aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
...
* Collect statistics about SLRU cachesTomas Vondra2020-04-02
| | | | | | | | | | | | | | | | | | | | | | | | | | | | There's a number of SLRU caches used to access important data like clog, commit timestamps, multixact, asynchronous notifications, etc. Until now we had no easy way to monitor these shared caches, compute hit ratios, number of reads/writes etc. This commit extends the statistics collector to track this information for a predefined list of SLRUs, and also introduces a new system view pg_stat_slru displaying the data. The list of built-in SLRUs is fixed, but additional SLRUs may be defined in extensions. Unfortunately, there's no suitable registry of SLRUs, so this patch simply defines a fixed list of SLRUs with entries for the built-in ones and one entry for all additional SLRUs. Extensions adding their own SLRU are fairly rare, so this seems acceptable. This patch only allows monitoring of SLRUs, not tuning. The SLRU sizes are still fixed (hard-coded in the code) and it's not entirely clear which of the SLRUs might need a GUC to tune size. In a way, allowing us to determine that is one of the goals of this patch. Bump catversion as the patch introduces new functions and system view. Author: Tomas Vondra Reviewed-by: Alvaro Herrera Discussion: https://www.postgresql.org/message-id/flat/20200119143707.gyinppnigokesjok@development
* Clean up parsing of ltree and lquery some more.Tom Lane2020-04-01
| | | | | | | | | | | | | | | | Fix lquery parsing to handle repeated flag characters correctly, and to enforce the max label length correctly in some cases where it did not before, and to detect empty labels in some cases where it did not before. In a more cosmetic vein, use a switch rather than if-then chains to handle the different states, and avoid unnecessary checks on charlen when looking for ASCII characters, and factor out multiple copies of the label length checking code. Tom Lane and Dmitry Belyavsky Discussion: https://postgr.es/m/CADqLbzLVkBuPX0812o+z=c3i6honszsZZ6VQOSKR3VPbB56P3w@mail.gmail.com
* Add support for binary I/O of ltree, lquery, and ltxtquery types.Tom Lane2020-04-01
| | | | | | | | | | | | | | | | | | Not much to say here --- does what it says on the tin. The "binary" representation in each case is really just the same as the text format, though we prefix a version-number byte in case anyone ever feels motivated to change that. Thus, there's not any expectation of improved speed or reduced space; the point here is just to allow clients to use binary format for all columns of a query result or COPY data. This makes use of the recently added ALTER TYPE support to add binary I/O functions to an existing data type. As in commit a80818605, we can piggy-back on there already being a new-for-v13 version of the ltree extension, so we don't need a new update script file. Nino Floris, reviewed by Alexander Korotkov and myself Discussion: https://postgr.es/m/CANmj9Vxx50jOo1L7iSRxd142NyTz6Bdcgg7u9P3Z8o0=HGkYyQ@mail.gmail.com
* Check equality semantics for unique indexes on partitioned tables.Tom Lane2020-04-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We require the partition key to be a subset of the set of columns being made unique, so that physically-separate indexes on the different partitions are sufficient to enforce the uniqueness constraint. The existing code checked that the listed columns appear, but did not inquire into the index semantics, which is a serious oversight given that different index opclasses might enforce completely different notions of uniqueness. Ideally, perhaps, we'd just match the partition key opfamily to the index opfamily. But hash partitioning uses hash opfamilies which we can't directly match to btree opfamilies. Hence, look up the equality operator in each family, and accept if it's the same operator. This should be okay in a fairly general sense, since the equality operator ought to precisely represent the opfamily's notion of uniqueness. A remaining weak spot is that we don't have a cross-index-AM notion of which opfamily member is "equality". But we know which one to use for hash and btree AMs, and those are the only two that are relevant here at present. (Any non-core AMs that know how to enforce equality are out of luck, for now.) Back-patch to v11 where this feature was introduced. Guancheng Luo, revised a bit by me Discussion: https://postgr.es/m/D9C3CEF7-04E8-47A1-8300-CA1DCD5ED40D@gmail.com
* Improve selectivity estimation for assorted match-style operators.Tom Lane2020-04-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Quite a few matching operators such as JSONB's @> used "contsel" and "contjoinsel" as their selectivity estimators. That was a bad idea, because (a) contsel is only a stub, yielding a fixed default estimate, and (b) that default is 0.001, meaning we estimate these operators as five times more selective than equality, which is surely pretty silly. There's a good model for improving this in ltree's ltreeparentsel(): for any "var OP constant" query, we can try applying the operator to all of the column's MCV and histogram values, taking the latter as being a random sample of the non-MCV values. That code is actually 100% generic, except for the question of exactly what default selectivity ought to be plugged in when we don't have stats. Hence, migrate the guts of ltreeparentsel() into the core code, provide wrappers "matchingsel" and "matchingjoinsel" with a more-appropriate default estimate, and use those for the non-geometric operators that formerly used contsel (mostly JSONB containment operators and tsquery matching). Also apply this code to some match-like operators in hstore, ltree, and pg_trgm, including the former users of ltreeparentsel as well as ones that improperly used contsel. Since commit 911e70207 just created new versions of those extensions that we haven't released yet, we can sneak this change into those new versions instead of having to create an additional generation of update scripts. Patch by me, reviewed by Alexey Bashtanov Discussion: https://postgr.es/m/12237.1582833074@sss.pgh.pa.us
* Refactor code to look up local replication tuplePeter Eisentraut2020-04-01
| | | | | | | This unifies some duplicate code. Author: Amit Langote <amitlangote09@gmail.com> Discussion: https://www.postgresql.org/message-id/CA+HiwqFjYE5anArxvkjr37AQMd52L-LZtz9Ld2QrLQ3YfcYhTw@mail.gmail.com
* Update SQL features countPeter Eisentraut2020-04-01
| | | | | | | | | | | The previously listed total of 179 does not appear to be correct for SQL:2016 anymore. (Previous SQL versions had slightly different feature sets, so it's plausible that it was once correct.) The currently correct count is the number of rows in the respective tables in appendix F in SQL parts 2 and 11, minus 2 features that are listed twice. Thus the correct count is currently 177. This also matches the number of Core entries the built documentation currently shows, so it's internally consistent.
* Fix typo in contrib/intarray documentationAlexander Korotkov2020-04-01
| | | | | Reported-by: Erik Rijkers Discussion: https://postgr.es/m/82529ecf9bcc58d5b5cf9f3ffb699881%40xs4all.nl
* Correct CREATE INDEX documentation for opclass parametersAlexander Korotkov2020-04-01
| | | | | | Old versions of opclass parameters patch supported ability to specify DEFAULT as the opclass name in CREATE INDEX command. This ability was removed in the final version, but 911e702077 still mentions that in the documentation.
* Documentation corrections for opclass parametersAlexander Korotkov2020-04-01
| | | | | Discussion: https://postgr.es/m/20200331024419.GB14618%40telsasoft.com Author: Justin Pryzby
* Fix crash in psql when attempting to reuse old connectionMichael Paquier2020-04-01
| | | | | | | | | | | | | | | | | In a psql session, if the connection to the server is abruptly cut, the referenced connection would become NULL as of CheckConnection(). This could cause a hard crash with psql if attempting to connect by reusing the past connection's data because of a null-pointer dereference with either PQhost() or PQdb(). This issue is fixed by making sure that no reuse of the past connection is done if it does not exist. Issue has been introduced by 6e5f8d4, so backpatch down to 12. Reported-by: Hugh Wang Author: Michael Paquier Reviewed-by: Álvaro Herrera, Tom Lane Discussion: https://postgr.es/m/16330-b34835d83619e25d@postgresql.org Backpatch-through: 12
* Fix coverity complaint about commit 40d964ec99.Amit Kapila2020-04-01
| | | | | | | | | | | The coverity complained that dividing integer expressions and then converting the integer quotient to type "double" would lose fractional part. Typecasting one of the arguments of expression with double should fix the report. Author: Mahendra Singh Thalor Reviewed-by: Amit Kapila Discussion: https://postgr.es/m/20200329224818.6phnhv7o2q2rfovf@alap3.anarazel.de
* psql: do file completion for \gxBruce Momjian2020-03-31
| | | | | | | | | | This was missed when the feature was added. Reported-by: Vik Fearing Discussion: https://postgr.es/m/eca20529-0b06-b493-ee38-f071a75dcd5b@postgresfriends.org Backpatch-through: 10
* Add -c/--restore-target-wal to pg_rewindMichael Paquier2020-04-01
| | | | | | | | | | | | | | | | | | | | | | | | | | pg_rewind needs to copy from the source cluster to the target cluster a set of relation blocks changed from the previous checkpoint where WAL forked up to the end of WAL on the target. Building this list of relation blocks requires a range of WAL segments that may not be present anymore on the target's pg_wal, causing pg_rewind to fail. It is possible to work around this issue by copying manually the WAL segments needed but this may lead to some extra and actually useless work. This commit introduces a new option allowing pg_rewind to use a restore_command while doing the rewind by grabbing the parameter value of restore_command from the target cluster configuration. This allows the rewind operation to be more reliable, so as only the WAL segments needed by the rewind are restored from the archives. In order to be able to do that, a new routine is added to src/common/ to allow frontend tools to restore files from archives using an already-built restore command. This version is more simple than the backend equivalent as there is no need to handle the non-recovery case. Author: Alexey Kondratov Reviewed-by: Andrey Borodin, Andres Freund, Alvaro Herrera, Alexander Korotkov, Michael Paquier Discussion: https://postgr.es/m/a3acff50-5a0d-9a2c-b3b2-ee36168955c1@postgrespro.ru
* doc: remove mention of bitwise operators as solely type-limitedBruce Momjian2020-03-31
| | | | | | | | | | | There are other operators that have limited number data type support, so just remove the sentence. Reported-by: Sergei Agalakov Discussion: https://postgr.es/m/158032651854.19851.16261832706661813796@wrigleys.postgresql.org Backpatch-through: 9.5
* doc: clarify hierarchy of objects: global, db, schema, etc.Bruce Momjian2020-03-31
| | | | | | | | | | | The previous wording was confusing because it wasn't in decreasing order and had to backtrack. Also clarify role/user wording. Reported-by: jbird@nuna.com Discussion: https://postgr.es/m/158057750885.1123.2806779262588618988@wrigleys.postgresql.org Backpatch-through: 9.5
* doc: clarify when row-level locks are releasedBruce Momjian2020-03-31
| | | | | | | | | | | They are released just like table-level locks. Also clean up wording. (Uses wording "rolled back to".) Reported-by: me@sillymon.ch Discussion: https://postgr.es/m/158074944048.1095.4309647363871637715@wrigleys.postgresql.org Backpatch-through: 9.5
* Add CREATE INDEX deduplication assertions.Peter Geoghegan2020-03-31
| | | | | Add two assertions that verify the assumptions about posting list tuple space accounting and suffix truncation made within nbtsort.c.
* Revert erroroneous commit 834b80464d; my apologiesBruce Momjian2020-03-31
| | | | Backpatch-through: master
* dummy commitBruce Momjian2020-03-31
|
* doc: add namespace column to pg_buffercache example queryBruce Momjian2020-03-31
| | | | | | | | | | Without the namespace, the table name could be ambiguous. Reported-by: adunham@arbormetrix.com Discussion: https://postgr.es/m/158155175140.23798.2189464781144503491@wrigleys.postgresql.org Backpatch-through: 9.5
* doc: clarify which table creation is used for inheritance part.Bruce Momjian2020-03-31
| | | | | | | | | | | | Previously people might assume that the partition syntax version of CREATE TABLE is to be used for the inheritance partition table example; mention that the non-partitioned version should be used. Reported-by: mib@nic.at Discussion: https://postgr.es/m/158089540905.1098.15071165437284409576@wrigleys.postgresql.org Backpatch-through: 10
* Fix race condition in statext_store().Tom Lane2020-03-31
| | | | | | | | | | | | | Must hold some lock on the pg_statistic_ext_data catalog *before* we look up the tuple we aim to replace. Otherwise a concurrent VACUUM FULL or similar operation could move it to a different TID, leaving us trying to replace the wrong tuple. Back-patch to v12 where this got broken. Credit goes to Dean Rasheed; I'm just doing the clerical work. Discussion: https://postgr.es/m/CAEZATCU0zHMDiQV0g8P2U+YSP9C1idUPrn79DajsbonwkN0xvQ@mail.gmail.com
* doc: adjust UPDATE/DELETE's FROM/USING to match SELECT's FROMBruce Momjian2020-03-31
| | | | | | | | | | Previously the syntax and wording were unclear. Reported-by: Alexey Bashtanov Discussion: https://postgr.es/m/968d4724-8e58-788f-7c45-f7b1813824cc@imap.cc Backpatch-through: 9.5
* Still another try at stabilizing stats_ext test results.Tom Lane2020-03-31
| | | | | | | | | | | | | | | | | The stats_ext test is not expecting that autovacuum will touch any of its tables; an expectation falsified by commit b07642dbc. Although I'm suspicious that there's something else going on that makes extended stats estimates not 100% reproducible, it's pretty easy to demonstrate that there are places in this test that fail if an autovacuum updates the table's stats unexpectedly. Hence, revert the band-aid changes made by 2dc16efed and 24566b359 in favor of summarily disabling autovacuum for all the tables that this test checks estimated rowcounts for. Also remove an evidently obsolete comment at the head of the test. Discussion: https://postgr.es/m/15012.1585623298@sss.pgh.pa.us
* Remove header noise from test_decoding testAlvaro Herrera2020-03-31
| | | | | | | | Use psql's expanded output to avoid a pointless header. Kyotaro Horiguchi, after an idea of Michael Paquier Discussion: https://postgr.es/m/20181120050744.GJ4400@paquier.xyz
* Improve the message logged when recovery is paused.Fujii Masao2020-04-01
| | | | | | | | | | | | | | | | | When recovery target is reached and recovery is paused because of recovery_target_action=pause, executing pg_wal_replay_resume() causes the standby to promote, i.e., the recovery to end. So, in this case, the previous message "Execute pg_wal_replay_resume() to continue" logged was confusing because pg_wal_replay_resume() doesn't cause the recovery to continue. This commit improves the message logged when recovery is paused, and the proper message is output based on what (pg_wal_replay_pause or recovery_target_action) causes recovery to be paused. Author: Sergei Kornilov, revised by Fujii Masao Reviewed-by: Robert Haas Discussion: https://postgr.es/m/19168211580382043@myt5-b646bde4b8f3.qloud-c.yandex.net
* Allow ecpg to be built stand-alone, allow parallel libpq makeBruce Momjian2020-03-31
| | | | | | | | | | | | | This change defines SHLIB_PREREQS for the libpgport dependency, rather than using a makefile rule. This was broken in PG 12. Reported-by: Filip Janus Discussion: https://postgr.es/m/E5Dc85EGUY4wyG8cjAU0qoEdCJxGK_qhW1s9qSuYq9A@mail.gmail.com Author: Dagfinn Ilmari Mannsåker (for libpq) Backpatch-through: 12
* Teach pg_ls_dir_files() to ignore ENOENT failures from stat().Tom Lane2020-03-31
| | | | | | | | | | | | | | | | Buildfarm experience shows that this function can fail with ENOENT if some other process unlinks a file between when we read the directory entry and when we try to stat() it. The problem is old but we had not noticed it until 085b6b667 added regression test coverage. To fix, just ignore ENOENT failures. There is one other case that this might hide: a symlink that points to nowhere. That seems okay though, at least better than erroring. Back-patch to v10 where this function was added, since the regression test cases were too. Discussion: https://postgr.es/m/20200308173103.GC1357@telsasoft.com
* Fix lquery's NOT handling, and add ability to quantify non-'*' items.Tom Lane2020-03-31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The existing implementation of the ltree ~ lquery match operator is sufficiently complex and undocumented that it's hard to tell exactly what it does. But one thing it clearly gets wrong is the combination of NOT symbols (!) and '*' symbols. A pattern such as '*.!foo.*' should, by any ordinary understanding of regular expression behavior, match any ltree that has at least one label that's not "foo". As best we can tell by experimentation, what it's actually matching is any ltree in which *no* label is "foo". That's surprising, and not at all what the documentation says. Now, that's arguably a useful behavior, so if we rewrite to fix the bug we should provide some other way to get it. To do so, add the ability to attach lquery quantifiers to non-'*' items as well as '*'s. Then the pattern '!foo{,}' expresses "any ltree in which no label is foo". For backwards compatibility, the default quantifier for non-'*' items has to be "{1}", although the default for '*' items is '{,}'. I wouldn't have done it like that in a green field, but it's not totally horrible. Armed with that, rewrite checkCond() from scratch. Treating '*' and non-'*' items alike makes it simpler, not more complicated, so that the function actually gets a lot shorter than it was. Filip Rembiałkowski, Tom Lane, Nikita Glukhov, per a very ancient bug report from M. Palm Discussion: https://postgr.es/m/CAP_rww=waX2Oo6q+MbMSiZ9ktdj6eaJj0cQzNu=Ry2cCDij5fw@mail.gmail.com
* Improve error messages in ltree_in and lquery_in.Tom Lane2020-03-31
| | | | | | | | | | | | | | | Ensure that the type name is mentioned in all cases (bare "syntax error" isn't that helpful). Avoid using the term "level", since that's not used in the documentation. Phrase error position reports as "at character N" not "in position N"; the latter seems ambiguous, and it's certainly not how we say it elsewhere. For the same reason, make the character position values be 1-based not 0-based. Provide a position in more cases. (I continued to leave that out of messages complaining about end-of-input, where it seemed pointless, as well as messages complaining about overall input complexity, where fingering any one part of the input would be arbitrary.) Discussion: https://postgr.es/m/15582.1585529626@sss.pgh.pa.us
* Improve error reporting in opclasscmds.cAlexander Korotkov2020-03-31
| | | | | | This commit improves error reporting introduced by 911e702077. It puts argument of errmsg() to the single line for easier grepping source for error text. Also it improves wording of errhint().
* Fix assorted typosMagnus Hagander2020-03-31
| | | | Author: Daniel Gustafsson <daniel@yesql.se>
* Fix INSERT OVERRIDING USER VALUE behaviorPeter Eisentraut2020-03-31
| | | | | | | | | | | | | The original implementation disallowed using OVERRIDING USER VALUE on identity columns defined as GENERATED ALWAYS, which is not per standard. So allow that now. Expand documentation and tests around this. Author: Dean Rasheed <dean.a.rasheed@gmail.com> Reviewed-by: Peter Eisentraut <peter.eisentraut@2ndquadrant.com> Reviewed-by: Vik Fearing <vik@postgresfriends.org> Discussion: https://www.postgresql.org/message-id/flat/CAEZATCVrh2ufCwmzzM%3Dk_OfuLhTTPBJCdFkimst2kry4oHepuQ%40mail.gmail.com
* Move routine definitions of xlogarchive.c to a new header fileMichael Paquier2020-03-31
| | | | | | | | | | | | | | The definitions of the routines defined in xlogarchive.c have been part of xlog_internal.h which is included by several frontend tools, but all those routines are only called by the backend. More cleanup could be done within xlog_internal.h, but that's already a nice cut. This will help a follow-up patch for pg_rewind where handling of restore_command is added for frontends. Author: Alexey Kondratov, Michael Paquier Reviewed-by: Álvaro Herrera, Alexander Korotkov Discussion: https://postgr.es/m/a3acff50-5a0d-9a2c-b3b2-ee36168955c1@postgrespro.ru
* Update SQL featuresPeter Eisentraut2020-03-31
| | | | Set T653 to supported. This has always been possible.
* Avoid calls to RelationGetRelationName() and RelationGetNamespace() inAmit Kapila2020-03-31
| | | | | | | | | | | | | | | | vacuum code. After commit b61d161c14, during vacuum, we cache the information of relation name and relation namespace in local structure LVRelStats so that we can use it in an error callback function. We can use the cached information to avoid the calls to RelationGetRelationName(), RelationGetNamespace() and get_namespace_name(). This is mainly for the consistent in vacuum code path but it will avoid the extra syscache lookup we do in get_namespace_name(). Author: Justin Pryzby Reviewed-by: Amit Kapila Discussion: https://www.postgresql.org/message-id/20191120210600.GC30362@telsasoft.com
* Further simplify nbtree high key truncation.Peter Geoghegan2020-03-30
| | | | | | | | | Commit 7c2dbc69 reorganized _bt_truncate() in a way that enables a further simplification that I (pgeoghegan) missed: Since we mark the tuple that is returned to the caller as a pivot tuple before the point where its heap TID is set as of 7c2dbc69, it is possible to use the high level BTreeTupleGetHeapTID() inline function to get an item pointer. Do it that way now. This approach is clearer and more maintainable.
* Revert "Skip redundant anti-wraparound vacuums"Michael Paquier2020-03-31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 2aa6e33, that added a fast path to skip anti-wraparound and non-aggressive autovacuum jobs (these have no sense as anti-wraparound implies aggressive). With a cluster using a high amount of relations with a portion of them being heavily updated, this could cause autovacuum to lock down, with autovacuum workers attempting repeatedly those jobs on the same relations for the same database, that just kept being skipped. This lock down can be solved with a manual VACUUM FREEZE. Justin King has reported one environment where the issue happened, and Julien Rouhaud and I have been able to reproduce it in a second environment. With a very aggressive autovacuum_freeze_max_age, triggering those jobs with pgbench is a matter of minutes, and hitting the lock down is a lot harder (my local tests failed to do that). Note that anti-wraparound and non-aggressive jobs can only be triggered on a subset of shared catalogs: - pg_auth_members - pg_authid - pg_database - pg_replication_origin - pg_shseclabel - pg_subscription - pg_tablespace While the lock down was possible down to v12, the root cause of those jobs is a much older issue, which needs more analysis. Bonus thanks to Andres Freund for the discussion. Reported-by: Justin King Discussion: https://postgr.es/m/CAE39h22zPLrkH17GrkDgAYL3kbjvySYD1io+rtnAUFnaJJVS4g@mail.gmail.com Backpatch-through: 12
* Refactor nbtree high key truncation.Peter Geoghegan2020-03-30
| | | | | | | | | | | | | Simplify _bt_truncate(), the routine that generates truncated leaf page high keys. Remove a micro-optimization that avoided a second palloc0() call (this was used when a heap TID was needed in the final pivot tuple, though only when the index happened to not be an INCLUDE index). Removing this dubious micro-optimization allows _bt_truncate() to use the index_truncate_tuple() indextuple.c utility routine in all cases. This was already the common case. This commit is a HEAD-only follow up to bugfix commit 4b42a899.
* Deduplicate PageIsNew() check in lazy_scan_heap().Andres Freund2020-03-30
| | | | | | | | | | | The recheck isn't needed anymore, as RelationGetBufferForTuple() now extends the relation with RBM_ZERO_AND_LOCK. Previously we needed to handle the fact that relation extension extended the relation and then separately acquired a lock on the page - while expecting that the page is empty. Reported-By: Ranier Vilela Discussion: https://postgr.es/m/CAEudQArA_=J0D5T258xsCY6Xtf6wiH4b=QDPDgVS+WZUN10WDw@mail.gmail.com
* Fix missing SP-GiST support in 911e702077Alexander Korotkov2020-03-30
| | | | | 911e702077 misses setting of amoptsprocnum for SP-GiST. This commit fixes that.
* Remove rudiments of supporting procnum == 0 from 911e702077Alexander Korotkov2020-03-30
| | | | | | Early versions of opclass options patch uses zero support procedure as opclass options procedure. This commit removes rudiments of it, which were committed in 911e702077. Also, it implements correct handling of amoptsprocnum == 0.
* Consistently truncate non-key suffix columns.Peter Geoghegan2020-03-30
| | | | | | | | | | | | | | | INCLUDE indexes failed to have their non-key attributes physically truncated away in certain rare cases. This led to physically larger pivot tuples that contained useless non-key attribute values. The impact on users should be negligible, but this is still clearly a regression (Postgres 11 supports INCLUDE indexes, and yet was not affected). The bug appeared in commit dd299df8, which introduced "true" suffix truncation of key attributes. Discussion: https://postgr.es/m/CAH2-Wz=E8pkV9ivRSFHtv812H5ckf8s1-yhx61_WrJbKccGcrQ@mail.gmail.com Backpatch: 12-, where "true" suffix truncation was introduced.
* Implement operator class parametersAlexander Korotkov2020-03-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PostgreSQL provides set of template index access methods, where opclasses have much freedom in the semantics of indexing. These index AMs are GiST, GIN, SP-GiST and BRIN. There opclasses define representation of keys, operations on them and supported search strategies. So, it's natural that opclasses may be faced some tradeoffs, which require user-side decision. This commit implements opclass parameters allowing users to set some values, which tell opclass how to index the particular dataset. This commit doesn't introduce new storage in system catalog. Instead it uses pg_attribute.attoptions, which is used for table column storage options but unused for index attributes. In order to evade changing signature of each opclass support function, we implement unified way to pass options to opclass support functions. Options are set to fn_expr as the constant bytea expression. It's possible due to the fact that opclass support functions are executed outside of expressions, so fn_expr is unused for them. This commit comes with some examples of opclass options usage. We parametrize signature length in GiST. That applies to multiple opclasses: tsvector_ops, gist__intbig_ops, gist_ltree_ops, gist__ltree_ops, gist_trgm_ops and gist_hstore_ops. Also we parametrize maximum number of integer ranges for gist__int_ops. However, the main future usage of this feature is expected to be json, where users would be able to specify which way to index particular json parts. Catversion is bumped. Discussion: https://postgr.es/m/d22c3a18-31c7-1879-fc11-4c1ce2f5e5af%40postgrespro.ru Author: Nikita Glukhov, revised by me Reviwed-by: Nikolay Shaplov, Robert Haas, Tom Lane, Tomas Vondra, Alvaro Herrera
* Allow using Unix-domain sockets on Windows in testsPeter Eisentraut2020-03-30
| | | | | | | | | | | | | | | The test suites currently don't use Unix-domain sockets on Windows. This optionally allows enabling that by setting the environment variable PG_TEST_USE_UNIX_SOCKETS. This should currently be considered experimental. In particular, pg_regress.c contains some comments that the cleanup code for Unix-domain sockets doesn't work correctly under Windows, which hasn't been an problem until now. But it's good enough for locally supervised testing of the functionality. Reviewed-by: Andrew Dunstan <andrew.dunstan@2ndquadrant.com> Discussion: https://www.postgresql.org/message-id/flat/54bde68c-d134-4eb8-5bd3-8af33b72a010@2ndquadrant.com
* Be more careful about extracting encoding from locale strings on Windows.Tom Lane2020-03-30
| | | | | | | | | | | | | | | | | | | | | GetLocaleInfoEx() can fail on strings that setlocale() was perfectly happy with. A common way for that to happen is if the locale string is actually a Unix-style string, say "et_EE.UTF-8". In that case, what's after the dot is an encoding name, not a Windows codepage number; blindly treating it as a codepage number led to failure, with a fairly silly error message. Hence, check to see if what's after the dot is all digits, and if not, treat it as a literal encoding name rather than a codepage number. This will do the right thing with many Unix-style locale strings, and produce a more sensible error message otherwise. Somewhat independently of that, treat a zero (CP_ACP) result from GetLocaleInfoEx() as meaning that we must use UTF-8 encoding. Back-patch to all supported branches. Juan José Santamaría Flecha Discussion: https://postgr.es/m/24905.1585445371@sss.pgh.pa.us
* Attempt to fix unstable regression tests, take 2David Rowley2020-03-30
| | | | | | | | | | | Following up on 2dc16efed, petalura has suffered some additional failures in stats_ext which again appear to be around the timing of an autovacuum during the test, causing instability in the row estimates. Again, let's fix this by explicitly performing a VACUUM on the table and not leave it to happen by chance of an autovacuum pass. Discussion: https://postgr.es/m/CAApHDvok5hmXr%2BbUbJe7%2B2sQzWo4B_QzSk7RKFR9fP6BjYXx5g%40mail.gmail.com
* Report waiting via PS while recovery is waiting for buffer pin in hot standby.Fujii Masao2020-03-30
| | | | | | | | | | | | | Previously while the startup process was waiting for the recovery conflict with snapshot, tablespace or lock to be resolved, waiting was reported in PS display, but not in the case of recovery conflict with buffer pin. This commit makes the startup process in hot standby report waiting via PS while waiting for the conflicts with other backends holding buffer pins to be resolved. Author: Masahiko Sawada Reviewed-by: Fujii Masao Discussion: https://postgr.es/m/CA+fd4k4mXWTwfQLS3RPwGr4xnfAEs1ysFfgYHvmmoUgv6Zxvmg@mail.gmail.com
* Improve handling of parameter differences in physical replicationPeter Eisentraut2020-03-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When certain parameters are changed on a physical replication primary, this is communicated to standbys using the XLOG_PARAMETER_CHANGE WAL record. The standby then checks whether its own settings are at least as big as the ones on the primary. If not, the standby shuts down with a fatal error. The correspondence of settings between primary and standby is required because those settings influence certain shared memory sizings that are required for processing WAL records that the primary might send. For example, if the primary sends a prepared transaction, the standby must have had max_prepared_transaction set appropriately or it won't be able to process those WAL records. However, fatally shutting down the standby immediately upon receipt of the parameter change record might be a bit of an overreaction. The resources related to those settings are not required immediately at that point, and might never be required if the activity on the primary does not exhaust all those resources. If we just let the standby roll on with recovery, it will eventually produce an appropriate error when those resources are used. So this patch relaxes this a bit. Upon receipt of XLOG_PARAMETER_CHANGE, we still check the settings but only issue a warning and set a global flag if there is a problem. Then when we actually hit the resource issue and the flag was set, we issue another warning message with relevant information. At that point we pause recovery, so a hot standby remains usable. We also repeat the last warning message once a minute so it is harder to miss or ignore. Reviewed-by: Sergei Kornilov <sk@zsrv.org> Reviewed-by: Masahiko Sawada <masahiko.sawada@2ndquadrant.com> Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com> Discussion: https://www.postgresql.org/message-id/flat/4ad69a4c-cc9b-0dfe-0352-8b1b0cd36c7b@2ndquadrant.com
generated by cgit v1.2.3 (git 2.25.1) at 2025-08-04 06:08:14 +0000