Commit c6f2f01611d4f2c412e92eb7893f76fa590818e8 purported to make
this work, but problems remained. In a plain-format backup, the
files from an in-place tablespace got included in the tar file for
the main tablespace, which is wrong but it's not clear that it
has any user-visible consequences. In a tar-format backup, the
TABLESPACE_MAP option is used, and so we never iterated over
pg_tblspc and thus never backed up the in-place tablespaces
anywhere at all.
To fix this, reverse the changes in that commit, so that when we scan
pg_tblspc during a backup, we create tablespaceinfo objects even for
in-place tablespaces. We set the field that would normally contain the
absolute pathname to the relative path pg_tblspc/${TSOID}, and that's
good enough to make basebackup.c happy without any further changes.
However, pg_basebackup needs a couple of adjustments to make it work.
First, it needs to understand that a relative path for a tablespace
means it's an in-place tablespace. Second, it needs to tolerate the
situation where restoring the main tablespace tries to create
pg_tblspc or a subdirectory and finds that it already exists, because
we restore user-defined tablespaces before the main tablespace.
Since in-place tablespaces are only intended for use in development
and testing, no back-patch.
Patch by me, reviewed by Thomas Munro and Michael Paquier.
Discussion: http://postgr.es/m/CA+TgmobwvbEp+fLq2PykMYzizcvuNv0a7gPMJtxOTMOuuRLMHg@mail.gmail.com
If the last few pages in the specified range are empty (all zero),
then log_newpage_range() could try to emit an empty WAL record
containing no FPIs. This at least upsets an Assert in
ReserveXLogInsertLocation, and might perhaps have bad real-world
consequences in non-assert builds.
This has been broken since log_newpage_range() was introduced,
but the case was hard if not impossible to hit before commit 3d6a98457
decided it was okay to leave VM and FSM pages intentionally zero.
Nonetheless, it seems prudent to back-patch. log_newpage_range()
was added in v12 but later back-patched, so this affects all
supported branches.
Matthias van de Meent, per report from Justin Pryzby
Discussion: https://postgr.es/m/ZD1daibg4RF50IOj@telsasoft.com
Clarify the goals of the recently added guidelines for rmgrdesc authors:
to avoid gratuitous inconsistencies across resource managers, and to
make it reasonably easy to write a reusable custom parser.
Beyond that, the guidelines leave rmgrdesc authors with a significant
amount of leeway. This even includes the leeway to invent custom
conventions (in cases where it's warranted).
Follow-up to commit 7d8219a4.
Author: Peter Geoghegan <pg@bowt.ie>
Reviewed-By: Melanie Plageman <melanieplageman@gmail.com>
Discussion: https://postgr.es/m/CAH2-WzkbYuvwYKm-Y-72QEh6SPMQcAo9uONv+mR3bMGcu9E_Cg@mail.gmail.com
Make heap desc routines that output status bit as arrays of constants
avoid outputting array literals that contain superfluous punctuation
characters that complicate parsing the output. Also make sure that no
heap desc routine repeats the same key name (at the same nesting level),
for the same reason. Arguably, these were both oversights in commit
7d8219a4.
In passing, make the desc output code (which covers Heap's DELETE,
UPDATE, HOT_UPDATE, LOCK, and LOCK_UPDATED record types) consistent in
terms of the output order of each field. This order also matches WAL
record struct order. Heap's DELETE desc output now shows the record's
xmax field for the first time (just like UPDATE/HOT_UPDATE records).
Author: Peter Geoghegan <pg@bowt.ie>
Reviewed-By: Melanie Plageman <melanieplageman@gmail.com>
Discussion: https://postgr.es/m/CAH2-Wz=pNYtxiJ2Jx5Lj=fKo1OEZ4GE0p_kct+ugAUTqBwU46g@mail.gmail.com
Make xl_heap_lock's infobits_set field of type uint8, not int8. Using
int8 isn't appropriate given that the field just holds status bits.
This fixes an oversight in commit 0ac5ad5134.
In passing rename the nearby TransactionId field to "xmax" to make
things consistency with related records, such as xl_heap_lock_updated.
Deliberately avoid a bump in XLOG_PAGE_MAGIC. No backpatch, either.
Author: Peter Geoghegan <pg@bowt.ie>
Discussion: https://postgr.es/m/CAH2-WzkCd3kOS8b7Rfxw7Mh1_6jvX=Nzo-CWR1VBTiOtVZkWHA@mail.gmail.com
We cannot use the generic array_desc approach with per-tuple nbtree
posting list update metadata because array_desc can only deal with fixed
width elements (e.g., page offset numbers). Using array_desc led to
incorrect rmgr descriptions for updates from nbtree DELETE/VACUUM WAL
records.
To fix, add specialized code to describe the update metadata as array
elements in desc output. We now iterate over the update metadata using
an approach that matches related REDO routines.
Also stop showing the updates offset number array separately in nbtree
DELETE/VACUUM desc output. It's redundant information, since the same
page offset numbers appear in the description of each individual update
element. Also make some small tweaks to the way that we format arrays
in all desc routines (not just nbtree desc routines) to make arrays a
little less verbose.
Oversight in commit 1c453cfd, which enhanced the nbtree rmgr desc
routines.
Author: Peter Geoghegan <pg@bowt.ie>
Discussion: https://postgr.es/m/CAH2-WzkbYuvwYKm-Y-72QEh6SPMQcAo9uONv+mR3bMGcu9E_Cg@mail.gmail.com
Unsurprisingly, this requires wal_level = logical to be set on the primary and
standby. The infrastructure added in 26669757b6a ensures that slots are
invalidated if the primary's wal_level is lowered.
Creating a slot on a standby waits for a xl_running_xact record to be
processed. If the primary is idle (and thus not emitting xl_running_xact
records), that can take a while. To make that faster, this commit also
introduces the pg_log_standby_snapshot() function. By executing it on the
primary, completion of slot creation on the standby can be accelerated.
Note that logical decoding on a standby does not itself enforce that required
catalog rows are not removed. The user has to use physical replication slots +
hot_standby_feedback or other measures to prevent that. If catalog rows
required for a slot are removed, the slot is invalidated.
See 6af1793954e for an overall design of logical decoding on a standby.
Bumps catversion, for the addition of the pg_log_standby_snapshot() function.
Author: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Author: Andres Freund <andres@anarazel.de> (in an older version)
Author: Amit Khandekar <amitdkhan.pg@gmail.com> (in an older version)
Reviewed-by: Andres Freund <andres@anarazel.de>
Reviewed-by: FabrÌzio de Royes Mello <fabriziomello@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Reviewed-By: Robert Haas <robertmhaas@gmail.com>
Physical walsenders can't send data until it's been flushed; logical
walsenders can't decode and send data until it's been applied. On the
standby, the WAL is flushed first, which will only wake up physical
walsenders; and then applied, which will only wake up logical
walsenders.
Previously, all walsenders were awakened when the WAL was flushed. That
was fine for logical walsenders on the primary; but on the standby the
flushed WAL would have been not applied yet, so logical walsenders were
awakened too early.
Per idea from Jeff Davis and Amit Kapila.
Author: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Reviewed-By: Jeff Davis <pgsql@j-davis.com>
Reviewed-By: Robert Haas <robertmhaas@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Discussion: https://postgr.es/m/CAA4eK1+zO5LUeisabX10c81LU-fWMKO4M9Wyg1cdkbW7Hqh6vQ@mail.gmail.com
During WAL replay on the standby, when a conflict with a logical slot is
identified, invalidate such slots. There are two sources of conflicts:
1) Using the information added in 6af1793954e, logical slots are invalidated if
required rows are removed
2) wal_level on the primary server is reduced to below logical
Uses the infrastructure introduced in the prior commit. FIXME: add commit
reference.
Change InvalidatePossiblyObsoleteSlot() to use a recovery conflict to
interrupt use of a slot, if called in the startup process. The new recovery
conflict is added to pg_stat_database_conflicts, as confl_active_logicalslot.
See 6af1793954e for an overall design of logical decoding on a standby.
Bumps catversion for the addition of the pg_stat_database_conflicts column.
Bumps PGSTAT_FILE_FORMAT_ID for the same reason.
Author: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Author: Andres Freund <andres@anarazel.de>
Author: Amit Khandekar <amitdkhan.pg@gmail.com> (in an older version)
Reviewed-by: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Reviewed-by: Andres Freund <andres@anarazel.de>
Reviewed-by: Robert Haas <robertmhaas@gmail.com>
Reviewed-by: Fabrízio de Royes Mello <fabriziomello@gmail.com>
Reviewed-by: Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Reviewed-by: Alvaro Herrera <alvherre@alvh.no-ip.org>
Discussion: https://postgr.es/m/20230407075009.igg7be27ha2htkbt@awork3.anarazel.de
Needed for logical decoding on a standby. Slots need to be invalidated because
of the horizon if rows required for logical decoding are removed. If the
primary's wal_level is lowered from 'logical', logical slots on the standby
need to be invalidated.
The new invalidation methods will be used in a subsequent commit.
Logical slots that have been invalidated can be identified via the new
pg_replication_slots.conflicting column.
See 6af1793954e for an overall design of logical decoding on a standby.
Bumps catversion for the addition of the new pg_replication_slots column.
Author: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Author: Andres Freund <andres@anarazel.de>
Author: Amit Khandekar <amitdkhan.pg@gmail.com> (in an older version)
Reviewed-by: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Reviewed-by: Andres Freund <andres@anarazel.de>
Reviewed-by: Robert Haas <robertmhaas@gmail.com>
Reviewed-by: Fabrízio de Royes Mello <fabriziomello@gmail.com>
Reviewed-by: Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Reviewed-by: Melanie Plageman <melanieplageman@gmail.com>
Reviewed-by: Alvaro Herrera <alvherre@alvh.no-ip.org>
Discussion: https://postgr.es/m/20230407075009.igg7be27ha2htkbt@awork3.anarazel.de
Provide a way to ask the kernel to use O_DIRECT (or local equivalent)
where available for data and WAL files, to avoid or minimize kernel
caching. This hurts performance currently and is not intended for end
users yet. Later proposed work would introduce our own I/O clustering,
read-ahead, etc to replace the facilities the kernel disables with this
option.
The only user-visible change, if the developer-only GUC is not used, is
that this commit also removes the obscure logic that would activate
O_DIRECT for the WAL when wal_sync_method=open_[data]sync and
wal_level=minimal (which also requires max_wal_senders=0). Those are
non-default and unlikely settings, and this behavior wasn't (correctly)
documented. The same effect can be achieved with io_direct=wal.
Author: Thomas Munro <thomas.munro@gmail.com>
Author: Andres Freund <andres@anarazel.de>
Author: Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>
Reviewed-by: Justin Pryzby <pryzby@telsasoft.com>
Reviewed-by: Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>
Discussion: https://postgr.es/m/CA%2BhUKGK1X532hYqJ_MzFWt0n1zt8trz980D79WbjwnT-yYLZpg%40mail.gmail.com
In order to have the option to use O_DIRECT/FILE_FLAG_NO_BUFFERING in a
later commit, we need the addresses of user space buffers to be well
aligned. The exact requirements vary by OS and file system (typically
sectors and/or memory pages). The address alignment size is set to
4096, which is enough for currently known systems: it matches modern
sectors and common memory page size. There is no standard governing
O_DIRECT's requirements so we might eventually have to reconsider this
with more information from the field or future systems.
Aligning I/O buffers on memory pages is also known to improve regular
buffered I/O performance.
Three classes of I/O buffers for regular data pages are adjusted:
(1) Heap buffers are now allocated with the new palloc_aligned() or
MemoryContextAllocAligned() functions introduced by commit 439f6175.
(2) Stack buffers now use a new struct PGIOAlignedBlock to respect
PG_IO_ALIGN_SIZE, if possible with this compiler. (3) The buffer
pool is also aligned in shared memory.
WAL buffers were already aligned on XLOG_BLCKSZ. It's possible for
XLOG_BLCKSZ to be configured smaller than PG_IO_ALIGNED_SIZE and thus
for O_DIRECT WAL writes to fail to be well aligned, but that's a
pre-existing condition and will be addressed by a later commit.
BufFiles are not yet addressed (there's no current plan to use O_DIRECT
for those, but they could potentially get some incidental speedup even
in plain buffered I/O operations through better alignment).
If we can't align stack objects suitably using the compiler extensions
we know about, we disable the use of O_DIRECT by setting PG_O_DIRECT to
0. This avoids the need to consider systems that have O_DIRECT but
can't align stack objects the way we want; such systems could in theory
be supported with more work but we don't currently know of any such
machines, so it's easier to pretend there is no O_DIRECT support
instead. That's an existing and tested class of system.
Add assertions that all buffers passed into smgrread(), smgrwrite() and
smgrextend() are correctly aligned, unless PG_O_DIRECT is 0 (= stack
alignment tricks may be unavailable) or the block size has been set too
small to allow arrays of buffers to be all aligned.
Author: Thomas Munro <thomas.munro@gmail.com>
Author: Andres Freund <andres@anarazel.de>
Reviewed-by: Justin Pryzby <pryzby@telsasoft.com>
Discussion: https://postgr.es/m/CA+hUKGK1X532hYqJ_MzFWt0n1zt8trz980D79WbjwnT-yYLZpg@mail.gmail.com
Show a detailed description of the page offset number arrays that appear
in certain nbtree WAL records.
Also brings nbtree desc routines in line with the guidelines established
by recent commit 7d8219a4.
Author: Melanie Plageman <melanieplageman@gmail.com>
Reviewed-By: Peter Geoghegan <pg@bowt.ie>
Discussion: https://postgr.es/m/flat/20230109215842.fktuhesvayno6o4g%40awork3.anarazel.de
Add helper functions that output arrays in a standard format, and use
the functions inside heapdesc routines. This allows tools like
pg_walinspect to show a detailed description of the page offset number
arrays for records like PRUNE and VACUUM (unless there was an FPI).
Also document the conventions that desc routines should follow. Only
the heapdesc routines follow the conventions for now, so they're just
guidelines for the time being.
Based on a suggestion from Andres Freund.
Author: Melanie Plageman <melanieplageman@gmail.com>
Reviewed-By: Peter Geoghegan <pg@bowt.ie>
Discussion: https://postgr.es/m/flat/20230109215842.fktuhesvayno6o4g%40awork3.anarazel.de
This commit adds a limit to the size of an XLogRecord at 1020MB, based
on a suggestion by Heikki Linnakangas. This counts for the overhead
needed by the XLogReader when allocating the memory it needs to read a
record in DecodeXLogRecordRequiredSpace(), based on the record size. An
assertion based on that is added to detect that any additions in the
XLogReader facilities would not cause any overflows. If that's ever the
case, the upper bound allowed would need to be adjusted.
Before this, it was possible for an external module to create WAL
records large enough to be assembled but not replayable, causing
failures when replaying such WAL records on standbys. One case
mentioned where this is possible is the in-core function
pg_logical_emit_message() (wrapper for LogLogicalMessage), that allows
to emit WAL records with an arbitrary amount of data potentially higher
than the replay limit of approximately 1GB (limit of a palloc, minus the
overhead needed by a XLogReader).
This commit is a follow-up of ffd1b6b that has added similar protections
for the block-level data. Here, the checks are extended to the whole
record length, mainrdata_len being extended from uint32 to uint64 with
the routines registering buffer and record data still limited to uint32
to minimize the checks when assembling a record. All the error messages
related to overflow checks are improved to provide more context about
the error happening.
Author: Matthias van de Meent
Reviewed-by: Andres Freund, Heikki Linnakangas, Michael Paquier
Discussion: https://postgr.es/m/CAEze2WgGiw+LZt+vHf8tWqB_6VxeLsMeoAuod0N=ij1q17n5pw@mail.gmail.com
Instead of extending the relation block-by-block, use ExtendBufferedRelTo(),
introduced in 31966b151e6. This is faster and simpler.
This also somewhat reduces the danger that disconnected segments pose (which
can be "discovered" once the previous segment reaches SEGSIZE), as
ExtendBufferedRelTo() won't extend past the block it has been asked. However,
the risk of the content of such a disconnected segment being invalid
remains.
Discussion: https://postgr.es/m/20221029025420.eplyow6k7tgu6he3@awork3.anarazel.de
Discussion: https://postgr.es/m/20230223010147.32oir7sb66slqnjk@awork3.anarazel.de
While we already had some form of bulk extension for relations, it was fairly
limited. It only amortized the cost of acquiring the extension lock, the
relation itself was still extended one-by-one. Bulk extension was also solely
triggered by contention, not by the amount of data inserted.
To address this, use ExtendBufferedRelBy(), introduced in 31966b151e6, to
extend the relation. We try to extend the relation by multiple blocks in two
situations:
1) The caller tells RelationGetBufferForTuple() that it will need multiple
pages. For now that's only used by heap_multi_insert(), see commit FIXME.
2) If there is contention on the extension lock, use the number of waiters for
the lock as a multiplier for the number of blocks to extend by. This is
similar to what we already did. Previously we additionally multiplied the
numbers of waiters by 20, but with the new relation extension
infrastructure I could not see a benefit in doing so.
Using the freespacemap to provide empty pages can cause significant
contention, and adds measurable overhead, even if there is no contention. To
reduce that, remember the blocks the relation was extended by in the
BulkInsertState, in the extending backend. In case 1) from above, the blocks
the extending backend needs are not entered into the FSM, as we know that we
will need those blocks.
One complication with using the FSM to record empty pages, is that we need to
insert blocks into the FSM, when we already hold a buffer content lock. To
avoid doing IO while holding a content lock, release the content lock before
recording free space. Currently that opens a small window in which another
backend could fill the block, if a concurrent VACUUM records the free
space. If that happens, we retry, similar to the already existing case when
otherBuffer is provided. In the future it might be worth closing the race by
preventing VACUUM from recording the space in newly extended pages.
This change provides very significant wins (3x at 16 clients, on my
workstation) for concurrent COPY into a single relation. Even single threaded
COPY is measurably faster, primarily due to not dirtying pages while
extending, if supported by the operating system (see commit 4d330a61bb1). Even
single-row INSERTs benefit, although to a much smaller degree, as the relation
extension lock rarely is the primary bottleneck.
Reviewed-by: Melanie Plageman <melanieplageman@gmail.com>
Discussion: https://postgr.es/m/20221029025420.eplyow6k7tgu6he3@awork3.anarazel.de
A future commit will use this information to determine how aggressively to
extend the relation by. In heap_multi_insert() we know accurately how many
pages we need once we need to extend the relation, providing an accurate lower
bound for how much to extend.
Reviewed-by: Melanie Plageman <melanieplageman@gmail.com>
Discussion: https://postgr.es/m/20221029025420.eplyow6k7tgu6he3@awork3.anarazel.de
Allow autovacuum to reload the config file more often so that cost-based
delay parameters can take effect while VACUUMing a relation. Previously,
autovacuum workers only reloaded the config file once per relation
vacuumed, so config changes could not take effect until beginning to
vacuum the next table.
Now, check if a reload is pending roughly once per block, when checking
if we need to delay.
In order for autovacuum workers to safely update their own cost delay
and cost limit parameters without impacting performance, we had to
rethink when and how these values were accessed.
Previously, an autovacuum worker's wi_cost_limit was set only at the
beginning of vacuuming a table, after reloading the config file.
Therefore, at the time that autovac_balance_cost() was called, workers
vacuuming tables with no cost-related storage parameters could still
have different values for their wi_cost_limit_base and wi_cost_delay.
Now that the cost parameters can be updated while vacuuming a table,
workers will (within some margin of error) have no reason to have
different values for cost limit and cost delay (in the absence of
cost-related storage parameters). This removes the rationale for keeping
cost limit and cost delay in shared memory. Balancing the cost limit
requires only the number of active autovacuum workers vacuuming a table
with no cost-based storage parameters.
Author: Melanie Plageman <melanieplageman@gmail.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Reviewed-by: Daniel Gustafsson <daniel@yesql.se>
Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Reviewed-by: Robert Haas <robertmhaas@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/CAAKRu_ZngzqnEODc7LmS1NH04Kt6Y9huSjz5pp7%2BDXhrjDA0gw%40mail.gmail.com
While vacuuming a table in failsafe mode, VacuumCostActive should
not be re-enabled. This currently isn't a problem because vacuum
cost parameters are only refreshed in between vacuuming tables and
failsafe status is reset for every table.
In preparation for allowing vacuum cost parameters to be updated
more frequently, elevate LVRelState->failsafe_active to a global,
VacuumFailsafeActive, which will be checked when determining whether
or not to re-enable vacuum cost-related delays.
Author: Melanie Plageman <melanieplageman@gmail.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Reviewed-by: Daniel Gustafsson <daniel@yesql.se>
Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Reviewed-by: Robert Haas <robertmhaas@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/CAAKRu_ZngzqnEODc7LmS1NH04Kt6Y9huSjz5pp7%2BDXhrjDA0gw%40mail.gmail.com
Starting with commit 7db0cd2145f, RelationGetBufferForTuple() did a
visibilitymap_pin() while holding an exclusive buffer content lock on a newly
extended page, when using COPY FREEZE. We elsewhere try hard to avoid to doing
IO while holding a content lock. And until 14f98e0af99, that happened while
holding the relation extension lock.
Practically, this isn't a huge issue, because COPY FREEZE is restricted to
relations created or truncated in the current session, so it's unlikely
there's a lot of contention.
We can't avoid doing IO while holding the content lock by pinning the VM
earlier, because we don't know which page it will be on.
While we could just ignore the issue in this case, a future commit will add
bulk relation extension, which needs to enter pages into the FSM while also
trying to hold onto a buffer lock.
To address this issue, use visibilitymap_pin_ok() to see if the relevant
buffer is already pinned. If not, release the buffer, pin the VM buffer, and
acquire the lock again. This opens up a small window for other backends to
insert data onto the page - as the page is not entered into the freespacemap,
other backends won't see it normally, but a concurrent vacuum could enter the
page, if it started just after the relation is extended. In case the page is
used by another backend, retry. This is very similar to how locking
"otherBuffer" is already dealt with.
Reviewed-by: Tomas Vondra <tomas.vondra@enterprisedb.com>
Discussion: http://postgr.es/m/20230325025740.wzvchp2kromw4zqz@awork3.anarazel.de
GetVisibilityMapPins() insisted on the buffer1/buffer2 being in a specific
order. This required checks at the callsite. As a subsequent patch will add
another callsite, move related logic into GetVisibilityMapPins().
Discussion: https://postgr.es/m/20230403190030.fk2frxv6faklrseb@awork3.anarazel.de
A few places are not converted. Some because they are tackled in later
commits (e.g. hio.c, xlogutils.c), some because they are more
complicated (e.g. brin_pageops.c). Having a few users of ReadBuffer(P_NEW) is
good anyway, to ensure the backward compat path stays working.
Discussion: https://postgr.es/m/20221029025420.eplyow6k7tgu6he3@awork3.anarazel.de
A future patch will add support for extending relations by multiple blocks at
once. To be concurrency safe, the buffers for those blocks need to be marked
as BM_IO_IN_PROGRESS. Until now we only had infrastructure for recovering from
an IO error for a single buffer. This commit extends that infrastructure to
multiple buffers by using the resource owner infrastructure.
This commit increases the size of the ResourceOwnerData struct, which appears
to have a just about measurable overhead in very extreme workloads. Medium
term we are planning to substantially shrink the size of
ResourceOwnerData. Short term the increase is small enough to not worry about
it for now.
Reviewed-by: Melanie Plageman <melanieplageman@gmail.com>
Discussion: https://postgr.es/m/20221029025420.eplyow6k7tgu6he3@awork3.anarazel.de
Discussion: https://postgr.es/m/20221029200025.w7bvlgvamjfo6z44@awork3.anarazel.de
The read path needs to be able to initialize pages anyway, as relation
extensions are not durable. By avoiding initializing pages, we can, in a
future patch, extend the relation by multiple blocks at once.
Using smgrextend() for {vm,fsm}_extend() is not a good idea in general, as at
least one page of the VM/FSM will be read immediately after, always causing a
cache miss, requiring us to read content we just wrote.
Discussion: https://postgr.es/m/20230301223515.pucbj7nb54n4i4nv@awork3.anarazel.de
Commit 61b313e4 made VACUUM pass down a heaprel to index AM bulkdelete
and vacuumcleanup routines. Although this was primarily intended as
preparation for logical decoding on standbys, it also made it easy to
correct an old deficiency in how we determine how to cleanup SP-GiST
redirect and placeholder tuples.
Pass the heaprel to GlobalVisTestFor() during cleanup of redirect and
placeholder tuples, rather than pessimistically passing NULL.
Author: Bertrand Drouvot <bertranddrouvot.pg@gmail.com>
Discussion: https://postgr.es/m/02392033-f030-a3c8-c7d0-5c27eb529fec@gmail.com
Commit 61b313e4 made nbtree consistently pass down a heaprel to low
level routines like _bt_getbuf(). Although this was primarily intended
as preparation for logical decoding on standbys, it also made it easy to
correct an old deficiency in how nbtree VACUUM determines whether or not
it's now safe to recycle deleted pages.
Pass the heaprel to GlobalVisTestFor() in nbtree routines that deal with
recycle safety. nbtree now makes less pessimistic assumptions about
recycle safety within non-catalog relations. This enhancement
complements the recycling enhancement added by commit 9dd963ae25.
nbtree remains just as pessimistic as ever when it comes to recycle
safety within indexes on catalog relations. There is no fundamental
reason why we need to treat catalog relations differently, though. The
behavioral inconsistency is a consequence of the way that nbtree uses
nextXID values to implement what Lanin and Shasha call "the drain
technique". Note in particular that it has nothing to do with whether
or not index tuples might still be required for an older MVCC snapshot.
Author: Bertrand Drouvot <bertranddrouvot.pg@gmail.com>
Discussion: https://postgr.es/m/CAH2-WzkaiDxCje0yPuH=3Uh2p1V_2pFGY==xfbZoZu7Ax_NB8g@mail.gmail.com
Traditionally, vacuum always makes use of a buffer access strategy 32
buffers in size. This means that running vacuums tend not to cause too
many shared buffers to become dirty, however, this can cause vacuums to
run much more slowly than they otherwise could as WAL flushes will occur
more frequently due to having to flush WAL out to the LSN of the dirty
page before that page can be written to disk.
When we are performing failsafe VACUUMs (as added in 1e55e7d17), we really
want to make the vacuum work go as quickly as possible, so here we disable
the buffer access strategy when entering failsafe mode while vacuuming a
relation.
Per idea and analyis from Andres Freund.
In passing, also include some changes I had intended for 32fbe0239.
Author: Melanie Plageman
Reviewed-by: Justin Pryzby, David Rowley
Discussion: https://postgr.es/m/20230111182720.ejifsclfwymw2reb%40awork3.anarazel.de
This commit only implements one prerequisite part for allowing logical
decoding. The commit message contains an explanation of the overall design,
which later commits will refer back to.
Overall design:
1. We want to enable logical decoding on standbys, but replay of WAL
from the primary might remove data that is needed by logical decoding,
causing error(s) on the standby. To prevent those errors, a new replication
conflict scenario needs to be addressed (as much as hot standby does).
2. Our chosen strategy for dealing with this type of replication slot
is to invalidate logical slots for which needed data has been removed.
3. To do this we need the latestRemovedXid for each change, just as we
do for physical replication conflicts, but we also need to know
whether any particular change was to data that logical replication
might access. That way, during WAL replay, we know when there is a risk of
conflict and, if so, if there is a conflict.
4. We can't rely on the standby's relcache entries for this purpose in
any way, because the startup process can't access catalog contents.
5. Therefore every WAL record that potentially removes data from the
index or heap must carry a flag indicating whether or not it is one
that might be accessed during logical decoding.
Why do we need this for logical decoding on standby?
First, let's forget about logical decoding on standby and recall that
on a primary database, any catalog rows that may be needed by a logical
decoding replication slot are not removed.
This is done thanks to the catalog_xmin associated with the logical
replication slot.
But, with logical decoding on standby, in the following cases:
- hot_standby_feedback is off
- hot_standby_feedback is on but there is no a physical slot between
the primary and the standby. Then, hot_standby_feedback will work,
but only while the connection is alive (for example a node restart
would break it)
Then, the primary may delete system catalog rows that could be needed
by the logical decoding on the standby (as it does not know about the
catalog_xmin on the standby).
So, it’s mandatory to identify those rows and invalidate the slots
that may need them if any. Identifying those rows is the purpose of
this commit.
Implementation:
When a WAL replay on standby indicates that a catalog table tuple is
to be deleted by an xid that is greater than a logical slot's
catalog_xmin, then that means the slot's catalog_xmin conflicts with
the xid, and we need to handle the conflict. While subsequent commits
will do the actual conflict handling, this commit adds a new field
isCatalogRel in such WAL records (and a new bit set in the
xl_heap_visible flags field), that is true for catalog tables, so as to
arrange for conflict handling.
The affected WAL records are the ones that already contain the
snapshotConflictHorizon field, namely:
- gistxlogDelete
- gistxlogPageReuse
- xl_hash_vacuum_one_page
- xl_heap_prune
- xl_heap_freeze_page
- xl_heap_visible
- xl_btree_reuse_page
- xl_btree_delete
- spgxlogVacuumRedirect
Due to this new field being added, xl_hash_vacuum_one_page and
gistxlogDelete do now contain the offsets to be deleted as a
FLEXIBLE_ARRAY_MEMBER. This is needed to ensure correct alignment.
It's not needed on the others struct where isCatalogRel has
been added.
This commit just introduces the WAL format changes mentioned above. Handling
the actual conflicts will follow in future commits.
Bumps XLOG_PAGE_MAGIC as the several WAL records are changed.
Author: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Author: Andres Freund <andres@anarazel.de> (in an older version)
Author: Amit Khandekar <amitdkhan.pg@gmail.com> (in an older version)
Reviewed-by: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Reviewed-by: Andres Freund <andres@anarazel.de>
Reviewed-by: Robert Haas <robertmhaas@gmail.com>
Reviewed-by: Fabrízio de Royes Mello <fabriziomello@gmail.com>
Reviewed-by: Melanie Plageman <melanieplageman@gmail.com>
This is done in preparation for logical decoding on standby, which needs to
include whether visibility affecting WAL records are about a (user) catalog
table. Which is only known for the table, not the indexes.
It's also nice to be able to pass the heap relation to GlobalVisTestFor() in
vacuumRedirectAndPlaceholder().
Author: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Discussion: https://postgr.es/m/21b700c3-eecf-2e05-a699-f8c78dd31ec7@gmail.com
If visibilitymap_set() is called with flags containing a higher bit than
VISIBILITYMAP_ALL_FROZEN, the state of neighboring pages is affected. While
there was an assertion that *some* valid bits were set, it did not check
that *only* valid bits were. Change that.
Discussion: https://postgr.es/m/20230331043300.gux3s5wzrapqi4oe@awork3.anarazel.de
PageInit() while holding the extension lock is unnecessary after 0d1fe9f74e3
started to use RBM_ZERO_AND_LOCK - nobody can look at the new page before we
release the page lock. PageInit() zeroes the page, which isn't that cheap, so
deferring it until after the extension lock is released seems like a good idea.
Doing visibilitymap_pin() while holding the extension lock, introduced in
7db0cd2145f2, looks like an accident. Due to the restrictions on
HEAP_INSERT_FROZEN it's unlikely to be a performance issue, but it still seems
better to move it out. We also are doing the visibilitymap_pin() while
holding the buffer lock, which will be fixed in a separate commit.
Reviewed-by: Heikki Linnakangas <hlinnaka@iki.fi>
Discussion: http://postgr.es/m/419312fd-9255-078c-c3e3-f0525f911d7f@iki.fi
In instr_time.h it is stated that:
* When summing multiple measurements, it's recommended to leave the
* running sum in instr_time form (ie, use INSTR_TIME_ADD or
* INSTR_TIME_ACCUM_DIFF) and convert to a result format only at the end.
The reason for that is that converting to microseconds is not cheap, and can
loose precision. Therefore this commit changes 'PendingWalStats' to use
'instr_time' instead of 'PgStat_Counter' while accumulating 'wal_write_time'
and 'wal_sync_time'.
Author: Nazir Bilal Yavuz <byavuz81@gmail.com>
Reviewed-by: Andres Freund <andres@anarazel.de>
Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Reviewed-by: Melanie Plageman <melanieplageman@gmail.com>
Discussion: https://postgr.es/m/1feedb83-7aa9-cb4b-5086-598349d3f555@gmail.com
gistBuildCallback tried to fetch the size of an index tuple that
might have already been freed by gistProcessEmptyingQueue.
While this seems to usually be harmless in production builds,
in principle it could result in a SIGSEGV, or more likely a bogus
value for indtuplesSize leading to poor page-split decisions later
in the build.
The memory management here is confusing and could stand to be
refactored, but for the moment it seems to be enough to fetch
the tuple size sooner. AFAICT the indtuples[Size] totals aren't
used in between these places; even if they were, the updated
values shouldn't be any worse to use. So just move the
incrementing of the totals up.
It's not very clear why our valgrind-using buildfarm animals
haven't noticed this problem, because the relevant code path
does seem to be exercised according to the code coverage report.
I think the reason that we didn't fix this bug after the first
report is that I'd wanted to try to understand that better.
However, now that it's been re-discovered let's just be pragmatic
and fix it already.
Original report by Alexander Lakhin (bug #16329),
later rediscovered by Egor Chindyaskin (bug #17874).
Patch by Alexander Lakhin (commentary by Pavel Borisov and me).
Back-patch to all supported branches.
Discussion: https://postgr.es/m/16329-7a6aa9b6fa1118a1@postgresql.org
Discussion: https://postgr.es/m/17874-63ca6c7ce42d2103@postgresql.org
This function has been a no-op for over a decade. Even if bufmgr
regains a need to be called during commit, it seems unlikely that
the most appropriate call points would be precisely here, so it's not
doing us much good as a placeholder either. Now, removing it probably
doesn't save any noticeable number of cycles --- but the main call is
inside the commit critical section, and the less work done there the
better.
Matthias van de Meent
Discussion: https://postgr.es/m/CAEze2Wi1=tLKbxZnXzcD+8fYKyKqBtivVakLQC_mYBsP4Y8qVA@mail.gmail.com
In ancient times, these belonged to arguments or fields that were
actually of type long, but now they are not anymore, so this "L"
decoration is just confusing. (Some other 0L and other "L" constants
remain, where they are actually associated with a long type.)
Change the columns attndims, attstattarget, and attinhcount from int32
to int16, and reorder a bit. This saves some space (currently 4
bytes) in pg_attribute and tuple descriptors, which translates into
small performance benefits and/or room for new columns in pg_attribute
needed by future features.
attndims and attinhcount are never realistically used with values
larger than int16. Just to be sure, add some overflow checks.
attstattarget is currently limited explicitly to 10000.
For consistency, pg_constraint.coninhcount is also changed like
attinhcount.
Discussion: https://www.postgresql.org/message-id/flat/d07ffc2b-e0e8-77f7-38fb-be921dff71af%40enterprisedb.com
When extracting an attr from a cached tuple in the syscache with
SysCacheGetAttr the isnull parameter must be checked in case the
attr cannot be NULL. For cases when this is known beforehand, a
wrapper is introduced which perform the errorhandling internally
on behalf of the caller, invoking an elog in case of a NULL attr.
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Reviewed-by: Peter Eisentraut <peter.eisentraut@enterprisedb.com>
Reviewed-by: David Rowley <dgrowleyml@gmail.com>
Discussion: https://postgr.es/m/AD76405E-DB45-46B6-941F-17B1EB3A9076@yesql.se
Add pgstat counter to track row updates that result in the successor
version going to a new heap page, leaving behind an original version
whose t_ctid points to the new version. The current count is shown by
the n_tup_newpage_upd column of each of the pg_stat_*_tables views.
The new n_tup_newpage_upd column complements the existing n_tup_hot_upd
and n_tup_upd columns. Tables that have high n_tup_newpage_upd values
(relative to n_tup_upd) are good candidates for tuning heap fillfactor.
Corey Huinker, with small tweaks by me.
Author: Corey Huinker <corey.huinker@gmail.com>
Reviewed-By: Peter Geoghegan <pg@bowt.ie>
Reviewed-By: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/CADkLM=ded21M9iZ36hHm-vj2rE2d=zcKpUQMds__Xm2pxLfHKA@mail.gmail.com
Currently, in read committed transaction isolation mode (default), we have the
following sequence of actions when tuple_update()/tuple_delete() finds
the tuple updated by concurrent transaction.
1. Attempt to update/delete tuple with tuple_update()/tuple_delete(), which
returns TM_Updated.
2. Lock tuple with tuple_lock().
3. Re-evaluate plan qual (recheck if we still need to update/delete and
calculate the new tuple for update).
4. Second attempt to update/delete tuple with tuple_update()/tuple_delete().
This attempt should be successful, since the tuple was previously locked.
This patch eliminates step 2 by taking the lock during first
tuple_update()/tuple_delete() call. Heap table access method saves some
efforts by checking the updated tuple once instead of twice. Future
undo-based table access methods, which will start from the latest row version,
can immediately place a lock there.
The code in nodeModifyTable.c is simplified by removing the nested switch/case.
Discussion: https://postgr.es/m/CAPpHfdua-YFw3XTprfutzGp28xXLigFtzNbuFY8yPhqeq6X5kg%40mail.gmail.com
Reviewed-by: Aleksander Alekseev, Pavel Borisov, Vignesh C, Mason Sharp
Reviewed-by: Andres Freund, Chris Travers
When determining whether an index update may be skipped by using HOT, we
can ignore attributes indexed by block summarizing indexes without
references to individual tuples that need to be cleaned up.
A new type TU_UpdateIndexes provides a signal to the executor to
determine which indexes to update - no indexes, all indexes, or only the
summarizing indexes.
This also removes rd_indexattr list, and replaces it with rd_attrsvalid
flag. The list was not used anywhere, and a simple flag is sufficient.
This was originally committed as 5753d4ee32, but then got reverted by
e3fcca0d0d because of correctness issues.
Original patch by Josef Simanek, various fixes and improvements by Tomas
Vondra and me.
Authors: Matthias van de Meent, Josef Simanek, Tomas Vondra
Reviewed-by: Tomas Vondra, Alvaro Herrera
Discussion: https://postgr.es/m/05ebcb44-f383-86e3-4f31-0a97a55634cf@enterprisedb.com
Discussion: https://postgr.es/m/CAFp7QwpMRGcDAQumN7onN9HjrJ3u4X3ZRXdGFT0K5G2JWvnbWg%40mail.gmail.com
When calculating distance in brin_minmax_multi_distance_inet(), the
netmask was applied incorrectly. This results in (seemingly) incorrect
ordering of values, triggering an assert.
For builds without asserts this is mostly harmless - we may merge other
ranges, possibly resulting in slightly less efficient index. But it's
still correct and the greedy algorithm doesn't guarantee optimality
anyway.
Backpatch to 14, where minmax-multi indexes were introduced.
Reported by Dmitry Dolgov, investigation and fix by me.
Reported-by: Dmitry Dolgov
Backpatch-through: 14
Discussion: https://postgr.es/m/17774-c6f3e36dd4471e67@postgresql.org
The following changes are made to pg_write_zeros(), the API able to
write series of zeros using vectored I/O:
- Add of an "offset" parameter, to write the size from this position
(the 'p' of "pwrite" seems to mean position, though POSIX does not
outline ythat directly), hence the name of the routine is incorrect if
it is not able to handle offsets.
- Avoid memset() of "zbuffer" on every call.
- Avoid initialization of the whole IOV array if not needed.
- Group the trailing write() call with the main write() call,
simplifying the function logic.
Author: Andres Freund
Reviewed-by: Michael Paquier, Bharath Rupireddy
Discussion: https://postgr.es/m/20230215005525.mrrlmqrxzjzhaipl@awork3.anarazel.de
Open long-lived data and WAL file descriptors with O_CLOEXEC. This flag
was introduced by SUSv4 (POSIX.1-2008), and by now all of our target
Unix systems have it. Our open() implementation for Windows already had
that behavior, so provide a dummy O_CLOEXEC flag on that platform.
For now, callers of open() and the "thin" wrappers in fd.c that deal in
raw descriptors need to pass in O_CLOEXEC explicitly if desired. This
commit does that for WAL files, and automatically for everything
accessed via VFDs including SMgrRelation and BufFile. (With more
discussion we might decide to turn it on automatically for the thin
open()-wrappers too to avoid risk of missing places that need it, but
these are typically used for short-lived descriptors where we don't
expect to fork/exec, and it's remotely possible that extensions could be
using these APIs and passing descriptors to subprograms deliberately, so
that hasn't been done here.)
Do the same for sockets and the postmaster pipe with FD_CLOEXEC. (Later
commits might use modern interfaces to remove these extra fcntl() calls
and more where possible, but we'll need them as a fallback for a couple
of systems, so do it that way in this initial commit.)
With this change, subprograms executed for archiving, copying etc will
no longer have access to the server's descriptors, other than the ones
that we decide to pass down.
Reviewed-by: Andres Freund <andres@anarazel.de> (earlier version)
Discussion: https://postgr.es/m/CA%2BhUKGKb6FsAdQWcRL35KJsftv%2B9zXqQbzwkfRf1i0J2e57%2BhQ%40mail.gmail.com
This function has been semi-deprecated ever since we invented
bms_next_member(). Its habit of scribbling on the input bitmapset
isn't great, plus for sufficiently large bitmapsets it would take
O(N^2) time to complete a loop. Now we have the additional problem
that reducing the input to empty while leaving it still accessible
would violate a planned invariant. So let's just get rid of it,
after updating the few extant callers to use bms_next_member().
Patch by me; thanks to Nathan Bossart and Richard Guo for review.
Discussion: https://postgr.es/m/1159933.1677621588@sss.pgh.pa.us
