abf8@enterprisedb.com
We have a few commands that "can't run in a transaction block",
meaning that if they complete their processing but then we fail
to COMMIT, we'll be left with inconsistent on-disk state.
However, the existing defenses for this are only watertight for
simple query protocol. In extended protocol, we didn't commit
until receiving a Sync message. Since the client is allowed to
issue another command instead of Sync, we're in trouble if that
command fails or is an explicit ROLLBACK. In any case, sitting
in an inconsistent state while waiting for a client message
that might not come seems pretty risky.
This case wasn't reachable via libpq before we introduced pipeline
mode, but it's always been an intended aspect of extended query
protocol, and likely there are other clients that could reach it
before.
To fix, set a flag in PreventInTransactionBlock that tells
exec_execute_message to force an immediate commit. This seems
to be the approach that does least damage to existing working
cases while still preventing the undesirable outcomes.
While here, add some documentation to protocol.sgml that explicitly
says how to use pipelining. That's latent in the existing docs if
you know what to look for, but it's better to spell it out; and it
provides a place to document this new behavior.
Per bug #17434 from Yugo Nagata. It's been wrong for ages,
so back-patch to all supported branches.
Discussion: https://postgr.es/m/17434-d9f7a064ce2a88a3@postgresql.org
The purpose of commit 8a54e12a38 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
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 515e3d84a0, 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
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
If a cross-partition UPDATE violates a constraint on the target partition,
and the columns in the new partition are in different physical order than
in the parent, the error message can reveal columns that the user does not
have SELECT permission on. A similar bug was fixed earlier in commit
804b6b6db4.
The cause of the bug is that the callers of the
ExecBuildSlotValueDescription() function got confused when constructing
the list of modified columns. If the tuple was routed from a parent, we
converted the tuple to the parent's format, but the list of modified
columns was grabbed directly from the child's RTE entry.
ExecUpdateLockMode() had a similar issue. That lead to confusion on which
columns are key columns, leading to wrong tuple lock being taken on tables
referenced by foreign keys, when a row is updated with INSERT ON CONFLICT
UPDATE. A new isolation test is added for that corner case.
With this patch, the ri_RangeTableIndex field is no longer set for
partitions that don't have an entry in the range table. Previously, it was
set to the RTE entry of the parent relation, but that was confusing.
NOTE: This modifies the ResultRelInfo struct, replacing the
ri_PartitionRoot field with ri_RootResultRelInfo. That's a bit risky to
backpatch, because it breaks any extensions accessing the field. The
change that ri_RangeTableIndex is not set for partitions could potentially
break extensions, too. The ResultRelInfos are visible to FDWs at least,
and this patch required small changes to postgres_fdw. Nevertheless, this
seem like the least bad option. I don't think these fields widely used in
extensions; I don't think there are FDWs out there that uses the FDW
"direct update" API, other than postgres_fdw. If there is, you will get a
compilation error, so hopefully it is caught quickly.
Backpatch to 11, where support for both cross-partition UPDATEs, and unique
indexes on partitioned tables, were added.
Reviewed-by: Amit Langote
Security: CVE-2021-3393
Every core SLRU wraps around. With the exception of pg_notify, the wrap
point can fall in the middle of a page. Account for this in the
PagePrecedes callback specification and in SimpleLruTruncate()'s use of
said callback. Update each callback implementation to fit the new
specification. This changes SerialPagePrecedesLogically() from the
style of asyncQueuePagePrecedes() to the style of CLOGPagePrecedes().
(Whereas pg_clog and pg_serial share a key space, pg_serial is nothing
like pg_notify.) The bug fixed here has the same symptoms and user
followup steps as 592a589a04. Back-patch
to 9.5 (all supported versions).
Reviewed by Andrey Borodin and (in earlier versions) by Tom Lane.
Discussion: https://postgr.es/m/20190202083822.GC32531@gust.leadboat.com
The previous coding imagined that it could call before_shmem_exit()
when a non-exclusive backup began and then remove the previously-added
handler by calling cancel_before_shmem_exit() when that backup
ended. However, this only works provided that nothing else in the
system has registered a before_shmem_exit() hook in the interim,
because cancel_before_shmem_exit() is documented to remove a callback
only if it is the latest callback registered. It also only works
if nothing can ERROR out between the time that sessionBackupState
is reset and the time that cancel_before_shmem_exit(), which doesn't
seem to be strictly true.
To fix, leave the handler installed for the lifetime of the session,
arrange to install it just once, and teach it to quietly do nothing if
there isn't a non-exclusive backup in process.
This was originally committed to master as
303640199d, but I did not back-patch
at the time because the consequences were minor. However, now
there's been a second report of this causing trouble with a slightly
different test case than the one I reported originally, so now
I'm back-patching as far as v11 where JIT was introduced.
Patch by me, reviewed by Kyotaro Horiguchi, Michael Paquier (who
preferred a different approach, but got outvoted), Fujii Masao,
and Tom Lane, and with comments by various others. New problem
report from Bharath Rupireddy.
Discussion: http://postgr.es/m/CA+TgmobMjnyBfNhGTKQEDbqXYE3_rXWpc4CM63fhyerNCes3mA@mail.gmail.com
Discussion: http://postgr.es/m/CALj2ACWk7j4F2v2fxxYfrroOF=AdFNPr1WsV+AGtHAFQOqm_pw@mail.gmail.com
Writing a trailing semicolon in a macro is almost never the right thing,
because you almost always want to write a semicolon after each macro
call instead. (Even if there was some reason to prefer not to, pgindent
would probably make a hash of code formatted that way; so within PG the
rule should basically be "don't do it".) Thus, if we have a semi inside
the macro, the compiler sees "something;;". Much of the time the extra
empty statement is harmless, but it could lead to mysterious syntax
errors at call sites. In perhaps an overabundance of neatnik-ism, let's
run around and get rid of the excess semicolons whereever possible.
The only thing worse than a mysterious syntax error is a mysterious
syntax error that only happens in the back branches; therefore,
backpatch these changes where relevant, which is most of them because
most of these mistakes are old. (The lack of reported problems shows
that this is largely a hypothetical issue, but still, it could bite
us in some future patch.)
John Naylor and Tom Lane
Discussion: https://postgr.es/m/CACPNZCs0qWTqJ2QUSGJ07B7uvAvzMb-KbG2q+oo+J3tsWN5cqw@mail.gmail.com
The logic for determining how many nbtree pages in an index are deleted
pages sometimes undercounted pages. Pages that were deleted by the
current VACUUM operation (as opposed to some previous VACUUM operation
whose deleted pages have yet to be reused) were sometimes overlooked.
The final count is exposed to users through VACUUM VERBOSE's "%u index
pages have been deleted" output.
btvacuumpage() avoided double-counting when _bt_pagedel() deleted more
than one page by assuming that only one page was deleted, and that the
additional deleted pages would get picked up during a future call to
btvacuumpage() by the same VACUUM operation. _bt_pagedel() can
legitimately delete pages that the btvacuumscan() scan will not visit
again, though, so that assumption was slightly faulty.
Fix the accounting by teaching _bt_pagedel() about its caller's
requirements. It now only reports on pages that it knows btvacuumscan()
won't visit again (including the current btvacuumpage() page), so
everything works out in the end.
This bug has been around forever. Only backpatch to v11, though, to
keep _bt_pagedel() is sync on the branches that have today's bugfix
commit b0229f26da. Note that this commit changes the signature of
_bt_pagedel(), just like commit b0229f26da.
Author: Peter Geoghegan
Reviewed-By: Masahiko Sawada
Discussion: https://postgr.es/m/CAH2-WzkrXBcMQWAYUJMFTTvzx_r4q=pYSjDe07JnUXhe+OZnJA@mail.gmail.com
Backpatch: 11-
Commit 857f9c36cd (which taught nbtree VACUUM to skip a scan of the
index from btcleanup in situations where it doesn't seem worth it) made
VACUUM maintain the oldest btpo.xact among all deleted pages for the
index as a whole. It failed to handle all the details surrounding pages
that are deleted by the current VACUUM operation correctly (though pages
deleted by some previous VACUUM operation were processed correctly).
The most immediate problem was that the special area of the page was
examined without a buffer pin at one point. More fundamentally, the
handling failed to account for the full range of _bt_pagedel()
behaviors. For example, _bt_pagedel() sometimes deletes internal pages
in passing, as part of deleting an entire subtree with btvacuumpage()
caller's page as the leaf level page. The original leaf page passed to
_bt_pagedel() might not be the page that it deletes first in cases where
deletion can take place.
It's unclear how disruptive this bug may have been, or what symptoms
users might want to look out for. The issue was spotted during
unrelated code review.
To fix, push down the logic for maintaining the oldest btpo.xact to
_bt_pagedel(). btvacuumpage() is now responsible for pages that were
fully deleted by a previous VACUUM operation, while _bt_pagedel() is now
responsible for pages that were deleted by the current VACUUM operation
(this includes half-dead pages from a previous interrupted VACUUM
operation that become fully deleted in _bt_pagedel()). Note that
_bt_pagedel() should never encounter an existing deleted page.
This commit theoretically breaks the ABI of a stable release by changing
the signature of _bt_pagedel(). However, if any third party extension
is actually affected by this, then it must already be completely broken
(since there are numerous assumptions made in _bt_pagedel() that cannot
be met outside of VACUUM). It seems highly unlikely that such an
extension actually exists, in any case.
Author: Peter Geoghegan
Reviewed-By: Masahiko Sawada
Discussion: https://postgr.es/m/CAH2-WzkrXBcMQWAYUJMFTTvzx_r4q=pYSjDe07JnUXhe+OZnJA@mail.gmail.com
Backpatch: 11-, where the "skip full scan" feature was introduced.
78ea8b5 has fixed an issue related to the recycling of WAL segments on
standbys depending on archive_mode. However, it has introduced a
regression with the handling of WAL segments ready to be archived during
crash recovery, causing those files to be recycled without getting
archived.
This commit fixes the regression by tracking in shared memory if a live
cluster is either in crash recovery or archive recovery as the handling
of WAL segments ready to be archived is different in both cases (those
WAL segments should not be removed during crash recovery), and by using
this new shared memory state to decide if a segment can be recycled or
not. Previously, it was not possible to know if a cluster was in crash
recovery or archive recovery as the shared state was able to track only
if recovery was happening or not, leading to the problem.
A set of TAP tests is added to close the gap here, making sure that WAL
segments ready to be archived are correctly handled when a cluster is in
archive or crash recovery with archive_mode set to "on" or "always", for
both standby and primary.
Reported-by: Benoît Lobréau
Author: Jehan-Guillaume de Rorthais
Reviewed-by: Kyotaro Horiguchi, Fujii Masao, Michael Paquier
Discussion: https://postgr.es/m/20200331172229.40ee00dc@firost
Backpatch-through: 9.5
Until now, only selected bulk operations (e.g. COPY) did this. If a
given relfilenode received both a WAL-skipping COPY and a WAL-logged
operation (e.g. INSERT), recovery could lose tuples from the COPY. See
src/backend/access/transam/README section "Skipping WAL for New
RelFileNode" for the new coding rules. Maintainers of table access
methods should examine that section.
To maintain data durability, just before commit, we choose between an
fsync of the relfilenode and copying its contents to WAL. A new GUC,
wal_skip_threshold, guides that choice. If this change slows a workload
that creates small, permanent relfilenodes under wal_level=minimal, try
adjusting wal_skip_threshold. Users setting a timeout on COMMIT may
need to adjust that timeout, and log_min_duration_statement analysis
will reflect time consumption moving to COMMIT from commands like COPY.
Internally, this requires a reliable determination of whether
RollbackAndReleaseCurrentSubTransaction() would unlink a relation's
current relfilenode. Introduce rd_firstRelfilenodeSubid. Amend the
specification of rd_createSubid such that the field is zero when a new
rel has an old rd_node. Make relcache.c retain entries for certain
dropped relations until end of transaction.
Back-patch to 9.5 (all supported versions). This introduces a new WAL
record type, XLOG_GIST_ASSIGN_LSN, without bumping XLOG_PAGE_MAGIC. As
always, update standby systems before master systems. This changes
sizeof(RelationData) and sizeof(IndexStmt), breaking binary
compatibility for affected extensions. (The most recent commit to
affect the same class of extensions was
089e4d405d0f3b94c74a2c6a54357a84a681754b.)
Kyotaro Horiguchi, reviewed (in earlier, similar versions) by Robert
Haas. Heikki Linnakangas and Michael Paquier implemented earlier
designs that materially clarified the problem. Reviewed, in earlier
designs, by Andrew Dunstan, Andres Freund, Alvaro Herrera, Tom Lane,
Fujii Masao, and Simon Riggs. Reported by Martijn van Oosterhout.
Discussion: https://postgr.es/m/20150702220524.GA9392@svana.org
Back-patch a subset of commit 9155580fd5
to v11, v10, 9.6, and 9.5. Include the latest repairs to this function.
Use a new XLOG_FPI_MULTI value instead of reusing XLOG_FPI. That way,
if an older server reads WAL from this function, that server will PANIC
instead of applying just one page of the record. The next commit adds a
call to this function.
Discussion: https://postgr.es/m/20200304.162919.898938381201316571.horikyota.ntt@gmail.com
This commit improves subject in two ways:
* It removes ugliness of 02f90879e7, which stores distance values and null
flags in two separate arrays after GISTSearchItem struct. Instead we pack
both distance value and null flag in IndexOrderByDistance struct. Alignment
overhead should be negligible, because we typically deal with at most few
"col op const" expressions in ORDER BY clause.
* It fixes handling of "col op NULL" expression in KNN-SP-GiST. Now, these
expression are not passed to support functions, which can't deal with them.
Instead, NULL result is implicitly assumed. It future we may decide to
teach support functions to deal with NULL arguments, but current solution is
bugfix suitable for backpatch.
Reported-by: Nikita Glukhov
Discussion: https://postgr.es/m/826f57ee-afc7-8977-c44c-6111d18b02ec%40postgrespro.ru
Author: Nikita Glukhov
Reviewed-by: Alexander Korotkov
Backpatch-through: 9.4
Include newitemoff in rmgr desc output for nbtree page split records.
In passing, correct an obsolete comment that claimed that newitemoff is
only logged for _L variant nbtree page split WAL records.
Both issues were oversights in commit 2c03216d83, which revamped the
WAL format.
Author: Peter Geoghegan
Backpatch: 9.5-, where the WAL format was revamped.
In order to implement NULL LAST semantic GiST previously assumed distance to
the NULL value to be Inf. However, our distance functions can return Inf and
NaN for non-null values. In such cases, NULL LAST semantic appears to be
broken. This commit fixes that by introducing separate array of null flags for
distances.
Backpatch to all supported versions.
Discussion: https://postgr.es/m/CAPpHfdsNvNdA0DBS%2BwMpFrgwT6C3-q50sFVGLSiuWnV3FqOJuQ%40mail.gmail.com
Author: Alexander Korotkov
Backpatch-through: 9.4
Commit 857f9c36cd, which taught nbtree VACUUM to avoid unnecessary
index scans, bumped the nbtree version number from 2 to 3, while adding
the ability for nbtree indexes to be upgraded on-the-fly. Various
assertions that assumed that an nbtree index was always on version 2 had
to be changed to accept any supported version (version 2 or 3 on
Postgres 11).
However, a few assertions were missed in the initial commit, all of
which were in code paths that cache a local copy of the metapage
metadata, where the index had been expected to be on the current version
(no longer version 2) as a generic sanity check. Rather than simply
update the assertions, follow-up commit 0a64b45152 intentionally made
the metapage caching code update the per-backend cached metadata version
without changing the on-disk version at the same time. This could even
happen when the planner needed to determine the height of a B-Tree for
costing purposes. The assertions only fail on Postgres v12 when
upgrading from v10, because they were adjusted to use the authoritative
shared memory metapage by v12's commit dd299df8.
To fix, remove the cache-only upgrade mechanism entirely, and update the
assertions themselves to accept any supported version (go back to using
the cached version in v12). The fix is almost a full revert of commit
0a64b45152 on the v11 branch.
VACUUM only considers the authoritative metapage, and never bothers with
a locally cached version, whereas everywhere else isn't interested in
the metapage fields that were added by commit 857f9c36cd. It seems
unlikely that this bug has affected any user on v11.
Reported-By: Christoph Berg
Bug: #15896
Discussion: https://postgr.es/m/15896-5b25e260fdb0b081%40postgresql.org
Backpatch: 11-, where VACUUM was taught to avoid unnecessary index scans.
52ac6cd2d0 added new field to ginxlogDeletePage and was backpatched to 9.4.
That led to problems when patched postgres instance applies WAL records
generated by non-patched one. WAL records generated by non-patched instance
don't contain new field, which patched one is expecting to see.
Thankfully, we can distinguish patched and non-patched WAL records by their data
size. If we see that WAL record is generated by non-patched instance, we skip
processing of new field. This commit comes with some assertions. In
particular, if it appears that on some platform struct data size didn't change
then static assertion will trigger.
Reported-by: Simon Riggs
Discussion: https://postgr.es/m/CANP8%2Bj%2BK4whxf7ET7%2BgO%2BG-baC3-WxqqH%3DnV4X2CgfEPA3Yu3g%40mail.gmail.com
Author: Alexander Korotkov
Reviewed-by: Simon Riggs, Alvaro Herrera
Backpatch-through: 9.4
Commit 6f6a6d8b1 introduced a delay of up to 2 seconds if we're trying
to request a checkpoint but the checkpointer hasn't started yet (or,
much less likely, our kill() call fails). However buildfarm experience
shows that that's not quite enough for slow or heavily-loaded machines.
There's no good reason to assume that the checkpointer won't start
eventually, so we may as well make the timeout much longer, say 60 sec.
However, if the caller didn't say CHECKPOINT_WAIT, it seems like a bad
idea to be waiting at all, much less for as long as 60 sec. We can
remove the need for that, and make this whole thing more robust, by
adjusting the code so that the existence of a pending checkpoint
request is clear from the contents of shared memory, and making sure
that the checkpointer process will notice it at startup even if it did
not get a signal. In this way there's no need for a non-CHECKPOINT_WAIT
call to wait at all; if it can't send the signal, it can nonetheless
assume that the checkpointer will eventually service the request.
A potential downside of this change is that "kill -INT" on the checkpointer
process is no longer enough to trigger a checkpoint, should anyone be
relying on something so hacky. But there's no obvious reason to do it
like that rather than issuing a plain old CHECKPOINT command, so we'll
assume that nobody is. There doesn't seem to be a way to preserve this
undocumented quasi-feature without introducing race conditions.
Since a principal reason for messing with this is to prevent intermittent
buildfarm failures, back-patch to all supported branches.
Discussion: https://postgr.es/m/27830.1552752475@sss.pgh.pa.us
Previously heap_getattr() returned NULL for attributes with a fast
default value (c.f. 16828d5c02), as it had no handling whatsoever
for that case.
A previous fix, 7636e5c60f, attempted to fix issues caused by this
oversight, but just expanding OLD tuples for triggers doesn't actually
solve the underlying issue.
One known consequence of this bug is that the check for HOT updates
can return the wrong result, when a previously fast-default'ed column
is set to NULL. Which in turn means that an index over a column with
fast default'ed columns might be corrupt if the underlying column(s)
allow NULLs.
Fix by handling fast default columns in heap_getattr(), remove now
superfluous expansion in GetTupleForTrigger().
Author: Andres Freund
Discussion: https://postgr.es/m/20190201162404.onngi77f26baem4g@alap3.anarazel.de
Backpatch: 11, where fast defaults were introduced
Attempting to use a temporary table within a two-phase transaction is
forbidden for ages. However, there have been uncovered grounds for
a couple of other object types and commands which work on temporary
objects with two-phase commit. In short, trying to create, lock or drop
an object on a temporary schema should not be authorized within a
two-phase transaction, as it would cause its state to create
dependencies with other sessions, causing all sorts of side effects with
the existing session or other sessions spawned later on trying to use
the same temporary schema name.
Regression tests are added to cover all the grounds found, the original
report mentioned function creation, but monitoring closer there are many
other patterns with LOCK, DROP or CREATE EXTENSION which are involved.
One of the symptoms resulting in combining both is that the session
which used the temporary schema is not able to shut down completely,
waiting for being able to drop the temporary schema, something that it
cannot complete because of the two-phase transaction involved with
temporary objects. In this case the client is able to disconnect but
the session remains alive on the backend-side, potentially blocking
connection backend slots from being used. Other problems reported could
also involve server crashes.
This is back-patched down to v10, which is where 9b013dc has introduced
MyXactFlags, something that this patch relies on.
Reported-by: Alexey Bashtanov
Author: Michael Paquier
Reviewed-by: Masahiko Sawada
Discussion: https://postgr.es/m/5d910e2e-0db8-ec06-dd5f-baec420513c3@imap.cc
Backpatch-through: 10
According to README we acquire predicate locks on entry tree leafs and posting
tree roots. However, when ginFindLeafPage() is going to lock leaf in exclusive
mode, then it checks root for conflicts regardless whether it's a entry or
posting tree. Assuming that we never place predicate lock on entry tree root
(excluding corner case when root is leaf), this check is redundant. This
commit removes this check. Now, root conflict checking is controlled by
separate argument of ginFindLeafPage().
Discussion: https://postgr.es/m/CAPpHfdv7rrDyy%3DMgsaK-L9kk0AH7az0B-mdC3w3p0FSb9uoyEg%40mail.gmail.com
Author: Alexander Korotkov
Backpatch-through: 11
013ebc0a7b implements so-called GiST microvacuum. That is gistgettuple() marks
index tuples as dead when kill_prior_tuple is set. Later, when new tuple
insertion claims page space, those dead index tuples are physically deleted
from page. When this deletion is replayed on standby, it might conflict with
read-only queries. But 013ebc0a7b doesn't handle this. That may lead to
disappearance of some tuples from read-only snapshots on standby.
This commit implements resolving of conflicts between replay of GiST microvacuum
and standby queries. On the master we implement new WAL record type
XLOG_GIST_DELETE, which comprises necessary information. On stable releases
we've to be tricky to keep WAL compatibility. Information required for conflict
processing is just appended to data of XLOG_GIST_PAGE_UPDATE record. So,
PostgreSQL version, which doesn't know about conflict processing, will just
ignore that.
Reported-by: Andres Freund
Diagnosed-by: Andres Freund
Discussion: https://postgr.es/m/20181212224524.scafnlyjindmrbe6%40alap3.anarazel.de
Author: Alexander Korotkov
Backpatch-through: 9.6
When GIN vacuum deletes a posting tree page, it assumes that no concurrent
searchers can access it, thanks to ginStepRight() locking two pages at once.
However, since 9.4 searches can skip parts of posting trees descending from the
root. That leads to the risk that page is deleted and reclaimed before
concurrent search can access it.
This commit prevents the risk of above by waiting for every transaction, which
might wait to reference this page, to finish. Due to binary compatibility
we can't change GinPageOpaqueData to store corresponding transaction id.
Instead we reuse page header pd_prune_xid field, which is unused in index pages.
Discussion: https://postgr.es/m/31a702a.14dd.166c1366ac1.Coremail.chjischj%40163.com
Author: Andrey Borodin, Alexander Korotkov
Reviewed-by: Alexander Korotkov
Backpatch-through: 9.4
The "rb" prefix is used by Ruby, so that our existing code results
in name collisions that break plruby. We discussed ways to prevent
that by adjusting dynamic linker options, but it seems that at best
we'd move the pain to other cases. Renaming to avoid the collision
is the only portable fix anyway. Fortunately, our rbtree code is
not (yet?) widely used --- in core, there's only a single usage
in GIN --- so it seems likely that we can get away with a rename.
I chose to do this basically as s/rb/rbt/g, except for places where
there already was a "t" after "rb". The patch could have been made
smaller by only touching linker-visible symbols, but it would have
resulted in oddly inconsistent-looking code. Better to make it look
like "rbt" was the plan all along.
Back-patch to v10. The rbtree.c code exists back to 9.5, but
rb_iterate() which is the actual immediate source of pain was added
in v10, so it seems like changing the names before that would have
more risk than benefit.
Per report from Pavel Raiskup.
Discussion: https://postgr.es/m/4738198.8KVIIDhgEB@nb.usersys.redhat.com
Repeatedly rewriting a mapped catalog table with VACUUM FULL or
CLUSTER could cause logical decoding to fail with:
ERROR, "could not map filenode \"%s\" to relation OID"
To trigger the problem the rewritten catalog had to have live tuples
with toasted columns.
The problem was triggered as during catalog table rewrites the
heap_insert() check that prevents logical decoding information to be
emitted for system catalogs, failed to treat the new heap's toast table
as a system catalog (because the new heap is not recognized as a
catalog table via RelationIsLogicallyLogged()). The relmapper, in
contrast to the normal catalog contents, does not contain historical
information. After a single rewrite of a mapped table the new relation
is known to the relmapper, but if the table is rewritten twice before
logical decoding occurs, the relfilenode cannot be mapped to a
relation anymore. Which then leads us to error out. This only
happens for toast tables, because the main table contents aren't
re-inserted with heap_insert().
The fix is simple, add a new heap_insert() flag that prevents logical
decoding information from being emitted, and accept during decoding
that there might not be tuple data for toast tables.
Unfortunately that does not fix pre-existing logical decoding
errors. Doing so would require not throwing an error when a filenode
cannot be mapped to a relation during decoding, and that seems too
likely to hide bugs. If it's crucial to fix decoding for an existing
slot, temporarily changing the ERROR in ReorderBufferCommit() to a
WARNING appears to be the best fix.
Author: Andres Freund
Discussion: https://postgr.es/m/20180914021046.oi7dm4ra3ot2g2kt@alap3.anarazel.de
Backpatch: 9.4-, where logical decoding was introduced
Previously, a worker process would establish values for these based on
its own start time. In v10 and up, this can trivially be shown to cause
misbehavior of transaction_timestamp(), timestamp_in(), and related
functions which are (perhaps unwisely?) marked parallel-safe. It seems
likely that other behaviors might diverge from what happens in the parent
as well.
It's not as trivial to demonstrate problems in 9.6 or 9.5, but I'm sure
it's still possible, so back-patch to all branches containing parallel
worker infrastructure.
In HEAD only, mark now() and statement_timestamp() as parallel-safe
(other affected functions already were). While in theory we could
still squeeze that change into v11, it doesn't seem important enough
to force a last-minute catversion bump.
Konstantin Knizhnik, whacked around a bit by me
Discussion: https://postgr.es/m/6406dbd2-5d37-4cb6-6eb2-9c44172c7e7c@postgrespro.ru
Historically we forbade datatype-specific comparison functions from
returning INT_MIN, so that it would be safe to invert the sort order
just by negating the comparison result. However, this was never
really safe for comparison functions that directly return the result
of memcmp(), strcmp(), etc, as POSIX doesn't place any such restriction
on those library functions. Buildfarm results show that at least on
recent Linux on s390x, memcmp() actually does return INT_MIN sometimes,
causing sort failures.
The agreed-on answer is to remove this restriction and fix relevant
call sites to not make such an assumption; code such as "res = -res"
should be replaced by "INVERT_COMPARE_RESULT(res)". The same is needed
in a few places that just directly negated the result of memcmp or
strcmp.
To help find places having this problem, I've also added a compile option
to nbtcompare.c that causes some of the commonly used comparators to
return INT_MIN/INT_MAX instead of their usual -1/+1. It'd likely be
a good idea to have at least one buildfarm member running with
"-DSTRESS_SORT_INT_MIN". That's far from a complete test of course,
but it should help to prevent fresh introductions of such bugs.
This is a longstanding portability hazard, so back-patch to all supported
branches.
Discussion: https://postgr.es/m/20180928185215.ffoq2xrq5d3pafna@alap3.anarazel.de
The commit 620b49a1 changed the value of HASH_MAX_BITMAPS with the intent
to allow many non-unique values in hash indexes without worrying to reach
the limit of the number of overflow pages. At that time, this didn't
occur to us that it can overrun the block for smaller block sizes.
Choose the value of HASH_MAX_BITMAPS based on BLCKSZ such that it gives
the same answer as now for the cases where the overrun doesn't occur, and
some other sufficiently-value for the cases where an overrun currently
does occur. This allows us not to change the behavior in any case that
currently works, so there's really no reason for a HASH_VERSION bump.
Author: Dilip Kumar
Reviewed-by: Amit Kapila
Backpatch-through: 10
Discussion: https://postgr.es/m/CAA4eK1LtF4VmU4mx_+i72ff1MdNZ8XaJMGkt2HV8+uSWcn8t4A@mail.gmail.com
Historically, the term procedure was used as a synonym for function in
Postgres/PostgreSQL. Now we have procedures as separate objects from
functions, so we need to clean up the documentation to not mix those
terms.
In particular, mentions of "trigger procedures" are changed to "trigger
functions", and access method "support procedures" are changed to
"support functions". (The latter already used FUNCTION in the SQL
syntax anyway.) Also, the terminology in the SPI chapter has been
cleaned up.
A few tests, examples, and code comments are also adjusted to be
consistent with documentation changes, but not everything.
Reported-by: Peter Geoghegan <pg@bowt.ie>
Reviewed-by: Jonathan S. Katz <jonathan.katz@excoventures.com>
Any changes on page should be done in critical section, so move
_bt_upgrademetapage into critical section. Improve comment. Found by Amit
Kapila during post-commit review of 857f9c36.
Author: Amit Kapila
If a continuation record is split so that its first half has already been
removed from the master, and is only present in pg_wal, and there is a
recycled WAL segment in the standby server that looks like it would
contain the second half, recovery would get stuck. The code in
XLogPageRead() incorrectly started streaming at the beginning of the
WAL record, even if we had already read the first page.
Backpatch to 9.4. In principle, older versions have the same problem, but
without replication slots, there was no straightforward mechanism to
prevent the master from recycling old WAL that was still needed by standby.
Without such a mechanism, I think it's reasonable to assume that there's
enough slack in how many old segments are kept around to not run into this,
or you have a WAL archive.
Reported by Jonathon Nelson. Analysis and patch by Kyotaro HORIGUCHI, with
some extra comments by me.
Discussion: https://www.postgresql.org/message-id/CACJqAM3xVz0JY1XFDKPP%2BJoJAjoGx%3DGNuOAshEDWCext7BFvCQ%40mail.gmail.com
The principle behind the locking was not very well thought-out, and not
documented. Add a section in the README to explain how it's supposed to
work, and change the code so that it actually works that way.
This fixes two bugs:
1. If fast update was turned on concurrently, subsequent inserts to the
pending list would not conflict with predicate locks that were acquired
earlier, on entry pages. The included 'predicate-gin-fastupdate' test
demonstrates that. To fix, make all scans acquire a predicate lock on
the metapage. That lock represents a scan of the pending list, whether
or not there is a pending list at the moment. Forget about the
optimization to skip locking/checking for locks, when fastupdate=off.
2. If a scan finds no match, it still needs to lock the entry page. The
point of predicate locks is to lock the gabs between values, whether
or not there is a match. The included 'predicate-gin-nomatch' test
tests that case.
In addition to those two bug fixes, this removes some unnecessary locking,
following the principle laid out in the README. Because all items in
a posting tree have the same key value, a lock on the posting tree root is
enough to cover all the items. (With a very large posting tree, it would
possibly be better to lock the posting tree leaf pages instead, so that a
"skip scan" with a query like "A & B", you could avoid unnecessary conflict
if a new tuple is inserted with A but !B. But let's keep this simple.)
Also, some spelling fixes.
Author: Heikki Linnakangas with some editorization by me
Review: Andrey Borodin, Alexander Korotkov
Discussion: https://www.postgresql.org/message-id/0b3ad2c2-2692-62a9-3a04-5724f2af9114@iki.fi
Previously a tuple that has been moved to a different partition (see
f16241bef7), set the block number on the old tuple to an invalid
value to indicate that fact. But the tuple offset was left
untouched. That turned out to trigger a wal_consistency_checking
failure as reported by Peter Geoghegan, as the offset wasn't
always overwritten during WAL replay.
Heikki observed that we're wasting valuable data by not putting
information also in the offset. Thus set that to
MovedPartitionsOffsetNumber when a tuple indicates it has moved.
We continue to set the block number to MovedPartitionsBlockNumber, as
that seems more likely to cause problems for code not updated to know
about moved tuples.
As t_ctid's offset number is now always set, this refinement also
fixes the wal_consistency_checking issue.
This technically is a minor disk format break, with previously created
moved tuples not being recognized anymore. But since there not even
has been a beta release since f16241bef7c...
Reported-By: Peter Geoghegan
Author: Heikki Linnakangas, Amul Sul
Discussion: https://postgr.es/m/CAH2-Wzm9ty+1BX7-GMNJ=xPRg67oJTVeDNdA9LSyJJtMgRiCMA@mail.gmail.com
After introducing usage of t_tid of inner or page high key for storing
number of attributes of tuple, validation of tuple's ItemPointer with
ItemPointerIsValid becomes incorrect, it's need to validate only blocknumber of
ItemPointer. Missing this causes a incorrect page deletion, fix that. Test is
added.
BTW, current contrib/amcheck doesn't fail on index corrupted by this way.
Also introduce BTreeTupleGetTopParent/BTreeTupleSetTopParent macroses to improve
code readability and to avoid possible confusion with page high key: high key
is used to store top-parent link for branch to remove.
Bug found by Michael Paquier, but bug doesn't exist in previous versions because
t_tid was set to P_HIKEY.
Author: Teodor Sigaev
Reviewer: Peter Geoghegan
Discussion: https://www.postgresql.org/message-id/flat/20180419052436.GA16000%40paquier.xyz
It appears that new fields introduced in 857f9c36 have inconsistent datatypes:
BTMetaPageData.btm_last_cleanup_num_heap_tuples is of float4 type,
while xl_btree_metadata.last_cleanup_num_heap_tuples is of double type.
IndexVacuumInfo.num_heap_tuples, which is a source of values for
both former fields is of double type. So, make both those fields in
BTMetaPageData and xl_btree_metadata use float8 type in order to match the
precision of the source. That shouldn't be double type, because we always
use types with explicit width in WAL.
Patch introduces incompatibility of on-disk format since 857f9c36 commit, but
that versions never was released, so just bump catalog version to avoid
possible confusion.
Author: Alexander Korortkov
Add several assertions that ensure that we're dealing with a pivot tuple
without non-key attributes where that's expected. Also, remove the
assertion within _bt_isequal(), restoring the v10 function signature. A
similar check will be performed for the page highkey within
_bt_moveright() in most cases. Also avoid dropping all objects within
regression tests, to increase pg_dump test coverage for INCLUDE indexes.
Rather than using infrastructure that's generally intended to be used
with reference counted heap tuple descriptors during truncation, use the
same function that was introduced to store flat TupleDescs in shared
memory (we use a temp palloc'd buffer). This isn't strictly necessary,
but seems more future-proof than the old approach. It also lets us
avoid including rel.h within indextuple.c, which was arguably a
modularity violation. Also, we now call index_deform_tuple() with the
truncated TupleDesc, not the source TupleDesc, since that's more robust,
and saves a few cycles.
In passing, fix a memory leak by pfree'ing truncated pivot tuple memory
during CREATE INDEX. Also pfree during a page split, just to be
consistent.
Refactor _bt_check_natts() to be more readable.
Author: Peter Geoghegan with some editorization by me
Reviewed by: Alexander Korotkov, Teodor Sigaev
Discussion: https://www.postgresql.org/message-id/CAH2-Wz%3DkCWuXeMrBCopC-tFs3FbiVxQNjjgNKdG2sHxZ5k2y3w%40mail.gmail.com
Review of commit 1eb6d652: It's pointless to add padding to the GID fields,
when the code that follows assumes that there is no alignment, and uses
memcpy(). Remove the pointless padding.
Update comments to note the new fields in the WAL records.
Reviewed-by: Michael Paquier
Discussion: https://www.postgresql.org/message-id/33b787bf-dc20-1161-54e9-3f3b607bf59d%40iki.fi
MaxIndexTuplesPerPage ignores the fact that btree indexes sometimes
store tuples with no data payload. But it also ignores the possibility
of "special space" on index pages, which offsets that, so that the
result isn't an underestimate. This all seems worth documenting, though.
In passing, remove #define MinIndexTupleSize, which was added by
commit 2c03216d8 but not used in that commit nor later ones.
Comment text by me; issue noticed by Peter Geoghegan.
Discussion: https://postgr.es/m/CAH2-WzkQmb54Kbx-YHXstRKXcNc+_87jwV3DRb54xcybLR7Oig@mail.gmail.com
This reverts commits d204ef6377,
83454e3c2b and a few more commits thereafter
(complete list at the end) related to MERGE feature.
While the feature was fully functional, with sufficient test coverage and
necessary documentation, it was felt that some parts of the executor and
parse-analyzer can use a different design and it wasn't possible to do that in
the available time. So it was decided to revert the patch for PG11 and retry
again in the future.
Thanks again to all reviewers and bug reporters.
List of commits reverted, in reverse chronological order:
f1464c5380 Improve parse representation for MERGE
ddb4158579 MERGE syntax diagram correction
530e69e59b Allow cpluspluscheck to pass by renaming variable
01b88b4df5 MERGE minor errata
3af7b2b0d4 MERGE fix variable warning in non-assert builds
a5d86181ec MERGE INSERT allows only one VALUES clause
4b2d44031f MERGE post-commit review
4923550c20 Tab completion for MERGE
aa3faa3c7a WITH support in MERGE
83454e3c2b New files for MERGE
d204ef6377 MERGE SQL Command following SQL:2016
Author: Pavan Deolasee
Reviewed-by: Michael Paquier
