Cause parallel workers to not check datallowconn, rolcanlogin, and
ACL_CONNECT privileges. The leader already checked these things
(except for rolcanlogin which might have been checked for a different
role). Re-checking can accomplish little except to induce unexpected
failures in applications that might not even be aware that their query
has been parallelized. We already had the principle that parallel
workers rely on their leader to pass a valid set of authorization
information, so this change just extends that a bit further.
Also, modify the ReservedConnections, datconnlimit and rolconnlimit
logic so that these limits are only enforced against regular backends,
and only regular backends are counted while checking if the limits
were already reached. Previously, background processes that had an
assigned database or role were subject to these limits (with rather
random exclusions for autovac workers and walsenders), and the set of
existing processes that counted against each limit was quite haphazard
as well. The point of these limits, AFAICS, is to ensure the
availability of PGPROC slots for regular backends. Since all other
types of processes have their own separate pools of PGPROC slots, it
makes no sense either to enforce these limits against them or to count
them while enforcing the limit.
While edge-case failures of these sorts have been possible for a
long time, the problem got a good deal worse with commit 5a2fed911
(CVE-2024-10978), which caused parallel workers to make some of these
checks using the leader's current role where before we had used its
AuthenticatedUserId, thus allowing parallel queries to fail after
SET ROLE. The previous behavior was fairly accidental and I have
no desire to return to it.
This patch includes reverting 73c9f91a1, which was an emergency hack
to suppress these same checks in some cases. It wasn't complete,
as shown by a recent bug report from Laurenz Albe. We can also revert
fd4d93d26 and 492217301, which hacked around the same problems in one
regression test.
In passing, remove the special case for autovac workers in
CheckMyDatabase; it seems cleaner to have AutoVacWorkerMain pass
the INIT_PG_OVERRIDE_ALLOW_CONNS flag, now that that does what's
needed.
Like 5a2fed911, back-patch to supported branches (which sadly no
longer includes v12).
Discussion: https://postgr.es/m/1808397.1735156190@sss.pgh.pa.us
Consistently reset so->scanBehind at the beginning of nbtree array
advancement, even during sktrig_required=false calls (calls where array
advancement is triggered by an unsatisfied non-required array scan key).
Otherwise, it's possible for queries to fail to return all relevant
tuples to the scan given a low-order required scan key that was
previously deemed "satisfied" by a truncated high key attribute value.
This only happened at the point where a later non-required array scan
key needed to be "advanced" once on the next leaf page (that is, once
the right sibling of the truncated high key page was reached).
The underlying issue was that later code within _bt_advance_array_keys
assumed that the so->scanBehind flag must have been set using the
current page's high key (not the previous page's high key). Any later
successful recheck call to _bt_check_compare would therefore spuriously
be prevented from making _bt_advance_array_keys return true, based on
the faulty belief that the truncated attribute must be from the scan's
current tuple (i.e. the non-pivot tuple at the start of the next page).
_bt_advance_array_keys would return false for the tuple, ultimately
resulting in _bt_checkkeys failing to return a matching tuple.
Oversight in commit 5bf748b8, which enhanced nbtree ScalarArrayOp
execution.
Author: Peter Geoghegan <pg@bowt.ie>
Discussion: https://postgr.es/m/CAH2-WzkJKncfqyAUTeuB5GgRhT1vhsWO2q11dbZNqKmvjopP_g@mail.gmail.com
Backpatch: 17-, where commit 5bf748b8 first appears.
parallel_vacuum_reset_dead_items used a local variable to hold a
pointer from the passed vacrel, purely as a shorthand. This pointer
was later freed and a new allocation was made and stored to the
struct. Then the local pointer was mistakenly referenced again.
This apparently happened not to break anything since the freed chunk
would have been put on the context's freelist, so it was accidentally
the same pointer anyway, in which case the DSA handle was correctly
updated. The minimal fix is to change two places so they access
dead_items through the vacrel. This coding style is a maintenance
hazard, so while at it get rid of most other similar usages, which
were inconsistently used anyway.
Analysis and patch by Vallimaharajan G, with further defensive coding
by me
Backpath to v17, when TidStore came in
Discussion: https://postgr.es/m/1936493cc38.68cb2ef27266.7456585136086197135@zohocorp.com
When using a pipeline, a transaction starts from the first command and
is committed with a Sync message or when the pipeline ends.
Functions like IsInTransactionBlock() or PreventInTransactionBlock()
were already able to understand a pipeline as being in a transaction
block, but it was not the case of CheckTransactionBlock(). This
function is called for example to generate a WARNING for SET LOCAL,
complaining that it is used outside of a transaction block.
The current state of the code caused multiple problems, like:
- SET LOCAL executed at any stage of a pipeline issued a WARNING, even
if the command was at least second in line where the pipeline is in a
transaction state.
- LOCK TABLE failed when invoked at any step of a pipeline, even if it
should be able to work within a transaction block.
The pipeline protocol assumes that the first command of a pipeline is
not part of a transaction block, and that any follow-up commands is
considered as within a transaction block.
This commit changes the backend so as an implicit transaction block is
started each time the first Execute message of a pipeline has finished
processing, with this implicit transaction block ended once a sync is
processed. The checks based on XACT_FLAGS_PIPELINING in the routines
checking if we are in a transaction block are not necessary: it is
enough to rely on the existing ones.
Some tests are added to pgbench, that can be backpatched down to v17
when \syncpipeline is involved and down to v14 where \startpipeline and
\endpipeline are available. This is unfortunately limited regarding the
error patterns that can be checked, but it provides coverage for various
pipeline combinations to check if these succeed or fail. These tests
are able to capture the case of SET LOCAL's WARNING. The author has
proposed a different feature to improve the coverage by adding similar
meta-commands to psql where error messages could be checked, something
more useful for the cases where commands cannot be used in transaction
blocks, like REINDEX CONCURRENTLY or VACUUM. This is considered as
future work for v18~.
Author: Anthonin Bonnefoy
Reviewed-by: Jelte Fennema-Nio, Michael Paquier
Discussion: https://postgr.es/m/CAO6_XqrWO8uNBQrSu5r6jh+vTGi5Oiyk4y8yXDORdE2jbzw8xw@mail.gmail.com
Backpatch-through: 13
The SQL spec mandates that SET SESSION AUTHORIZATION implies
SET ROLE NONE. We tried to implement that within the lowest-level
functions that manipulate these settings, but that was a bad idea.
In particular, guc.c assumes that it doesn't matter in what order
it applies GUC variable updates, but that was not the case for these
two variables. This problem, compounded by some hackish attempts to
work around it, led to some security-grade issues:
* Rolling back a transaction that had done SET SESSION AUTHORIZATION
would revert to SET ROLE NONE, even if that had not been the previous
state, so that the effective user ID might now be different from what
it had been.
* The same for SET SESSION AUTHORIZATION in a function SET clause.
* If a parallel worker inspected current_setting('role'), it saw
"none" even when it should see something else.
Also, although the parallel worker startup code intended to cope
with the current role's pg_authid row having disappeared, its
implementation of that was incomplete so it would still fail.
Fix by fully separating the miscinit.c functions that assign
session_authorization from those that assign role. To implement the
spec's requirement, teach set_config_option itself to perform "SET
ROLE NONE" when it sets session_authorization. (This is undoubtedly
ugly, but the alternatives seem worse. In particular, there's no way
to do it within assign_session_authorization without incompatible
changes in the API for GUC assign hooks.) Also, improve
ParallelWorkerMain to directly set all the relevant user-ID variables
instead of relying on some of them to get set indirectly. That
allows us to survive not finding the pg_authid row during worker
startup.
In v16 and earlier, this includes back-patching 9987a7bf3 which
fixed a violation of GUC coding rules: SetSessionAuthorization
is not an appropriate place to be throwing errors from.
Security: CVE-2024-10978
Commit ac04aa84a put the shutoff for this into the planner, which is
not ideal because it doesn't prevent us from re-using a previously
made parallel plan. Revert the planner change and instead put the
shutoff into InitializeParallelDSM, modeling it on the existing code
there for recovering from failure to allocate a DSM segment.
However, that code path is mostly untested, and testing a bit harder
showed there's at least one bug: ExecHashJoinReInitializeDSM is not
prepared for us to have skipped doing parallel DSM setup. I also
thought the Assert in ReinitializeParallelWorkers is pretty
ill-advised, and replaced it with a silent Min() operation.
The existing test case added by ac04aa84a serves fine to test this
version of the fix, so no change needed there.
Patch by me, but thanks to Noah Misch for the core idea that we
could shut off worker creation when !INTERRUPTS_CAN_BE_PROCESSED.
Back-patch to v12, as ac04aa84a was.
Discussion: https://postgr.es/m/CAC-SaSzHUKT=vZJ8MPxYdC_URPfax+yoA1hKTcF4ROz_Q6z0_Q@mail.gmail.com
This reverts commit 95c5acb3fc (v17) and
counterparts in each other non-master branch. If released, that commit
would have caused a worst-in-years minor release regression, via
undetected LWLock self-deadlock. This commit and its self-deadlock fix
warrant more bake time in the master branch.
Reported by Alexander Lakhin.
Discussion: https://postgr.es/m/10ec0bc3-5933-1189-6bb8-5dec4114558e@gmail.com
A bug in nbtree's handling of primitive index scan scheduling could lead
to wrong answers when a scrollable cursor was used with an index scan
that had a SAOP index qual. Wrong answers were only possible when the
scan direction changed after a primitive scan was scheduled, but before
_bt_next was asked to fetch the next tuple in line (i.e. for things to
break, _bt_next had to be denied the opportunity to step off the page in
the same direction as the one used when the primscan was scheduled).
Furthermore, the issue only occurred when the page in question happened
to be the first page to be visited by the entire top-level scan; the
issue hinged upon the cursor backing up to the absolute beginning of the
key space that it returns tuples from (fetching in the opposite scan
direction across a "primitive scan boundary" always worked correctly).
To fix, make _bt_next unset the "needs primitive index scan" flag when
it detects that the current scan direction is not the one that was used
by _bt_readpage back when the primitive scan in question was scheduled.
This fixes the cases that are known to be faulty, and also seems like a
good idea on general robustness grounds.
Affected scrollable cursor cases now avoid a spurious primitive index
scan when they fetch backwards to the absolute start of the key space to
be visited by their cursor. Fetching backwards now only returns those
tuples at the start of the scan, as expected. It'll also be okay to
once again fetch forwards from the start at that point, since the scan
will be left in a state that's exactly consistent with the state it was
in before any tuples were ever fetched, as expected.
Oversight in commit 5bf748b8, which enhanced nbtree ScalarArrayOp
execution.
Author: Peter Geoghegan <pg@bowt.ie>
Discussion: https://postgr.es/m/CAH2-Wznv49bFsE2jkt4GuZ0tU2C91dEST=50egzjY2FeOcHL4Q@mail.gmail.com
Backpatch: 17-, where commit 5bf748b8 first appears.
Commit a07e03fd8f changed inplace updates
to wait for heap_update() commands like GRANT TABLE and GRANT DATABASE.
By keeping the pin during that wait, a sequence of autovacuum workers
and an uncommitted GRANT starved one foreground LockBufferForCleanup()
for six minutes, on buildfarm member sarus. Prevent, at the cost of a
bit of complexity. Back-patch to v12, like the earlier commit. That
commit and heap_inplace_lock() have not yet appeared in any release.
Discussion: https://postgr.es/m/20241026184936.ae.nmisch@google.com
A buffer lock won't stop a reader having already checked tuple
visibility. If a vac_update_datfrozenid() and then a crash happened
during inplace update of a relfrozenxid value, datfrozenxid could
overtake relfrozenxid. That could lead to "could not access status of
transaction" errors. Back-patch to v12 (all supported versions). In
v14 and earlier, this also back-patches the assertion removal from
commit 7fcf2faf9c.
Discussion: https://postgr.es/m/20240620012908.92.nmisch@google.com
The inplace update survives ROLLBACK. The inval didn't, so another
backend's DDL could then update the row without incorporating the
inplace update. In the test this fixes, a mix of CREATE INDEX and ALTER
TABLE resulted in a table with an index, yet relhasindex=f. That is a
source of index corruption. Back-patch to v12 (all supported versions).
The back branch versions don't change WAL, because those branches just
added end-of-recovery SIResetAll(). All branches change the ABI of
extern function PrepareToInvalidateCacheTuple(). No PGXN extension
calls that, and there's no apparent use case in extensions.
Reviewed by Nitin Motiani and (in earlier versions) Andres Freund.
Discussion: https://postgr.es/m/20240523000548.58.nmisch@google.com
An inplace update's invalidation messages are part of its transaction's
commit record. However, the update survives even if its transaction
aborts or we stop recovery before replaying its transaction commit.
After recovery, a backend that started in recovery could update the row
without incorporating the inplace update. That could result in a table
with an index, yet relhasindex=f. That is a source of index corruption.
This bulk invalidation avoids the functional consequences. A future
change can fix the !RecoveryInProgress() scenario without changing the
WAL format. Back-patch to v17 - v12 (all supported versions). v18 will
instead add invalidations to WAL.
Discussion: https://postgr.es/m/20240618152349.7f.nmisch@google.com
During logical decoding of in-progress transactions, we perform the toast
table scan while fetching the default toast value for an attribute. We
forgot to initialize the flag during this scan to indicate that the system
table scan is in progress. We need this flag to ensure that during logical
decoding we never directly access the tableam or heap APIs because we check
for concurrent aborts only in systable_* APIs.
Reported-by: Alexander Lakhin
Author: Takeshi Ideriha, Hou Zhijie
Reviewed-by: Amit Kapila, Hou Zhijie
Backpatch-through: 14
Discussion: https://postgr.es/m/18641-6687273b7f15269d@postgresql.org
COMMIT PREPARED removes on-disk 2PC files near its end, but the state
checked if a file is on-disk or not gets read from shared memory while
not holding the two-phase state lock.
Because of that, there was a small window where a second backend doing a
PREPARE TRANSACTION could reuse the GlobalTransaction put back into the
2PC free list by the COMMIT PREPARED, overwriting the "ondisk" flag read
afterwards by the COMMIT PREPARED to decide if its on-disk two-phase
state file should be removed, preventing the file deletion.
This commit fixes this issue so as the "ondisk" flag in the
GlobalTransaction is read while holding the two-phase state lock, not
from shared memory after its entry has been added to the free list.
Orphaned two-phase state files flushed to disk after a checkpoint are
discarded at the beginning of recovery. However, a truncation of
pg_xact/ would make the startup process issue a FATAL when it cannot
read the SLRU page holding the state of the transaction whose 2PC file
was orphaned, which is a necessary step to decide if the 2PC file should
be removed or not. Removing manually the file would be necessary in
this case.
Issue introduced by effe7d9552, so backpatch all the way down.
Mea culpa.
Author: wuchengwen
Discussion: https://postgr.es/m/tencent_A7F059B5136A359625C7B2E4A386B3C3F007@qq.com
Backpatch-through: 12
The previous commit fixed some ways of losing an inplace update. It
remained possible to lose one when a backend working toward a
heap_update() copied a tuple into memory just before inplace update of
that tuple. In catalogs eligible for inplace update, use LOCKTAG_TUPLE
to govern admission to the steps of copying an old tuple, modifying it,
and issuing heap_update(). This includes MERGE commands. To avoid
changing most of the pg_class DDL, don't require LOCKTAG_TUPLE when
holding a relation lock sufficient to exclude inplace updaters.
Back-patch to v12 (all supported versions). In v13 and v12, "UPDATE
pg_class" or "UPDATE pg_database" can still lose an inplace update. The
v14+ UPDATE fix needs commit 86dc90056d,
and it wasn't worth reimplementing that fix without such infrastructure.
Reviewed by Nitin Motiani and (in earlier versions) Heikki Linnakangas.
Discussion: https://postgr.es/m/20231027214946.79.nmisch@google.com
As previously-added tests demonstrated, heap_inplace_update() could
instead update an unrelated tuple of the same catalog. It could lose
the update. Losing relhasindex=t was a source of index corruption.
Inplace-updating commands like VACUUM will now wait for heap_update()
commands like GRANT TABLE and GRANT DATABASE. That isn't ideal, but a
long-running GRANT already hurts VACUUM progress more just by keeping an
XID running. The VACUUM will behave like a DELETE or UPDATE waiting for
the uncommitted change.
For implementation details, start at the systable_inplace_update_begin()
header comment and README.tuplock. Back-patch to v12 (all supported
versions). In back branches, retain a deprecated heap_inplace_update(),
for extensions.
Reported by Smolkin Grigory. Reviewed by Nitin Motiani, (in earlier
versions) Heikki Linnakangas, and (in earlier versions) Alexander
Lakhin.
Discussion: https://postgr.es/m/CAMp+ueZQz3yDk7qg42hk6-9gxniYbp-=bG2mgqecErqR5gGGOA@mail.gmail.com
Commit 5bf748b8, which enhanced nbtree ScalarArrayOp execution, made
parallel index scans work with the new design for arrays via explicit
scheduling of primitive index scans. Under this scheme a parallel index
scan with array keys will perform the same number of index descents as
an equivalent serial index scan (barring corner cases where an
individual parallel worker discovers that it can advance the scan's
array keys without anybody needing to perform another descent of the
index to get to the relevant page on the leaf level).
Despite all this, the pgstats accounting wasn't updated; it continued to
increment the total number of index scans for the rel once per _bt_first
call, no matter the details. As a result, the number of (primitive)
index scans could be over-counted during parallel scans.
To fix, delay incrementing the count of index scans until after we've
established that another descent of the index (using either _bt_search
or _bt_endpoint) is required. That way pg_stat_user_tables.idx_scan
always advances in the same way, regardless of whether or not the scan
makes use of parallelism.
Oversight in commit 5bf748b8, which enhanced nbtree ScalarArrayOp
execution.
Author: Peter Geoghegan <pg@bowt.ie>
Reviewed-By: Tomas Vondra <tomas@vondra.me>
Discussion: https://postgr.es/m/CAH2-Wz=E7XrkvscBN0U6V81NK3Q-dQOmivvbEsjG-zwEfDdFpg@mail.gmail.com
Discussion: https://postgr.es/m/CAH2-WzkRqvaqR2CTNqTZP0z6FuL4-3ED6eQB0yx38XBNj1v-4Q@mail.gmail.com
Backpatch: 17-, where nbtree SAOP execution was enhanced.
Commit 5bf748b8, which enhanced nbtree ScalarArrayOp execution, made
parallel index scans work with the new design for arrays via explicit
scheduling of primitive index scans. A backend that successfully
scheduled the scan's next primitive index scan saved its backend local
array keys in shared memory. Any backend could pick up the scheduled
primitive scan within _bt_first. This scheme decouples scheduling a
primitive scan from starting the scan (by performing another descent of
the index via a _bt_search call from _bt_first) to make things robust.
The scheme had a deadlock hazard, at least when the leader process
participated in the scan. _bt_parallel_seize had a code path that made
backends that were not in an immediate position to start a scheduled
primitive index scan wait for some other backend to do so instead.
Under the right circumstances, the leader process could wait here
forever: the leader would wait for any other backend to start the
primitive scan, while every worker was busy waiting on the leader to
consume tuples from the scan's tuple queue.
To fix, don't wait for a scheduled primitive index scan to be started by
some other eligible backend from within _bt_parallel_seize (when the
calling backend isn't in a position to do so itself). Return false
instead, while recording that the scan has a scheduled primitive index
scan in backend local state. This leaves the backend in the same state
as the existing case where a backend schedules (or tries to schedule)
another primitive index scan from within _bt_advance_array_keys, before
calling _bt_parallel_seize. _bt_parallel_seize already handles that
case by returning false without waiting, and without unsetting the
backend local state. Leaving the backend in this state enables it to
start a previously scheduled primitive index scan once it gets back to
_bt_first.
Oversight in commit 5bf748b8, which enhanced nbtree ScalarArrayOp
execution.
Matthias van de Meent, with tweaks by me.
Author: Matthias van de Meent <boekewurm+postgres@gmail.com>
Reported-By: Tomas Vondra <tomas@vondra.me>
Reviewed-By: Peter Geoghegan <pg@bowt.ie>
Discussion: https://postgr.es/m/CAH2-WzmMGaPa32u9x_FvEbPTUkP5e95i=QxR8054nvCRydP-sw@mail.gmail.com
Backpatch: 17-, where nbtree SAOP execution was enhanced.
When we are building a hash index that is large enough to need
pre-sorting (larger than either maintenance_work_mem or NBuffers),
the initial sorting phase is interruptible, but the insertion
phase wasn't. Add the missing CHECK_FOR_INTERRUPTS().
Per bug #18616 from Alexander Lakhin. Back-patch to all
supported branches.
Pavel Borisov
Discussion: https://postgr.es/m/18616-acbb9e5caf41e964@postgresql.org
Add bounds checking to nbtree's lookahead/skip-within-a-page mechanism.
Otherwise it's possible for cases with lots of before-array-keys tuples
to overflow an int16 variable, causing the mechanism to generate an out
of bounds page offset number.
Oversight in commit 5bf748b8, which enhanced nbtree ScalarArrayOp
execution.
Reported-By: Alexander Lakhin <exclusion@gmail.com>
Discussion: https://postgr.es/m/6c68ac42-bbb5-8b24-103e-af0e279c536f@gmail.com
Backpatch: 17-, where nbtree SAOP execution was enhanced.
This reverts commit 5887dd4894.
Some buildfarm animals are failing with "cannot change
"client_encoding" during a parallel operation". It looks like
assign_client_encoding is unhappy at being asked to roll back a
client_encoding setting after a parallel worker encounters a
failure. There must be more to it though: why didn't I see this
during local testing? In any case, it's clear that moving the
RestoreGUCState() call is not as side-effect-free as I thought.
Given that the bug f5f30c22e intended to fix has gone unreported
for years, it's not something that's urgent to fix; I'm not
willing to risk messing with it further with only days to our
next release wrap.
Parallel workers failed after a sequence like
BEGIN;
CREATE USER foo;
SET SESSION AUTHORIZATION foo;
because check_session_authorization could not see the uncommitted
pg_authid row for "foo". This is because we ran RestoreGUCState()
in a separate transaction using an ordinary just-created snapshot.
The same disease afflicts any other GUC that requires catalog lookups
and isn't forgiving about the lookups failing.
To fix, postpone RestoreGUCState() into the worker's main transaction
after we've set up a snapshot duplicating the leader's. This affects
check_transaction_isolation and check_transaction_deferrable, which
think they should only run during transaction start. Make them
act like check_transaction_read_only, which already knows it should
silently accept the value when InitializingParallelWorker.
Per bug #18545 from Andrey Rachitskiy. Back-patch to all
supported branches, because this has been wrong for awhile.
Discussion: https://postgr.es/m/18545-feba138862f19aaa@postgresql.org
When a standby is promoted, CleanupAfterArchiveRecovery() may decide
to rename the final WAL file from the old timeline by adding ".partial"
to the name. If WAL summarization is enabled and this file is renamed
before its partial contents are summarized, WAL summarization breaks:
the summarizer gets stuck at that point in the WAL stream and just
errors out.
To fix that, first make the startup process wait for WAL summarization
to catch up before renaming the file. Generally, this should be quick,
and if it's not, the user can shut off summarize_wal and try again.
To make this fix work, also teach the WAL summarizer that after a
promotion has occurred, no more WAL can appear on the previous
timeline: previously, the WAL summarizer wouldn't switch to the new
timeline until we actually started writing WAL there, but that meant
that when the startup process was waiting for the WAL summarizer, it
was waiting for an action that the summarizer wasn't yet prepared to
take.
In the process of fixing these bugs, I realized that the logic to wait
for WAL summarization to catch up was spread out in a way that made
it difficult to reuse properly, so this code refactors things to make
it easier.
Finally, add a test case that would have caught this bug and the
previously-fixed bug that WAL summarization sometimes needs to back up
when the timeline changes.
Discussion: https://postgr.es/m/CA+TgmoZGEsZodXC4f=XZNkAeyuDmWTSkpkjCEOcF19Am0mt_OA@mail.gmail.com
clog.c, async.c and predicate.c included some SLRU page numbers still
handled as 4-byte integers, while int64 should be used for this purpose.
These holes have been introduced in 4ed8f0913b, that has introduced
the use of 8-byte integers for SLRU page numbers, still forgot about the
code paths updated by this commit.
Reported-by: Noah Misch
Author: Aleksander Alekseev, Michael Paquier
Discussion: https://postgr.es/m/20240626002747.dc.nmisch@google.com
Backpatch-through: 17
slru.h described incorrectly how SLRU segment names are formatted
depending on the segment number and if long or short segment names are
used. This commit closes the gap with a better description, fitting
with the reality.
Reported-by: Noah Misch
Author: Aleksander Alekseev
Discussion: https://postgr.es/m/20240626002747.dc.nmisch@google.com
Backpatch-through: 17
If vacuum fails to remove a tuple with xmax older than
VacuumCutoffs->OldestXmin and younger than GlobalVisState->maybe_needed,
it may attempt to freeze the tuple's xmax and then ERROR out in
pre-freeze checks with "cannot freeze committed xmax".
Fix this by having vacuum always remove tuples older than OldestXmin.
It is possible for GlobalVisState->maybe_needed to precede OldestXmin if
maybe_needed is forced to go backward while vacuum is running. This can
happen if a disconnected standby with a running transaction older than
VacuumCutoffs->OldestXmin reconnects to the primary after vacuum
initially calculates GlobalVisState and OldestXmin.
In back branches starting with 14, the first version using
GlobalVisState, failing to remove tuples older than OldestXmin during
pruning caused vacuum to infinitely loop in lazy_scan_prune(), as
investigated on this [1] thread. After 1ccc1e05ae removed the retry loop
in lazy_scan_prune() and stopped comparing tuples to OldestXmin, the
hang could no longer happen, but we could still attempt to freeze dead
tuples with xmax older than OldestXmin -- resulting in an ERROR.
Fix this by always removing dead tuples with xmax older than
VacuumCutoffs->OldestXmin. This is okay because the standby won't replay
the tuple removal until the tuple is removable. Thus, the worst that can
happen is a recovery conflict.
[1] https://postgr.es/m/20240415173913.4zyyrwaftujxthf2%40awork3.anarazel.de#1b216b7768b5bd577a3d3d51bd5aadee
Back-patch through 14
Author: Melanie Plageman
Reviewed-by: Peter Geoghegan, Robert Haas, Andres Freund, Heikki Linnakangas, and Noah Misch
Discussion: https://postgr.es/m/CAAKRu_bDD7oq9ZwB2OJqub5BovMG6UjEYsoK2LVttadjEqyRGg%40mail.gmail.com
To do this, we must include the wal_level in the first WAL record
covered by each summary file; so add wal_level to struct Checkpoint
and the payload of XLOG_CHECKPOINT_REDO and XLOG_END_OF_RECOVERY.
This, in turn, requires bumping XLOG_PAGE_MAGIC and, since the
Checkpoint is also stored in the control file, also
PG_CONTROL_VERSION. It's not great to do that so late in the release
cycle, but the alternative seems to ship v17 without robust
protections against this scenario, which could result in corrupted
incremental backups.
A side effect of this patch is that, when a server with
wal_level=replica is started with summarize_wal=on for the first time,
summarization will no longer begin with the oldest WAL that still
exists in pg_wal, but rather from the first checkpoint after that.
This change should be harmless, because a WAL summary for a partial
checkpoint cycle can never make an incremental backup possible when
it would otherwise not have been.
Report by Fujii Masao. Patch by me. Review and/or testing by Jakub
Wartak and Fujii Masao.
Discussion: http://postgr.es/m/6e30082e-041b-4e31-9633-95a66de76f5d@oss.nttdata.com
Commit f4b54e1ed9, which introduced macros for protocol characters,
missed updating a few places. It also did not introduce macros for
messages sent from parallel workers to their leader processes.
This commit adds a new section in protocol.h for those.
Author: Aleksander Alekseev
Discussion: https://postgr.es/m/CAJ7c6TNTd09AZq8tGaHS3LDyH_CCnpv0oOz2wN1dGe8zekxrdQ%40mail.gmail.com
Backpatch-through: 17
Before this commit, when the WAL summarizer started up or recovered
from an error, it would resume summarization from wherever it left
off. That was OK normally, but wrong if summarize_wal=off had been
turned off temporary, allowing some WAL to be removed, and then turned
back on again. In such cases, the WAL summarizer would simply hang
forever. This commit changes the reinitialization sequence for WAL
summarizer to rederive the starting position in the way we were
already doing at initial startup, fixing the problem.
Per report from Israel Barth Rubio. Reviewed by Tom Lane.
Discussion: http://postgr.es/m/CA+TgmoYN6x=YS+FoFOS6=nr6=qkXZFWhdiL7k0oatGwug2hcuA@mail.gmail.com
Make the isolation harness recognize injection_points wait events as a
type of blocked state. Test an extant inplace-update bug.
Reviewed by Robert Haas and Michael Paquier.
Discussion: https://postgr.es/m/20240512232923.aa.nmisch@google.com
We did not recover the subtransaction IDs of prepared transactions
when starting a hot standby from a shutdown checkpoint. As a result,
such subtransactions were considered as aborted, rather than
in-progress. That would lead to hint bits being set incorrectly, and
the subtransactions suddenly becoming visible to old snapshots when
the prepared transaction was committed.
To fix, update pg_subtrans with prepared transactions's subxids when
starting hot standby from a shutdown checkpoint. The snapshots taken
from that state need to be marked as "suboverflowed", so that we also
check the pg_subtrans.
Backport to all supported versions.
Discussion: https://www.postgresql.org/message-id/6b852e98-2d49-4ca1-9e95-db419a2696e0@iki.fi
1. TruncateMultiXact() performs the SLRU truncations in a critical
section. Deleting the SLRU segments calls ForwardSyncRequest(), which
will try to compact the request queue if it's full
(CompactCheckpointerRequestQueue()). That in turn allocates memory,
which is not allowed in a critical section. Backtrace:
TRAP: failed Assert("CritSectionCount == 0 || (context)->allowInCritSection"), File: "../src/backend/utils/mmgr/mcxt.c", Line: 1353, PID: 920981
postgres: autovacuum worker template0(ExceptionalCondition+0x6e)[0x560a501e866e]
postgres: autovacuum worker template0(+0x5dce3d)[0x560a50217e3d]
postgres: autovacuum worker template0(ForwardSyncRequest+0x8e)[0x560a4ffec95e]
postgres: autovacuum worker template0(RegisterSyncRequest+0x2b)[0x560a50091eeb]
postgres: autovacuum worker template0(+0x187b0a)[0x560a4fdc2b0a]
postgres: autovacuum worker template0(SlruDeleteSegment+0x101)[0x560a4fdc2ab1]
postgres: autovacuum worker template0(TruncateMultiXact+0x2fb)[0x560a4fdbde1b]
postgres: autovacuum worker template0(vac_update_datfrozenxid+0x4b3)[0x560a4febd2f3]
postgres: autovacuum worker template0(+0x3adf66)[0x560a4ffe8f66]
postgres: autovacuum worker template0(AutoVacWorkerMain+0x3ed)[0x560a4ffe7c2d]
postgres: autovacuum worker template0(+0x3b1ead)[0x560a4ffecead]
postgres: autovacuum worker template0(+0x3b620e)[0x560a4fff120e]
postgres: autovacuum worker template0(+0x3b3fbb)[0x560a4ffeefbb]
postgres: autovacuum worker template0(+0x2f724e)[0x560a4ff3224e]
/lib/x86_64-linux-gnu/libc.so.6(+0x27c8a)[0x7f62cc642c8a]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0x85)[0x7f62cc642d45]
postgres: autovacuum worker template0(_start+0x21)[0x560a4fd16f31]
To fix, bail out in CompactCheckpointerRequestQueue() without doing
anything, if it's called in a critical section. That covers the above
call path, as well as any other similar cases where
RegisterSyncRequest might be called in a critical section.
2. After fixing that, another problem became apparent: Autovacuum
process doing that truncation can deadlock with the checkpointer
process. TruncateMultiXact() sets "MyProc->delayChkptFlags |=
DELAY_CHKPT_START". If the sync request queue is full and cannot be
compacted, the process will repeatedly sleep and retry, until there is
room in the queue. However, if the checkpointer is trying to start a
checkpoint at the same time, and is waiting for the DELAY_CHKPT_START
processes to finish, the queue will never shrink.
More concretely, the autovacuum process is stuck here:
#0 0x00007fc934926dc3 in epoll_wait () from /lib/x86_64-linux-gnu/libc.so.6
#1 0x000056220b24348b in WaitEventSetWaitBlock (set=0x56220c2e4b50, occurred_events=0x7ffe7856d040, nevents=1, cur_timeout=<optimized out>) at ../src/backend/storage/ipc/latch.c:1570
#2 WaitEventSetWait (set=0x56220c2e4b50, timeout=timeout@entry=10, occurred_events=<optimized out>, occurred_events@entry=0x7ffe7856d040, nevents=nevents@entry=1,
wait_event_info=wait_event_info@entry=150994949) at ../src/backend/storage/ipc/latch.c:1516
#3 0x000056220b243224 in WaitLatch (latch=<optimized out>, latch@entry=0x0, wakeEvents=wakeEvents@entry=40, timeout=timeout@entry=10, wait_event_info=wait_event_info@entry=150994949)
at ../src/backend/storage/ipc/latch.c:538
#4 0x000056220b26cf46 in RegisterSyncRequest (ftag=ftag@entry=0x7ffe7856d0a0, type=type@entry=SYNC_FORGET_REQUEST, retryOnError=true) at ../src/backend/storage/sync/sync.c:614
#5 0x000056220af9db0a in SlruInternalDeleteSegment (ctl=ctl@entry=0x56220b7beb60 <MultiXactMemberCtlData>, segno=segno@entry=11350) at ../src/backend/access/transam/slru.c:1495
#6 0x000056220af9dab1 in SlruDeleteSegment (ctl=ctl@entry=0x56220b7beb60 <MultiXactMemberCtlData>, segno=segno@entry=11350) at ../src/backend/access/transam/slru.c:1566
#7 0x000056220af98e1b in PerformMembersTruncation (oldestOffset=<optimized out>, newOldestOffset=<optimized out>) at ../src/backend/access/transam/multixact.c:3006
#8 TruncateMultiXact (newOldestMulti=newOldestMulti@entry=3221225472, newOldestMultiDB=newOldestMultiDB@entry=4) at ../src/backend/access/transam/multixact.c:3201
#9 0x000056220b098303 in vac_truncate_clog (frozenXID=749, minMulti=<optimized out>, lastSaneFrozenXid=749, lastSaneMinMulti=3221225472) at ../src/backend/commands/vacuum.c:1917
#10 vac_update_datfrozenxid () at ../src/backend/commands/vacuum.c:1760
#11 0x000056220b1c3f76 in do_autovacuum () at ../src/backend/postmaster/autovacuum.c:2550
#12 0x000056220b1c2c3d in AutoVacWorkerMain (startup_data=<optimized out>, startup_data_len=<optimized out>) at ../src/backend/postmaster/autovacuum.c:1569
and the checkpointer is stuck here:
#0 0x00007fc9348ebf93 in clock_nanosleep () from /lib/x86_64-linux-gnu/libc.so.6
#1 0x00007fc9348fe353 in nanosleep () from /lib/x86_64-linux-gnu/libc.so.6
#2 0x000056220b40ecb4 in pg_usleep (microsec=microsec@entry=10000) at ../src/port/pgsleep.c:50
#3 0x000056220afb43c3 in CreateCheckPoint (flags=flags@entry=108) at ../src/backend/access/transam/xlog.c:7098
#4 0x000056220b1c6e86 in CheckpointerMain (startup_data=<optimized out>, startup_data_len=<optimized out>) at ../src/backend/postmaster/checkpointer.c:464
To fix, add AbsorbSyncRequests() to the loops where the checkpointer
waits for DELAY_CHKPT_START or DELAY_CHKPT_COMPLETE operations to
finish.
Backpatch to v14. Before that, SLRU deletion didn't call
RegisterSyncRequest, which avoided this failure. I'm not sure if there
are other similar scenarios on older versions, but we haven't had
any such reports.
Discussion: https://www.postgresql.org/message-id/ccc66933-31c1-4f6a-bf4b-45fef0d4f22e@iki.fi
_bt_advance_array_keys didn't take sufficient care at the point where it
decides whether to start a new primitive index scan based on a call to
_bt_check_compare against finaltup (a call with the scan direction
flipped around). The final decision was conditioned on rules about how
the scan key offset sktrig that initially triggered array advancement
(passed to _bt_advance_array_keys from its _bt_checkkeys caller)
compared to the offset set by its own _bt_check_compare finaltup call.
This approach was faulty, in that it allowed _bt_advance_array_keys to
incorrectly start a new primitive index scan, that landed on the same
leaf page (on assert-enabled builds it led to an assertion failure).
In general, scans with array keys are expected to never have to read the
same leaf page more than once (barring cases involving cursors, and
cases where the scan restores a marked position for the inner side of a
merge join). This principle was established by commit 5bf748b8.
To fix, make the final decision based on whether the scan key offset set
by the _bt_check_compare finaltup call is an offset to an inequality
strategy scan key. An unsatisfied required inequality strategy scan key
indicates that all of the scan's required equality strategy scan keys
must also be satisfied by finaltup (not just by caller's tuple), and
that there is a decent chance that _bt_first will be able to reposition
the scan to a position many leaf pages ahead of the current leaf page.
Oversight in commit 5bf748b8.
Discussion: https://postgr.es/m/CAH2-Wz=DyHbcg7o6zXqzyiin8WE8vzk4tvU8Lrnh-a=EAvO0TQ@mail.gmail.com
Reconstruction of an SP-GiST index by REINDEX CONCURRENTLY may
insert some REDIRECT tuples. This will typically happen in
a transaction that lacks an XID, which leads either to assertion
failure in spgFormDeadTuple or to insertion of a REDIRECT tuple
with zero xid. The latter's not good either, since eventually
VACUUM will apply GlobalVisTestIsRemovableXid() to the zero xid,
resulting in either an assertion failure or a garbage answer.
In practice, since REINDEX CONCURRENTLY locks out index scans
till it's done, it doesn't matter whether it inserts REDIRECTs
or PLACEHOLDERs; and likewise it doesn't matter how soon VACUUM
reduces such a REDIRECT to a PLACEHOLDER. So in non-assert builds
there's no observable problem here, other than perhaps a little
index bloat. But it's not behaving as intended.
To fix, remove the failing Assert in spgFormDeadTuple, acknowledging
that we might sometimes insert a zero XID; and guard VACUUM's
GlobalVisTestIsRemovableXid() call with a test for valid XID,
ensuring that we'll reduce such a REDIRECT the first time VACUUM
sees it. (Versions before v14 use TransactionIdPrecedes here,
which won't fail on zero xid, so they really have no bug at all
in non-assert builds.)
Another solution could be to not create REDIRECTs at all during
REINDEX CONCURRENTLY, making the relevant code paths treat that
case like index build (which likewise knows that no concurrent
index scans can be happening). That would allow restoring the
Assert in spgFormDeadTuple, but we'd still need the VACUUM change
because redirection tuples with zero xid may be out there already.
But there doesn't seem to be a nice way for spginsert() to tell that
it's being called in REINDEX CONCURRENTLY without some API changes,
so we'll leave that as a possible future improvement.
In HEAD, also rename the SpGistState.myXid field to redirectXid,
which seems less misleading (since it might not in fact be our
transaction's XID) and is certainly less uninformatively generic.
Per bug #18499 from Alexander Lakhin. Back-patch to all supported
branches.
Discussion: https://postgr.es/m/18499-8a519c280f956480@postgresql.org
Commit 667e65aac3 changed both num_dead_tuples and max_dead_tuples
columns to dead_tuple_bytes and max_dead_tuple_bytes columns,
respectively. But as per discussion, the number of dead tuples
collected still provides meaningful insights for users.
This commit reintroduces the column for the count of dead tuples,
renamed as num_dead_item_ids. It avoids confusion with the number of
dead tuples removed by VACUUM, which includes dead heap-only tuples
but excludes any pre-existing LP_DEAD items left behind by
opportunistic pruning.
Bump catalog version.
Reviewed-by: Peter Geoghegan, Álvaro Herrera, Andrey Borodin
Discussion: https://postgr.es/m/CAD21AoBL5sJE9TRWPyv%2Bw7k5Ee5QAJqDJEDJBUdAaCzGWAdvZw%40mail.gmail.com
The purpose of the function is to reduce the effective
autovacuum_multixact_freeze_max_age if the multixact members SLRU is
approaching wraparound, to make multixid freezing more aggressive.
The returned value should therefore never be greater than plain
autovacuum_multixact_freeze_max_age.
Reviewed-by: Robert Haas
Discussion: https://www.postgresql.org/message-id/85fb354c-f89f-4d47-b3a2-3cbd461c90a3@iki.fi
Backpatch-through: 12, all supported versions
Make sure that function declarations use names that exactly match the
corresponding names from function definitions in a few places. These
inconsistencies were all introduced during Postgres 17 development.
pg_bsd_indent still has a couple of similar inconsistencies, which I
(pgeoghegan) have left untouched for now.
This commit was written with help from clang-tidy, by mechanically
applying the same rules as similar clean-up commits (the earliest such
commit was commit 035ce1fe).
ReorderBufferImmediateInvalidation() executes invalidation messages in
an aborted transaction. However, RelationFlushRelation sometimes
required a valid resource owner, to temporarily increment the refcount
of the relache entry. Commit b8bff07daa worked around that in the main
subtransaction abort function, AbortSubTransaction(), but missed this
similar case in ReorderBufferImmediateInvalidation().
To fix, introduce a separate function to invalidate a relcache
entry. It does the same thing as RelationClearRelation(rebuild==true)
does when outside a transaction, but can be called without
incrementing the refcount.
Add regression test. Before this fix, it failed with:
ERROR: ResourceOwnerEnlarge called after release started
Reported-by: Alexander Lakhin <exclusion@gmail.com>
Discussion: https://www.postgresql.org/message-id/e56be7d9-14b1-664d-0bfc-00ce9772721c@gmail.com
