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>
When building hash indexes using the spool method, tuples are added to the
index page in hashkey order. Because of this, we can safely skip
performing the binary search on the existing tuples on the page to find
the location to insert the tuple based on its hashkey value. For this
case, we can just always put the tuple at the end of the item array as the
tuples will always arrive in hashkey order.
Testing has shown that this can improve hash index build speeds by 5-15%
with a unique set of integer values.
Author: Simon Riggs
Reviewed-by: David Rowley
Tested-by: David Zhang, Tomas Vondra
Discussion: https://postgr.es/m/CANbhV-GBc5JoG0AneUGPZZW3o4OK5LjBGeKe_icpC3R1McrZWQ@mail.gmail.com
Standardize on the name snapshotConflictHorizon for all XID fields from
WAL records that generate recovery conflicts when in hot standby mode.
This supersedes the previous latestRemovedXid naming convention.
The new naming convention places emphasis on how the values are actually
used by REDO routines. How the values are generated during original
execution (details of which vary by record type) is deemphasized. Users
of tools like pg_waldump can now grep for snapshotConflictHorizon to see
all potential sources of recovery conflicts in a standardized way,
without necessarily having to consider which specific record types might
be involved.
Also bring a couple of WAL record types that didn't follow any kind of
naming convention into line. These are heapam's VISIBLE record type and
SP-GiST's VACUUM_REDIRECT record type. Now every WAL record whose REDO
routine calls ResolveRecoveryConflictWithSnapshot() passes through the
snapshotConflictHorizon field from its WAL record. This is follow-up
work to the refactoring from commit 9e540599 that made FREEZE_PAGE WAL
records use a standard snapshotConflictHorizon style XID cutoff.
No bump in XLOG_PAGE_MAGIC, since the underlying format of affected WAL
records doesn't change.
Author: Peter Geoghegan <pg@bowt.ie>
Reviewed-By: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/CAH2-Wzm2CQUmViUq7Opgk=McVREHSOorYaAjR1ZpLYkRN7_dPw@mail.gmail.com
This makes the code more consistent with SpGiST, GiST and GIN, that
already use this style, and the idea is to make easier the introduction
of more sanity checks for each of these AM-specific macros. BRIN uses a
different set of macros to get a page's type and flags, so it has no
need for something similar.
Author: Matthias van de Meent
Discussion: https://postgr.es/m/CAEze2WjE3+tGO9Fs9+iZMU+z6mMZKo54W1Zt98WKqbEUHbHOBg@mail.gmail.com
xlog.h is directly and indirectly #included in a lot of places. With
this change, xloginsert.h is no longer unnecessarily included in the
large number of them that don't need it.
Author: Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>
Discussion: https://postgr.es/m/CALj2ACVe-W+WM5P44N7eG9C2_FmaeM8Dq5aCnD3fHt0Ba=WR6w@mail.gmail.com
All the code paths simplified here were already using a boolean or used
an expression that led to zero or one, making the extra bits
unnecessary.
Author: Justin Pryzby
Reviewed-by: Tom Lane, Michael Paquier, Peter Smith
Discussion: https://postgr.es/m/20210428182936.GE27406@telsasoft.com
The following changes make the predicate locking functions more
generic and suitable for use by future access methods:
- PredicateLockTuple() is renamed to PredicateLockTID(). It takes
ItemPointer and inserting transaction ID instead of HeapTuple.
- CheckForSerializableConflictIn() takes blocknum instead of buffer.
- CheckForSerializableConflictOut() no longer takes HeapTuple or buffer.
Author: Ashwin Agrawal
Reviewed-by: Andres Freund, Kuntal Ghosh, Thomas Munro
Discussion: https://postgr.es/m/CALfoeiv0k3hkEb3Oqk%3DziWqtyk2Jys1UOK5hwRBNeANT_yX%2Bng%40mail.gmail.com
Previously the xid horizon was only computed during WAL replay. That
had two major problems:
1) It relied on knowing what the table pointed to looks like. That was
easy enough before the introducing of tableam (we knew it had to be
heap, although some trickery around logging the heap relfilenodes
was required). But to properly handle table AMs we need
per-database catalog access to look up the AM handler, which
recovery doesn't allow.
2) Not knowing the xid horizon also makes it hard to support logical
decoding on standbys. When on a catalog table, we need to be able
to conflict with slots that have an xid horizon that's too old. But
computing the horizon by visiting the heap only works once
consistency is reached, but we always need to be able to detect
conflicts.
There's also a secondary problem, in that the current method performs
redundant work on every standby. But that's counterbalanced by
potentially computing the value when not necessary (either because
there's no standby, or because there's no connected backends).
Solve 1) and 2) by moving computation of the xid horizon to the
primary and by involving tableam in the computation of the horizon.
To address the potentially increased overhead, increase the efficiency
of the xid horizon computation for heap by sorting the tids, and
eliminating redundant buffer accesses. When prefetching is available,
additionally perform prefetching of buffers. As this is more of a
maintenance task, rather than something routinely done in every read
only query, we add an arbitrary 10 to the effective concurrency -
thereby using IO concurrency, when not globally enabled. That's
possibly not the perfect formula, but seems good enough for now.
Bumps WAL format, as latestRemovedXid is now part of the records, and
the heap's relfilenode isn't anymore.
Author: Andres Freund, Amit Khandekar, Robert Haas
Reviewed-By: Robert Haas
Discussion:
https://postgr.es/m/20181212204154.nsxf3gzqv3gesl32@alap3.anarazel.dehttps://postgr.es/m/20181214014235.dal5ogljs3bmlq44@alap3.anarazel.dehttps://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
Hash index searches acquire predicate locks on the primary
page of a bucket. It acquires a lock on both the old and new buckets
for scans that happen concurrently with page splits. During a bucket
split, a predicate lock is copied from the primary page of an old
bucket to the primary page of a new bucket.
Author: Shubham Barai, Amit Kapila
Reviewed by: Amit Kapila, Alexander Korotkov, Thomas Munro
Discussion: https://www.postgresql.org/message-id/flat/CALxAEPvNsM2GTiXdRgaaZ1Pjd1bs+sxfFsf7Ytr+iq+5JJoYXA@mail.gmail.com
Use IndexTupleSize everywhere, instead. Also, remove IndexTupleSize's
internal typecast, as that's not really needed and might mask coding
errors. Change some pointer variable datatypes in the call sites
to compensate for that and make it clearer what we're assuming.
Ildar Musin, Robert Haas, Stephen Frost
Discussion: https://postgr.es/m/0274288e-9e88-13b6-c61c-7b36928bf221@postgrespro.ru
Don't move parenthesized lines to the left, even if that means they
flow past the right margin.
By default, BSD indent lines up statement continuation lines that are
within parentheses so that they start just to the right of the preceding
left parenthesis. However, traditionally, if that resulted in the
continuation line extending to the right of the desired right margin,
then indent would push it left just far enough to not overrun the margin,
if it could do so without making the continuation line start to the left of
the current statement indent. That makes for a weird mix of indentations
unless one has been completely rigid about never violating the 80-column
limit.
This behavior has been pretty universally panned by Postgres developers.
Hence, disable it with indent's new -lpl switch, so that parenthesized
lines are always lined up with the preceding left paren.
This patch is much less interesting than the first round of indent
changes, but also bulkier, so I thought it best to separate the effects.
Discussion: https://postgr.es/m/E1dAmxK-0006EE-1r@gemulon.postgresql.org
Discussion: https://postgr.es/m/30527.1495162840@sss.pgh.pa.us
Most seriously, fix use of incorrect block ID, per a report from
Jeff Janes that it causes a crash and a diagnosis from Amit Kapila.
Improve consistency between the hash and btree versions of this
code by adding back a PANIC that btree has, and by registering
data in the xlog record in the same way, per complaints from
Jeff Janes and Amit Kapila.
Tidy up some minor cosmetic points, per complaints from Amit
Kapila.
Patch by Ashutosh Sharma, reviewed by Amit Kapila, and tested by
Jeff Janes.
Discussion: http://postgr.es/m/CAMkU=1w-9Qe=Ff1o6bSaXpNO9wqpo7_9GL8_CVhw4BoVVHasqg@mail.gmail.com
Clear LH_PAGE_HAS_DEAD_TUPLES during replay, similar to what gets done
for btree. Update hashdesc.c for xl_hash_vacuum_one_page.
Oversights in commit 6977b8b7f4 spotted
by Amit Kapila. Patch by Ashutosh Sharma.
Bump WAL version. The original patch to make hash indexes write-ahead
logged probably should have done this, and the single page vacuuming
patch probably should have done it again, but better late than never.
Discussion: http://postgr.es/m/CAA4eK1Kd=mJ9xreovcsh0qMiAj-QqCphHVQ_Lfau1DR9oVjASQ@mail.gmail.com
The warning about hash indexes not being write-ahead logged and their
use being discouraged has been removed. "snapshot too old" is now
supported for tables with hash indexes. Most importantly, barring
bugs, hash indexes will now be crash-safe and usable on standbys.
This commit doesn't yet add WAL consistency checking for hash
indexes, as we now have for other index types; a separate patch has
been submitted to cure that lack.
Amit Kapila, reviewed and slightly modified by me. The larger patch
series of which this is a part has been reviewed and tested by Álvaro
Herrera, Ashutosh Sharma, Mark Kirkwood, Jeff Janes, and Jesper
Pedersen.
Discussion: http://postgr.es/m/CAA4eK1JOBX=YU33631Qh-XivYXtPSALh514+jR8XeD7v+K3r_Q@mail.gmail.com
In preparation for adding write-ahead logging to hash indexes,
refactor _hash_freeovflpage and _hash_squeezebucket so that all
related page modifications happen in a single section of code. The
previous coding assumed that it would be fine to move tuples one at a
time, and also that the various operations involved in freeing an
overflow page didn't necessarily all need to be done together, all
of which is true if you don't care about write-ahead logging.
Amit Kapila, with slight changes by me.
This avoids a very significant amount of buffer manager traffic and
contention when scanning hash indexes, because it's no longer
necessary to lock and pin the metapage for every scan. We do need
some way of figuring out when the cache is too stale to use any more,
so that when we lock the primary bucket page to which the cached
metapage points us, we can tell whether a split has occurred since we
cached the metapage data. To do that, we use the hash_prevblkno field
in the primary bucket page, which would otherwise always be set to
InvalidBuffer.
This patch contains code so that it will continue working (although
less efficiently) with hash indexes built before this change, but
perhaps we should consider bumping the hash version and ripping out
the compatibility code. That decision can be made later, though.
Mithun Cy, reviewed by Jesper Pedersen, Amit Kapila, and by me.
Before committing, I made a number of cosmetic changes to the last
posted version of the patch, adjusted _hash_getcachedmetap to be more
careful about order of operation, and made some necessary updates to
the pageinspect documentation and regression tests.
This is basically for the same reasons I got rid of _hash_wrtbuf()
in commit 25216c98938495fd741bf585dcbef45b3a9ffd40: it's not
convenient to have a function which encapsulates MarkBufferDirty(),
especially as we move towards having hash indexes be WAL-logged.
Patch by me, reviewed (but not entirely endorsed) by Amit Kapila.
You can't just cast a HashMetaPage to a Page, because the meta page
data is stored after the page header, not at offset 0. Fortunately,
this didn't break anything because it happens to find hashm_bsize
at the offset at which it expects to find pd_pagesize_version, and
the values are close enough to the same that this works out.
Still, it's a bug, so back-patch to all supported versions.
Mithun Cy, revised a bit by me.
The whole concept of _hash_wrtbuf() is that we need to know at the
time we're releasing the buffer lock (and pin) whether we dirtied the
buffer, but this is easy to get wrong. This patch actually fixes one
non-obvious bug of that form: hashbucketcleanup forgot to signal
_hash_squeezebucket, which gets the primary bucket page already
locked, as to whether it had already dirtied the page. Calling
MarkBufferDirty() at the places where we dirty the buffer is more
intuitive and lets us simplify the code in various places as well.
On top of all that, the ultimate goal here is to make hash indexes
WAL-logged, and as the comments to _hash_wrtbuf() note, it should
go away when that happens. Making it go away a little earlier than
that seems like a good preparatory step.
Report by Jeff Janes. Diagnosis by Amit Kapila, Kuntal Ghosh,
and Dilip Kumar. Patch by me, after studying an alternative patch
submitted by Amit Kapila.
Discussion: http://postgr.es/m/CAA4eK1Kf6tOY0oVz_SEdngiNFkeXrA3xUSDPPORQvsWVPdKqnA@mail.gmail.com
Previously, the right to split a bucket was represented by a
heavyweight lock on the page number of the primary bucket page.
Unfortunately, this meant that every scan needed to take a heavyweight
lock on that bucket also, which was bad for concurrency. Instead, use
a cleanup lock on the primary bucket page to indicate the right to
begin a split, so that scans only need to retain a pin on that page,
which is they would have to acquire anyway, and which is also much
cheaper.
In addition to reducing the locking cost, this also avoids locking out
scans and inserts for the entire lifetime of the split: while the new
bucket is being populated with copies of the appropriate tuples from
the old bucket, scans and inserts can happen in parallel. There are
minor concurrency improvements for vacuum operations as well, though
the situation there is still far from ideal.
This patch also removes the unworldly assumption that a split will
never be interrupted. With the new code, a split is done in a series
of small steps and the system can pick up where it left off if it is
interrupted prior to completion. While this patch does not itself add
write-ahead logging for hash indexes, it is clearly a necessary first
step, since one of the things that could interrupt a split is the
removal of electrical power from the machine performing it.
Amit Kapila. I wrote the original design on which this patch is
based, and did a good bit of work on the comments and README through
multiple rounds of review, but all of the code is Amit's. Also
reviewed by Jesper Pedersen, Jeff Janes, and others.
Discussion: http://postgr.es/m/CAA4eK1LfzcZYxLoXS874Ad0+S-ZM60U9bwcyiUZx9mHZ-KCWhw@mail.gmail.com
The reverted changes were intended to force a choice of whether any
newly-added BufferGetPage() calls needed to be accompanied by a
test of the snapshot age, to support the "snapshot too old"
feature. Such an accompanying test is needed in about 7% of the
cases, where the page is being used as part of a scan rather than
positioning for other purposes (such as DML or vacuuming). The
additional effort required for back-patching, and the doubt whether
the intended benefit would really be there, have indicated it is
best just to rely on developers to do the right thing based on
comments and existing usage, as we do with many other conventions.
This change should have little or no effect on generated executable
code.
Motivated by the back-patching pain of Tom Lane and Robert Haas
This patch is a no-op patch which is intended to reduce the chances
of failures of omission once the functional part of the "snapshot
too old" patch goes in. It adds parameters for snapshot, relation,
and an enum to specify whether the snapshot age check needs to be
done for the page at this point. This initial patch passes NULL
for the first two new parameters and BGP_NO_SNAPSHOT_TEST for the
third. The follow-on patch will change the places where the test
needs to be made.
Since C99, it's been standard for printf and friends to accept a "z" size
modifier, meaning "whatever size size_t has". Up to now we've generally
dealt with printing size_t values by explicitly casting them to unsigned
long and using the "l" modifier; but this is really the wrong thing on
platforms where pointers are wider than longs (such as Win64). So let's
start using "z" instead. To ensure we can do that on all platforms, teach
src/port/snprintf.c to understand "z", and add a configure test to force
use of that implementation when the platform's version doesn't handle "z".
Having done that, modify a bunch of places that were using the
unsigned-long hack to use "z" instead. This patch doesn't pretend to have
gotten everyplace that could benefit, but it catches many of them. I made
an effort in particular to ensure that all uses of the same error message
text were updated together, so as not to increase the number of
translatable strings.
It's possible that this change will result in format-string warnings from
pre-C99 compilers. We might have to reconsider if there are any popular
compilers that will warn about this; but let's start by seeing what the
buildfarm thinks.
Andres Freund, with a little additional work by me
Avoid using LockPage(rel, 0, lockmode) to protect against changes to
the bucket mapping. Instead, an exclusive buffer content lock is now
viewed as sufficient permission to modify the metapage, and a shared
buffer content lock is used when such modifications need to be
prevented. This more relaxed locking regimen makes it possible that,
when we're busy getting a heavyweight bucket on the bucket we intend
to search or insert into, a bucket split might occur underneath us.
To compenate for that possibility, we use a loop-and-retry system:
release the metapage content lock, acquire the heavyweight lock on the
target bucket, and then reacquire the metapage content lock and check
that the bucket mapping has not changed. Normally it hasn't, and
we're done. But if by chance it has, we simply unlock the metapage,
release the heavyweight lock we acquired previously, lock the new
bucket, and loop around again. Even in the worst case we cannot loop
very many times here, since we don't split the same bucket again until
we've split all the other buckets, and 2^N gets big pretty fast.
This results in greatly improved concurrency, because we're
effectively replacing two lwlock acquire-and-release cycles in
exclusive mode (on one of the lock manager locks) with a single
acquire-and-release cycle in shared mode (on the metapage buffer
content lock). Testing shows that it's still not quite as good as
btree; for that, we'd probably have to find some way of getting rid
of the heavyweight bucket locks as well, which does not appear
straightforward.
Patch by me, review by Jeff Janes.
hash indexes keep entries sorted by hash value. First, the original plans for
concurrency assumed that insertions would happen only at the end of a page,
which is no longer true; this could cause scans to transiently fail to find
index entries in the presence of concurrent insertions. We can compensate
by teaching scans to re-find their position after re-acquiring read locks.
Second, neither the bucket split nor the bucket compaction logic had been
fixed to preserve hashvalue ordering, so application of either of those
processes could lead to permanent corruption of an index, in the sense
that searches might fail to find entries that are present.
This patch fixes the split and compaction logic to preserve hashvalue
ordering, but it cannot do anything about pre-existing corruption. We will
need to recommend reindexing all hash indexes in the 8.4.2 release notes.
To buy back the performance loss hereby induced in split and compaction,
fix them to use PageIndexMultiDelete instead of retail PageIndexDelete
operations. We might later want to do something with qsort'ing the
page contents rather than doing a binary search for each insertion,
but that seemed more invasive than I cared to risk in a back-patch.
Per bug #5157 from Jeff Janes and subsequent investigation.
Andres Freund
Bruce Momjian
David Rowley
Peter Geoghegan
Michael Paquier
Alvaro Herrera
Thomas Munro
Amit Kapila
Tom Lane
Teodor Sigaev
Robert Haas
Kevin Grittner
Magnus Hagander