By default, $PGUSER has permission to unlink $PGLOG. If $PGUSER
replaces $PGLOG with a symbolic link, the server will corrupt the
link-targeted file by appending log messages. Since these scripts open
$PGLOG as root, the attack works regardless of target file ownership.
"make install" does not install these scripts anywhere. Users having
manually installed them in the past should repeat that process to
acquire this fix. Most script users have $PGLOG writable to root only,
located in $PGDATA. Just before updating one of these scripts, such
users should rename $PGLOG to $PGLOG.old. The script will then recreate
$PGLOG with proper ownership.
Reviewed by Peter Eisentraut. Reported by Antoine Scemama.
Security: CVE-2017-12172
The logical decoding functions do BeginInternalSubTransaction and
RollbackAndReleaseCurrentSubTransaction to clean up after themselves.
It turns out that AtEOSubXact_SPI has an unrecognized assumption that
we always need to cancel the active SPI operation in the SPI context
that surrounds the subtransaction (if there is one). That's true
when the RollbackAndReleaseCurrentSubTransaction call is coming from
the SPI-using function itself, but not when it's happening inside
some unrelated function invoked by a SPI query. In practice the
affected callers are the various PLs.
To fix, record the current subtransaction ID when we begin a SPI
operation, and clean up only if that ID is the subtransaction being
canceled.
Also, remove AtEOSubXact_SPI's assertion that it must have cleaned
up the surrounding SPI context's active tuptable. That's proven
wrong by the same test case.
Also clarify (or, if you prefer, reinterpret) the calling conventions
for _SPI_begin_call and _SPI_end_call. The memory context cleanup
in the latter means that these have always had the flavor of a matched
resource-management pair, but they weren't documented that way before.
Per report from Ben Chobot.
Back-patch to 9.4 where logical decoding came in. In principle,
the SPI changes should go all the way back, since the problem dates
back to commit 7ec1c5a86. But given the lack of field complaints
it seems few people are using internal subtransactions in this way.
So I don't feel a need to take any risks in 9.2/9.3.
Discussion: https://postgr.es/m/73FBA179-C68C-4540-9473-71E865408B15@silentmedia.com
If the user modifies a view that has CHECK OPTIONs and this gets
translated into a modification to an underlying relation which happens
to be a foreign table, the check options should be enforced. In the
normal code path, that was happening properly, but it was not working
properly for "direct" modification because the whole operation gets
pushed to the remote side in that case and we never have an option to
enforce the constraint against individual tuples. Fix by disabling
direct modification when there is a need to enforce CHECK OPTIONs.
Etsuro Fujita, reviewed by Kyotaro Horiguchi and by me.
Discussion: http://postgr.es/m/f8a48f54-6f02-9c8a-5250-9791603171ee@lab.ntt.co.jp
The new test cases added in commit 8bf58c0d9 turn out to have output
that can vary depending on the lc_messages setting prevailing on the
test server. Hide the remote end's error messages to ensure stable
output. This isn't a terribly desirable solution; we'd rather know
that the connection failed for the expected reason and not some other
one. But there seems little choice for the moment.
Per buildfarm.
Discussion: https://postgr.es/m/18419.1500658570@sss.pgh.pa.us
Previously, postgres_fdw would keep on using an existing connection even
if the user did ALTER SERVER or ALTER USER MAPPING commands that should
affect connection parameters. Teach it to watch for catcache invals
on these catalogs and re-establish connections when the relevant catalog
entries change. Per bug #14738 from Michal Lis.
In passing, clean up some rather crufty decisions in commit ae9bfc5d6
about where fields of ConnCacheEntry should be reset. We now reset
all the fields whenever we open a new connection.
Kyotaro Horiguchi, reviewed by Ashutosh Bapat and myself.
Back-patch to 9.3 where postgres_fdw appeared.
Discussion: https://postgr.es/m/20170710113917.7727.10247@wrigleys.postgresql.org
We had three occurrences of essentially the same coding pattern
wherein we tried to retrieve a query result from a libpq connection
without blocking. In the case where PQconsumeInput failed (typically
indicating a lost connection), all three loops simply gave up and
returned, forgetting to clear any previously-collected PGresult
object. Since those are malloc'd not palloc'd, the oversight results
in a process-lifespan memory leak.
One instance, in libpqwalreceiver, is of little significance because
the walreceiver process would just quit anyway if its connection fails.
But we might as well fix it.
The other two instances, in postgres_fdw, are somewhat more worrisome
because at least in principle the scenario could be repeated, allowing
the amount of memory leaked to build up to something worth worrying
about. Moreover, in these cases the loops contain CHECK_FOR_INTERRUPTS
calls, as well as other calls that could potentially elog(ERROR),
providing another way to exit without having cleared the PGresult.
Here we need to add PG_TRY logic similar to what exists in quite a
few other places in postgres_fdw.
Coverity noted the libpqwalreceiver bug; I found the other two cases
by checking all calls of PQconsumeInput.
Back-patch to all supported versions as appropriate (9.2 lacks
postgres_fdw, so this is really quite unexciting for that branch).
Discussion: https://postgr.es/m/22620.1497486981@sss.pgh.pa.us
Commit f039eaac71, later back-patched
with commit 1b812afb0e, allowed many of
the queries issued by postgres_fdw to fetch remote data to respond to
cancel interrupts in a timely fashion. However, it didn't do anything
about the transaction control commands, which remained
noninterruptible.
Improve the situation by changing do_sql_command() to retrieve query
results using pgfdw_get_result(), which uses the asynchronous
interface to libpq so that it can check for interrupts every time
libpq returns control. Since this might result in a situation
where we can no longer be sure that the remote transaction state
matches the local transaction state, add a facility to force all
levels of the local transaction to abort if we've lost track of
the remote state; without this, an apparently-successful commit of
the local transaction might fail to commit changes made on the
remote side. Also, add a 60-second timeout for queries issue during
transaction abort; if that expires, give up and mark the state of
the connection as unknown. Drop all such connections when we exit
the local transaction. Together, these changes mean that if we're
aborting the local toplevel transaction anyway, we can just drop the
remote connection in lieu of waiting (possibly for a very long time)
for it to complete an abort.
This still leaves quite a bit of room for improvement. PQcancel()
has no asynchronous interface, so if we get stuck sending the cancel
request we'll still hang. Also, PQsetnonblocking() is not used, which
means we could block uninterruptibly when sending a query. There
might be some other optimizations possible as well. Nonetheless,
this allows us to escape a wait for an unresponsive remote server
quickly in many more cases than previously.
Report by Suraj Kharage. Patch by me and Rafia Sabih. Review
and testing by Amit Kapila and Tushar Ahuja.
Discussion: http://postgr.es/m/CAF1DzPU8Kx+fMXEbFoP289xtm3bz3t+ZfxhmKavr98Bh-C0TqQ@mail.gmail.com
Ultra-modern versions of the perl Data::Dumper module have apparently
changed how they indent output. Instead of trying to keep up we choose
to tell it to supporess all indentation in the hstore_plperl regression
tests.
Backpatch to 9.5 where this feature was introduced.
The snapshot assembly during the creation of logical slots relied
waiting for transactions in xl_running_xacts to end, by checking for
their commit/abort records. Unfortunately, despite locking, it is
possible to see an xl_running_xact record listing transactions as
ready, that have already WAL-logged an commit/abort record, as the
locking just prevents the ProcArray to be adjusted, and the commit
record has to be logged first.
That lead to either delayed or hanging snapshot creation, because
snapbuild.c would wait "forever" to see commit/abort records for some
transactions. That hang resolved only if a xl_running_xacts record
without any running transactions happened to be logged, far from
certain on a busy server.
It's impractical to prevent that via more heavyweight locking, the
likelihood of deadlocks and significantly increased contention would
be too big.
Instead change the initial snapshot creation to be solely based on
tracking the oldest running transaction via
xl_running_xacts->oldestRunningXid - that actually ends up
significantly simplifying the code. That has two disadvantages:
1) Because we cannot fully "trust" the contents of xl_running_xacts,
we cannot use it to build the initial snapshot. Instead we have to
wait twice for all running transactions to finish.
2) Previously a slot, unless the race occurred, could be created when
the all transaction perceived as running based on commit/abort
records, now we have to wait for the next xl_running_xacts record.
To address that, trigger logging new xl_running_xacts record from
within snapbuild.c exactly when necessary.
Unfortunately snabuild.c's SnapBuild is stored on disk, one of the
stupider ideas of a certain Mr Freund, so we can't change it in a
minor release. As this is going to be backpatched, we have to hack
around a bit to keep on-disk compatibility. A later commit will
rejigger that on master.
Author: Andres Freund, based on a quite different patch from Petr Jelinek
Analyzed-By: Petr Jelinek
Reviewed-By: Petr Jelinek
Discussion: https://postgr.es/m/f37e975c-908f-858e-707f-058d3b1eb214@2ndquadrant.com
Backpatch: 9.4-, where logical decoding has been introduced
Objects in an extension are shippable to a foreign server if the
extension is part of the foreign server definition's shippable
extensions list. But this was not properly considered in some cases
when checking whether a join condition can be pushed to a foreign server
and the join condition uses an object from a shippable extension. So
the join would never be pushed down in those cases.
So, the list of extensions needs to be made available in fpinfo of the
relation being considered to be pushed down before any expressions are
assessed for being shippable. Fix foreign_join_ok() to do that for a
join relation.
David Rowley and Ashutosh Bapat, per report from David Rowley
reduced version of patch 332bec1e60 for
backpatch to 9.6, first release with join push down
Commit 9e43e8714 turns out to have been insufficient: not only is it
necessary to track tentative parent links while considering a set of
arc removals, but it's necessary to track tentative flag additions
as well. This is because we always merge arc target states into
arc source states; therefore, when considering a merge of the final
state with some other, it is the other state that will acquire a new
TSTATE_FIN bit. If there's another arc for the same color trigram
that would cause merging of that state with the initial state, we
failed to recognize the problem. The test cases for the prior commit
evidently only exercised situations where a tentative merge with the
initial state occurs before one with the final state. If it goes the
other way around, we'll happily merge the initial and final states,
either producing a broken final graph that would never match anything,
or triggering the Assert added by the prior commit.
It's tempting to consider switching the merge direction when the merge
involves the final state, but I lack the time to analyze that idea in
detail. Instead just keep track of the flag changes that would result
from proposed merges, in the same way that the prior commit tracked
proposed parent links.
Along the way, add some more debugging support, because I'm not entirely
confident that this is the last bug here. And tweak matters so that
the transformed.dot file uses small integers rather than pointer values
to identify states; that makes it more readable if you're just eyeballing
it rather than fooling with Graphviz. And rename a couple of identically
named struct fields to reduce confusion.
Per report from Corey Csuhta. Add a test case based on his example.
(Note: this case does not trigger the bug under 9.3, apparently because
its different measurement of costs causes it to stop merging states before
it hits the failure. I spent some time trying to find a variant that would
fail in 9.3, without success; but I'm sure such cases exist.)
Like the previous patch, back-patch to 9.3 where this code was added.
Report: https://postgr.es/m/E2B01A4B-4530-406B-8D17-2F67CF9A16BA@csuhta.com
regexport.c thought it could just ignore LACON arcs, but the correct
behavior is to treat them as satisfiable while consuming zero input
(rather reminiscently of commit 9f1e642d5). Otherwise, the emitted
simplified-NFA representation may contain no paths leading from initial
to final state, which unsurprisingly confuses pg_trgm, as seen in
bug #14623 from Jeff Janes.
Since regexport's output representation has no concept of an arc that
consumes zero input, recurse internally to find the next normal arc(s)
after any LACON transitions. We'd be forced into changing that
representation if a LACON could be the last arc reaching the final
state, but fortunately the regex library never builds NFAs with such
a configuration, so there always is a next normal arc.
Back-patch to 9.3 where this logic was introduced.
Discussion: https://postgr.es/m/20170413180503.25948.94871@wrigleys.postgresql.org
<selinux/label.h> includes <stdbool.h>, which creates an incompatible
We don't care if <stdbool.h> redefines "true"/"false"; those are close
enough.
Complaint and initial patch by Mike Palmiotto. Final approach per
Tom Lane's suggestion, as discussed on hackers. Backpatching to
all supported branches.
Discussion: https://postgr.es/m/flat/623bcaae-112e-ced0-8c22-a84f75ae0c53%40joeconway.com
Detect fclose() failures; given "ln -s /dev/full $PGDATA/devfull",
"pg_file_write('devfull', 'x', true)" now fails as it should. Don't
leak a stream when fwrite() fails. Remove a born-ineffective test that
aimed to skip zero-length writes. Back-patch to 9.2 (all supported
versions).
The logic for removing excess trigrams from the result was faulty.
It intends to avoid merging the initial and final states of the NFA,
which is necessary, but in testing whether removal of a specific trigram
would cause that, it failed to consider the combined effects of all the
state merges that that trigram's removal would cause. This could result
in a broken final graph that would never match anything, leading to GIN
or GiST indexscans not finding anything.
To fix, add a "tentParent" field that is used only within this loop,
and set it to show state merges that we are tentatively going to do.
While examining a particular arc, we must chase up through tentParent
links as well as regular parent links (the former can only appear atop
the latter), and we must account for state init/fin flag merges that
haven't actually been done yet.
To simplify the latter, combine the separate init and fin bool fields
into a bitmap flags field. I also chose to get rid of the "children"
state list, which seems entirely inessential.
Per bug #14563 from Alexey Isayko, which the added test cases are based on.
Back-patch to 9.3 where this code was added.
Report: https://postgr.es/m/20170222111446.1256.67547@wrigleys.postgresql.org
Discussion: https://postgr.es/m/8816.1487787594@sss.pgh.pa.us
This fixes problems where a plan must change but fails to do so,
as seen in a bug report from Rajkumar Raghuwanshi.
For ALTER FOREIGN TABLE OPTIONS, do this through the standard method of
forcing a relcache flush on the table. For ALTER FOREIGN DATA WRAPPER
and ALTER SERVER, just flush the whole plan cache on any change in
pg_foreign_data_wrapper or pg_foreign_server. That matches the way
we handle some other low-probability cases such as opclass changes, and
it's unclear that the case arises often enough to be worth working harder.
Besides, that gives a patch that is simple enough to back-patch with
confidence.
Back-patch to 9.3. In principle we could apply the code change to 9.2 as
well, but (a) we lack postgres_fdw to test it with, (b) it's doubtful that
anyone is doing anything exciting enough with FDWs that far back to need
this desperately, and (c) the patch doesn't apply cleanly.
Patch originally by Amit Langote, reviewed by Etsuro Fujita and Ashutosh
Bapat, who each contributed substantial changes as well.
Discussion: https://postgr.es/m/CAKcux6m5cA6rRPTKkqVdJ-R=KKDfe35Q_ZuUqxDSV_4hwga=og@mail.gmail.com
When libpq encounters a connection-level error, e.g. runs out of memory
while forming a result, there will be no error associated with PGresult,
but a message will be placed into PGconn's error buffer. postgres_fdw
takes care to use the PGconn error message when PGresult does not have
one, but dblink has been negligent in that regard. Modify dblink to mirror
what postgres_fdw has been doing.
Back-patch to all supported branches.
Author: Joe Conway
Reviewed-By: Tom Lane
Discussion: https://postgr.es/m/02fa2d90-2efd-00bc-fefc-c23c00eb671e%40joeconway.com
When dblink uses a postgres_fdw server name for its connection, it
is possible for the connection to have options that are invalid
with dblink (e.g. "updatable"). The recommended way to avoid this
problem is to use dblink_fdw servers instead. However there are use
cases for using postgres_fdw, and possibly other FDWs, for dblink
connection options, therefore protect against trying to use any
options that do not apply by using is_valid_dblink_option() when
building the connection string from the options.
Back-patch to 9.3. Although 9.2 supports FDWs for connection info,
is_valid_dblink_option() did not yet exist, and neither did
postgres_fdw, at least in the postgres source tree. Given the lack
of previous complaints, fixing that seems too invasive/not worth it.
Author: Corey Huinker
Reviewed-By: Joe Conway
Discussion: https://postgr.es/m/CADkLM%3DfWyXVEyYcqbcRnxcHutkP45UHU9WD7XpdZaMfe7S%3DRwA%40mail.gmail.com
Before commit b8cc8f947, it was possible to build contrib/uuid-ossp without
having told configure you meant to; you could just cd into that directory
and "make". That no longer works because the code depends on configure to
have done header and library probes, but the ensuing error messages are
not so easy to interpret if you're not an old C hand. We've gotten a
couple of complaints recently from people trying to do this the low-tech
way, so add an explicit #error directing the user to use --with-uuid.
(In principle we might want to do something similar in the other
optionally-built contrib modules; but I don't think any of the others have
ever worked without preconfiguration, so there are no bad habits to break
people of.)
Back-patch to 9.4 where the previous commit came in.
Report: https://postgr.es/m/CAHeEsBf42AWTnk=1qJvFv+mYgRFm07Knsfuc86Ono8nRjf3tvQ@mail.gmail.com
Report: https://postgr.es/m/CAKYdkBrUaZX+F6KpmzoHqMtiUqCtAW_w6Dgvr6F0WTiopuGxow@mail.gmail.com
When dblink or postgres_fdw detects an error on the remote side of the
connection, it will try to construct a local error message as best it
can using libpq's PQresultErrorField(). When no primary message is
available, it was bailing out with an unhelpful "unknown error". Make
that message better and more style guide compliant. Per discussion
on hackers.
Backpatch to 9.2 except postgres_fdw which didn't exist before 9.3.
Discussion: https://postgr.es/m/19872.1482338965%40sss.pgh.pa.us
Oversight in commit b8cc8f947. I just noticed this causes compiler
warnings on FreeBSD, and it really ought to cause warnings elsewhere too:
all references I can find say that <arpa/inet.h> is required for these.
We have a lot of code elsewhere that thinks that both <netinet/in.h>
and <arpa/inet.h> should be included for these functions, so do it that
way here too, even though <arpa/inet.h> ought to be sufficient according
to the references I consulted.
Back-patch to 9.4 where the previous commit landed.
Previously, it was thought that this only needed to be done for the
benefit of possible standbys, so wal_level = minimal skipped it.
But that's not safe, because during crash recovery we might replay
XLOG_DBASE_CREATE or XLOG_TBLSPC_CREATE record which recursively
removes the directory that contains the new init fork. So log it
always.
The user-visible effect of this bug is that if you create a database
or tablespace, then create an unlogged table, then crash without
checkpointing, then restart, accessing the table will fail, because
the it won't have been properly reset. This commit fixes that.
Michael Paquier, per a report from Konstantin Knizhnik. Wording of
the comments per a suggestion from me.
A little COLLATE "C" goes a long way.
Pavel Stehule, per suggestion from Craig Ringer
Discussion: <CAFj8pRA8nJZcozgxN=RMSqMmKuHVOkcGAAKPKdFeiMWGDSUDLA@mail.gmail.com>
Somebody apparently thought that "if Int32GetDatum is good,
Int64GetDatum must be better". Per buildfarm failures now
that Peter has added some regression tests here.
Before initializing iteration over a subtransaction's changes, the last
few changes were not spilled to disk. That's correct if the transaction
didn't spill to disk, but otherwise... This bug can lead to missed or
misorderd subtransaction contents when they were spilled to disk.
Move spilling of the remaining in-memory changes to
ReorderBufferIterTXNInit(), where it can easily be applied to the top
transaction and, if present, subtransactions.
Since this code had too many bugs already, noticeably increase test
coverage.
Fixes: #14319
Reported-By: Huan Ruan
Discussion: <20160909012610.20024.58169@wrigleys.postgresql.org>
Backport: 9,4-, where logical decoding was added
collect_corrupt_items() failed to initialize tuple.t_self. While
HeapTupleSatisfiesVacuum() doesn't actually use that value, it does
Assert that it's valid, so that the code would dump core if ip_posid
chanced to be zero. (That's somewhat unlikely, which probably explains
how this got missed. In any case it wouldn't matter for field use.)
Also, collect_corrupt_items was returning the wrong TIDs, that is the
contents of t_ctid rather than the tuple's own location. This would
be the same thing in simple cases, but it could be wrong if, for
example, a past update attempt had been rolled back, leaving a live
tuple whose t_ctid doesn't point at itself.
Also, in pg_visibility(), guard against trying to read a page past
the end of the rel. The VM code handles inquiries beyond the end
of the map by silently returning zeroes, and it seems like we should
do the same thing here.
I ran into the assertion failure while using pg_visibility to check
pg_upgrade's behavior, and then noted the other problems while
reading the code.
Report: <29043.1475288648@sss.pgh.pa.us>
LibreSSL defines OPENSSL_VERSION_NUMBER to claim that it is version 2.0.0,
but it doesn't have the functions added in OpenSSL 1.1.0. Add autoconf
checks for the individual functions we need, and stop relying on
OPENSSL_VERSION_NUMBER.
Backport to 9.5 and 9.6, like the patch that broke this. In the
back-branches, there are still a few OPENSSL_VERSION_NUMBER checks left,
to check for OpenSSL 0.9.8 or 0.9.7. I left them as they were - LibreSSL
has all those functions, so they work as intended.
Per buildfarm member curculio.
Discussion: <2442.1473957669@sss.pgh.pa.us>
Otherwise, users who have configured shared_buffers >= 256GB won't
be able to use this module. There probably aren't many of those, but
it doesn't hurt anything to fix it so that it works.
Backpatch to 9.4, where MemoryContextAllocHuge was introduced. The
same problem exists in older branches, but there's no easy way to
fix it there.
KaiGai Kohei
Changes needed to build at all:
- Check for SSL_new in configure, now that SSL_library_init is a macro.
- Do not access struct members directly. This includes some new code in
pgcrypto, to use the resource owner mechanism to ensure that we don't
leak OpenSSL handles, now that we can't embed them in other structs
anymore.
- RAND_SSLeay() -> RAND_OpenSSL()
Changes that were needed to silence deprecation warnings, but were not
strictly necessary:
- RAND_pseudo_bytes() -> RAND_bytes().
- SSL_library_init() and OpenSSL_config() -> OPENSSL_init_ssl()
- ASN1_STRING_data() -> ASN1_STRING_get0_data()
- DH_generate_parameters() -> DH_generate_parameters()
- Locking callbacks are not needed with OpenSSL 1.1.0 anymore. (Good
riddance!)
Also change references to SSLEAY_VERSION_NUMBER with OPENSSL_VERSION_NUMBER,
for the sake of consistency. OPENSSL_VERSION_NUMBER has existed since time
immemorial.
Fix SSL test suite to work with OpenSSL 1.1.0. CA certificates must have
the "CA:true" basic constraint extension now, or OpenSSL will refuse them.
Regenerate the test certificates with that. The "openssl" binary, used to
generate the certificates, is also now more picky, and throws an error
if an X509 extension is specified in "req_extensions", but that section
is empty.
Backpatch to 9.5 and 9.6, per popular demand. The file structure was
somewhat different in earlier branches, so I didn't bother to go further
than that. In back-branches, we still support OpenSSL 0.9.7 and above.
OpenSSL 0.9.6 should still work too, but I didn't test it. In master, we
only support 0.9.8 and above.
Patch by Andreas Karlsson, with additional changes by me.
Discussion: <20160627151604.GD1051@msg.df7cb.de>
I found that half a dozen (nearly 5%) of our AllocSetContextCreate calls
had typos in the context-sizing parameters. While none of these led to
especially significant problems, they did create minor inefficiencies,
and it's now clear that expecting people to copy-and-paste those calls
accurately is not a great idea. Let's reduce the risk of future errors
by introducing single macros that encapsulate the common use-cases.
Three such macros are enough to cover all but two special-purpose contexts;
those two calls can be left as-is, I think.
While this patch doesn't in itself improve matters for third-party
extensions, it doesn't break anything for them either, and they can
gradually adopt the simplified notation over time.
In passing, change TopMemoryContext to use the default allocation
parameters. Formerly it could only be extended 8K at a time. That was
probably reasonable when this code was written; but nowadays we create
many more contexts than we did then, so that it's not unusual to have a
couple hundred K in TopMemoryContext, even without considering various
dubious code that sticks other things there. There seems no good reason
not to let it use growing blocks like most other contexts.
Back-patch to 9.6, mostly because that's still close enough to HEAD that
it's easy to do so, and keeping the branches in sync can be expected to
avoid some future back-patching pain. The bugs fixed by these changes
don't seem to be significant enough to justify fixing them further back.
Discussion: <21072.1472321324@sss.pgh.pa.us>
As implemented, -e ran an EXPLAIN but then discarded the output, which
certainly seems pointless. Make it print to stdout instead. It's been
like that forever, so back-patch to all supported branches.
Daniel Gustafsson, reviewed by Andreas Scherbaum
Patch: <B97BDCB7-A3B3-4734-90B5-EDD586941629@yesql.se>
In blinsert(), cope with the possibility that a page we pull from the
notFullPage list is marked BLOOM_DELETED. This could happen if VACUUM
recently marked it deleted but hasn't (yet) updated the metapage.
We can re-use such a page safely, but we *must* reinitialize it so that
it's no longer marked deleted.
Fix blvacuum() so that it updates the notFullPage list even if it's
going to update it to empty. The previous "optimization" of skipping
the update seems pretty dubious, since it means that the next blinsert()
will uselessly visit whatever pages we left in the list.
Uniformly treat PageIsNew pages the same as deleted pages. This should
allow proper recovery if a crash occurs just after relation extension.
Properly use vacuum_delay_point, not assorted ad-hoc CHECK_FOR_INTERRUPTS
calls, in the blvacuum() main loop.
Fix broken tuple-counting logic: blvacuum.c counted the number of live
index tuples over again in each scan, leading to VACUUM VERBOSE reporting
some multiple of the actual number of surviving index tuples after any
vacuum that removed any tuples (since they'd be counted in blvacuum, maybe
more than once, and then again in blvacuumcleanup, without ever zeroing the
counter). It's sufficient to count them in blvacuumcleanup.
stats->estimated_count is a boolean, not a counter, and we don't want
to set it true, so don't add tuple counts to it.
Add a couple of Asserts that we don't overrun available space on a bloom
page. I don't think there's any bug there today, but the way the
FreeBlockNumberArray size calculation is set up is scarily fragile, and
BloomPageGetFreeSpace isn't much better. The Asserts should help catch
any future mistakes.
Per investigation of a report from Jeff Janes. I think the first item
above may explain his report; the other changes were things I noticed
while casting about for an explanation.
Report: <CAMkU=1xEUuBphDwDmB1WjN4+td4kpnEniFaTBxnk1xzHCw8_OQ@mail.gmail.com>
Per discussion, we should provide such functions to replace the lost
ability to discover AM properties by inspecting pg_am (cf commit
65c5fcd35). The added functionality is also meant to displace any code
that was looking directly at pg_index.indoption, since we'd rather not
believe that the bit meanings in that field are part of any client API
contract.
As future-proofing, define the SQL API to not assume that properties that
are currently AM-wide or index-wide will remain so unless they logically
must be; instead, expose them only when inquiring about a specific index
or even specific index column. Also provide the ability for an index
AM to override the behavior.
In passing, document pg_am.amtype, overlooked in commit 473b93287.
Andrew Gierth, with kibitzing by me and others
Discussion: <87mvl5on7n.fsf@news-spur.riddles.org.uk>
This oversight could cause logical decoding to fail to decode an outer
transaction containing changes, if a subtransaction had an XID but no
actual changes. Per bug #14279 from Marko Tiikkaja. Patch by Marko
based on analysis by Andrew Gierth.
Discussion: <20160804191757.1430.39011@wrigleys.postgresql.org>
Commits 4452000f3 et al established semantics for NullTest.argisrow that
are a bit different from its initial conception: rather than being merely
a cache of whether we've determined the input to have composite type,
the flag now has the further meaning that we should apply field-by-field
testing as per the standard's definition of IS [NOT] NULL. If argisrow
is false and yet the input has composite type, the construct instead has
the semantics of IS [NOT] DISTINCT FROM NULL. Update the comments in
primnodes.h to clarify this, and fix ruleutils.c and deparse.c to print
such cases correctly. In the case of ruleutils.c, this merely results in
cosmetic changes in EXPLAIN output, since the case can't currently arise
in stored rules. However, it represents a live bug for deparse.c, which
would formerly have sent a remote query that had semantics different
from the local behavior. (From the user's standpoint, this means that
testing a remote nested-composite column for null-ness could have had
unexpected recursive behavior much like that fixed in 4452000f3.)
In a related but somewhat independent fix, make plancat.c set argisrow
to false in all NullTest expressions constructed to represent "attnotnull"
constructs. Since attnotnull is actually enforced as a simple null-value
check, this is a more accurate representation of the semantics; we were
previously overpromising what it meant for composite columns, which might
possibly lead to incorrect planner optimizations. (It seems that what the
SQL spec expects a NOT NULL constraint to mean is an IS NOT NULL test, so
arguably we are violating the spec and should fix attnotnull to do the
other thing. If we ever do, this part should get reverted.)
Back-patch, same as the previous commit.
Discussion: <10682.1469566308@sss.pgh.pa.us>
That script is incorrect in that it sets the combine function for
max(citext) twice instead of setting the combine function for
max(citext) once and the combine functon for min(citext) once. The
consequence is that if you install 1.0 or 1.1 and then update to 1.2,
you end up with min(citext) not having a combine function, contrary to
what was intended. If you install 1.2 directly, you're OK.
Fix things up by defining a new 1.3 version. Upgrading from 1.2 to
1.3 won't change anything for people who first installed the 1.2
version, but people upgrading from 1.0 or 1.1 will get the right
catalog contents once they reach 1.3.
Report and patch by David Rowley, reviewed by Andreas Karlsson.