threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "Hi all,\n\nWhile working on the recent issues with 2PC and temporary objects I\nhave added a test case based on current_schema() for the first time in\nhistory, and the buildfarm complained about it, as mentioned here:\nhttps://www.postgresql.org/message-id/20190118005949.GD1883@paquier.xyz\n... |
[
{
"msg_contents": "Hi,\n\nAttached small patch fixes the function name heap_vacuum_rel in the comment.\n\ns/vacuum_heap_rel/heap_vacuum_rel/\n\nRegards,\n\n--\nMasahiko Sawada\nNIPPON TELEGRAPH AND TELEPHONE CORPORATION\nNTT Open Source Software Center",
"msg_date": "Fri, 18 Jan 2019 12:50:15 +0900",
"m... |
[
{
"msg_contents": "I've found that a couple of different OpenBSD 6.4 machines fall over\nbadly in the ecpg regression tests, with output like\n\ntest sql/parser ... ok\ntest thread/thread ... stdout stderr FAILED (test process was terminated by signal 6: Abort trap)\ntest thread... |
[
{
"msg_contents": "Hi all,\n\nc5660e0 has introduced a new flag for MyXactFlags to restrict\ntwo-phase commit from working with temporary objects, and as a matter\nof fact XACT_FLAGS_ACCESSEDTEMPREL has been kept around to keep the\nerror handling message compatible with past versions, still it is\nweird to kee... |
[
{
"msg_contents": "Hi,\n\nAs mentioned in the subject, if the $PATH where you are building postgres\ncontains a perl's special character [1], for example a `+`, pgbench's tap\ntests will fail.\n\nThe outputs looks something like this (full at [2]):\n```\n# Failed test 'file name format'\n# at t/001_pgbench_... |
[
{
"msg_contents": "Continuing the project of getting check-world to pass on OpenBSD 6.4,\nI find that the bin/scripts/ tests fall over:\n\nt/090_reindexdb.pl (Wstat: 7424 Tests: 10 Failed: 2)\n Failed tests: 9-10\n Non-zero exit status: 29\n Parse errors: Bad plan. You planned 23 tests but ran 10.\nt/... |
[
{
"msg_contents": "Hi,\n\nCurrently TestForOldSnapshot() is defined as\n\n\nextern void TestForOldSnapshot_impl(Snapshot snapshot, Relation relation);\n...\n/*\n * Check whether the given snapshot is too old to have safely read the given\n * page from the given table. If so, throw a \"snapshot too old\" error.... |
[
{
"msg_contents": "Hello,\r\nWhile locking heap pages (function RelationGetBufferForTuple() in file hio.c) we acquire locks in increasing block number order to avoid deadlock. In certain cases, we do have to drop and reacquire lock on heap pages (when we have to pin visibility map pages) to avoid doing IO while... |
[
{
"msg_contents": "pg_stat_statements considers a plain select and a select for update to\nbe equivalent, which seems quite wrong to me as they will have very\ndifferent performance characteristics due to locking.\n\nThe only comment about it in the code is:\n\n /* we ignore rowMarks */\n\nI propose that it ... |
[
{
"msg_contents": "I tried to build cmake on OpenBSD running on my old HPPA hardware,\nnot because I cared about cmake per se but because it's a dependency\nof something I did care about. Failed miserably. It turns out cmake has\na hard dependency on libuv which (a) has a hard dependency on atomic ops\nand (b... |
[
{
"msg_contents": "Hello!\n\nI have a proposal for a feature to add to Postgres. I believe it is a natural\nextension to the current standard SQL ON DELETE SET NULL behavior when using\ncomposite foreign keys. The basic idea is that you can specify which columns to\nset to NULL in the DELETE trigger created by ... |
[
{
"msg_contents": "In our PG 10.2(CentOS 7.3) database, the following error is reported when querying a table. We have already restored the production data through backup, but i want to confirm what may be the reason and how to avoid it in the future.\n\n\tlma=# select count(*) from bi_dm.tdm_ttk_site_on_way_rt... |
[
{
"msg_contents": "Hi,\n\nCurrently RelationFindReplTupleByIndex(), RelationFindReplTupleSeq()\nlock the found tuple. I don't quite get what that achieves - why isn't\ndealing with concurrency in the table_update/delete calls at the\ncallsites sufficient? As far as I can tell there's no meaningful\nconcurrency ... |
[
{
"msg_contents": "Hi\n\nremove_useless_join does not prove uniqueness if the unique index is partial, and therefore wont remove the join if no columns are referenced (see example in bottom).\n\nI have been trying to look around the source code and from what I have identified the problem seems to be that \"chec... |
[
{
"msg_contents": "Over in the thread at [1], we realized that PostGIS has been thrashing\naround trying to fake its way to having \"special index operators\", ie\na way to automatically convert WHERE clauses into lossy index quals.\nThat's existed in a non-extensible way inside indxpath.c for twenty\nyears com... |
[
{
"msg_contents": "In [1] I propose that we should allow extensions to get their hands\non the ability to transform function calls as per \"protransform\" and\nto generate lossy index quals based on items in WHERE clauses. The\nAPIs to give an extension control at the right points seem pretty\nstraightforward,... |
[
{
"msg_contents": "When libpq is loaded in the server (libpqwalreceiver, dblink,\npostgres_fdw), it may use libpq environment variables set in the\npostmaster environment for connection parameter defaults. I have\nnoticed that this has some confusing effects in our test suites. I\nwonder if this is a good ide... |
[
{
"msg_contents": "Hi,\n\nI found that ALTER TRIGGER .. RENAME TO is broken and it's\nunable to rename the trigger. Looking further seems renametrig()\nfunction, copy the new trigger name into wrong tuple. This is\nbroken with below commit:\n\ncommit 578b229718e8f15fa779e20f086c4b6bb3776106\nAuthor: Andres Fr... |
[
{
"msg_contents": "Hi,\n Please let me know how to deal with procedures with out parameter and\nreturning out parameters.\nThanks & Regards\nNavneet maddheshiya\n\nHi, Please let me know how to deal with procedures with out parameter and returning out parameters.Thanks & RegardsNavneet maddheshiya",
"... |
[
{
"msg_contents": "---------- Forwarded message ---------\nFrom: Vikramsingh Kushwaha <vskushwaha@mitaoe.ac.in>\nDate: Mon, Jan 21, 2019 at 11:56 PM\nSubject: Fwd: Google Summer Of Code\nTo: <developers@phpmyadmin.net>\n\n\n\n\n---------- Forwarded message ---------\nFrom: Vikramsingh Kushwaha <vskushwaha@mitao... |
[
{
"msg_contents": "Hello\n\nWhile working on bugfixes for FK problems in partitioned tables, I came\nacross some behavior that appears to stem from our inclusion of foreign\nkeys in relcache, without sufficient care for invalidating the relcache\nentries when the foreign key set for the table changes. (Namely,... |
[
{
"msg_contents": "I knew that XLogRecordPageWithFreeSpace() is only called by heapam.c,\nbut I figured indexes had their own logic to xlog free space. While\ninvestigating something else I found that brin indexes, at least, on\nthe secondary have no FSMs. Is this deliberate?\n\n-- \nJohn Naylor ... |
[
{
"msg_contents": "(This is pretty minor, but I struggled to ignore it)\n\nIn RelationGetIndexAttrBitmap() a comment claims /* We return our\noriginal working copy for caller to play with */. 3 of the 4 possible\nBitmapsets follow that comment but for some reason, we make a copy of\nthe primary key attrs before... |
[
{
"msg_contents": "Hi,\n\nAttached find a patch to fix $subject.\n\nThanks,\nAmit",
"msg_date": "Tue, 22 Jan 2019 13:43:32 +0900",
"msg_from": "Amit Langote <Langote_Amit_f8@lab.ntt.co.jp>",
"msg_from_op": true,
"msg_subject": "Typo: llvm*.cpp files identified as llvm*.c"
},
{
"msg_conte... |
[
{
"msg_contents": "Hi all,\n\nWhile working on REINDEX CONCURRENTLY, I have noticed that\nindex_build() does not need its argument isprimary. Perhaps it is\nnot worth bothering, but for the REINDEX CONCURRENTLY business this\nremoves the need to open an index when triggering a concurrent\nbuild.\n\nThe flag wa... |
[
{
"msg_contents": "(Moving this discussion from the pgsql-committers thread \"pgsql:\nUpdate ssl test certificates and keys\", which is innocent.)\n\nOn Wed, Jan 16, 2019 at 10:37 AM Thomas Munro\n<thomas.munro@enterprisedb.com> wrote:\n> On Fri, Jan 4, 2019 at 10:08 AM Thomas Munro\n> <thomas.munro@enterprised... |
[
{
"msg_contents": "I'm finding a massive difference in query execution time between two\nqueries that should be identical:\n\nVery slow:\nselect ... from x natural left join y where y is null\n\nFast:\nselect ... from x natural left join y where y.primary_key_column is null\n\nA fact that I suspect is important... |
[
{
"msg_contents": "Hi,\n\nIn the standard_planner, we set the proper JIT flags on the resulting\nPlannedStmt node. We also offered a planer hook so extensions can\nadd customized planners. The JIT flags in jit/jit.h however, is\nnot present on the system. I think it's probably better to\ninstall the jit headers... |
[
{
"msg_contents": "hello,\n\nI'm Maayan, I'm in a DBA team that uses postgresql.\nI saw in the documentation on wals:\nhttps://www.postgresql.org/docs/10/wal-intro.html\nIn the tip box that, it's better not to use a journaling filesystem. and I\nwanted to ask how it works?\ncan't we get corruption that we can'... |
[
{
"msg_contents": "Dear hackers.\n\nI found a minor typo in the comment pgbench.c.\n\n-\t\t * always allocate one more in order to accomodate the NULL terminator\n+\t\t * always allocate one more in order to accommodate the NULL terminator\n\nAttached fix it.\n\nRegards.\nMoon.\n--------------------------------... |
[
{
"msg_contents": "Hi,\n\nI'm working on an extension which analyzes all possible plans\ngenerated by the planner. I believe this extension would become\nuseful for benchmarking the planner (e.g. the performance of the\nestimation and the cost model) and better understanding the cases\nwhere the planners would ... |
[
{
"msg_contents": "An assorted set of command typos is fixed in the attached patch.\n\n-- \nFabien.",
"msg_date": "Wed, 23 Jan 2019 09:48:12 +0100 (CET)",
"msg_from": "Fabien COELHO <coelho@cri.ensmp.fr>",
"msg_from_op": true,
"msg_subject": "yet another comment typo patch"
},
{
"msg_con... |
[
{
"msg_contents": "Hi,\n\nIt seems that ATExecDetachPartition misses to mark a child's primary key\nconstraint entry (if any) as local (conislocal=true, coninhcount=0), which\ncauses this:\n\ncreate table p (a int primary key) partition by list (a);\ncreate table p2 partition of p for values in (2);\nalter tabl... |
[
{
"msg_contents": "Hi,\n\nI happened to notice that when CopyReadLineText() calls mblen(), it \npasses only the first byte of the multi-byte characters. However, \npg_gb18030_mblen() looks at the first and the second byte. \nCopyReadLineText() always passes \\0 as the second byte, so \npg_gb18030_mblen() will i... |
[
{
"msg_contents": "Detach constraints when partitions are detached\n\nI (Álvaro) forgot to do this in eb7ed3f30634, leading to undroppable\nconstraints after partitions are detached. Repair.\n\nReported-by: Amit Langote\nAuthor: Amit Langote\nDiscussion: https://postgr.es/m/c1c9b688-b886-84f7-4048-1e4ebe9b1d06... |
[
{
"msg_contents": "Hi hacker,\n\nI am working for some time under built-in connection pooler for Postgres:\nhttps://www.postgresql.org/message-id/flat/a866346d-5582-e8e8-2492-fd32732b0783%40postgrespro.ru#b1c354d5cf937893a75954e1e77f81f5\n\nUnlike existed external pooler, this solution supports session semantic... |
[
{
"msg_contents": "We are working to migrate several large tables from the timestamp to the\ntimestamptz data type by using logical replication (so as to avoid long\ndowntime for type conversions). We are using pglogical but curious if what\nI share below applies to native logical replication as well.\n\nBoth ... |
[
{
"msg_contents": "Hi,\n\nIn parse_fcall_arguments_20():\n\nc0a8c3ac13f8 (Tom Lane 2003-05-08 18:16:37 +0000 579) argsize = pq_getmsgint(msgBuf, 4);\n0ac6298bb8ac (Tom Lane 2003-05-09 18:08:48 +0000 580) if (argsize == -1)\n0ac6298bb8ac (Tom Lane 2003-05-09 18:08:48... |
[
{
"msg_contents": "I added some code to selfuncs.c to estimate the selectivity of CTID,\nincluding nullness, in my ongoing attempt to add TID range scans [1]. And\nas Tom pointed out [2], no system attribute can be null, so we might as\nwell handle them all.\n\nThat's what the attached patch does.\nI observed ... |
[
{
"msg_contents": "Hi,\n\nI noticed yet another thing while updating the patch for pushing down\nORDER BY LIMIT. Let me explain. When costing foreign paths on the\nbasis of local statistics, we calculate/cache the costs of an unsorted\nforeign path, and re-use them to estimate the costs of presorted foreign\n... |
[
{
"msg_contents": "Hi, hackers.\n\nWhen running pg_basebackup with -X s with network file system as target we\nwould consistently get \"could not receive data from WAL stream: server\nclosed the connection unexpectedly\".\n\nOn server we see this error firing: \"terminating walsender process due to\nreplication... |
[
{
"msg_contents": "I just noticed (once again) that we have hash_any() declared in\nsrc/access/hash.h (the header file for the hash index AM) rather than\nsomewhere in utils/, for no good reason other than perhaps there was no\nbetter place when it was introduced. This means that some files that\nseem to just ... |
[
{
"msg_contents": "Hackers,\n\nWe have a strong need to make a variant to the crosstab interface so\nthat data that is pivoted one way would be sent through a crosstab\nlike function so that it would be pivoted another way. For example,\nif you had\n\nrow 0: a1, a2, a3, k1, c1, c2, ...\nrow 1: a1, a2, a3, k2, ... |
[
{
"msg_contents": "With our scanner keywords list now more cache-aware, and with us\nplanning to use Bison for years to come, has anyone ever looked at\nreordering the bison state machine array to be more cache aware, e.g.,\nhaving common states next to each other rather than scattered around the\narray?\n\n-- ... |
[
{
"msg_contents": "Hello hackers,\n\nIn his blog post (What's new in SQL 2016)[\nhttps://modern-sql.com/blog/2017-06/whats-new-in-sql-2016], Markus Winand\nexplained some of the changes added to SQL:2016. I spotted that Postgres\nwas behind other RDBMS on hyperbolic functions and log10 function.\nThe log10 func... |
[
{
"msg_contents": "I see that opossum's owner just resurrected it after a period of being\nMIA ... and it's failing a few regression tests with symptoms like\nthis:\n\n SELECT 'NaN'::float4;\n- float4 \n---------\n- NaN\n+ float4 \n+----------\n+ Infinity\n (1 row)\n\nI have no doubt that this is caused by... |
[
{
"msg_contents": "Recently, buildfarm member curculio has started to show a semi-repeatable\nfailure in src/test/recovery/t/013_crash_restart.pl:\n\n# aborting wait: program died\n# stream contents: >>psql:<stdin>:8: no connection to the server\n# psql:<stdin>:8: connection to server was lost\n# <<\n# pattern ... |
[
{
"msg_contents": "I understand it's not fully supported to replicate to a differently\npartitioned setup on a subscriber with either pglogical or the native\nlogical replication, however I also know that INSERT triggers can be fired\nin replication mode. I have an insert-only OLTP table that I want\npartition... |
[
{
"msg_contents": "Hello,\n\nUsing either the .dir-locals.el settings or the \"more complete\"\nsrc/tools/editors/emacs.samples, I have never convinced Emacs to\nproduce multi-line function declarations in .h files that satisfy\npg_indent.\n\nFor example, latch.c has the following definition:\n\nint\nWaitEventS... |
[
{
"msg_contents": "Hello,\n\nI made a utility program to dump PostgreSQL database in Apache Arrow format.\n\nApache Arrow is a kind of data format for columnar-based structured\ndata; actively\ndeveloped by Spark and comprehensive communities.\nIt is suitable data representation for static and read-only but lar... |
[
{
"msg_contents": "Avoid creation of the free space map for small heap relations.\n\nPreviously, all heaps had FSMs. For very small tables, this means that the\nFSM took up more space than the heap did. This is wasteful, so now we\nrefrain from creating the FSM for heaps with 4 pages or fewer. If the last\nknow... |
[
{
"msg_contents": "In the pluggable-storage discussion, there was some talk of renaming\nnodes/relation.h to avoid confusion with the new access/relation.h\nheader. I think this is a fine idea, not only because of that conflict\nbut because \"relation.h\" has never made a lot of sense as the file's\nname.\n\nA... |
[
{
"msg_contents": "Since the LLVM stuff went in, src/tools/pginclude/cpluspluscheck\nfails on my main devel machine, because\n\nIn file included from /tmp/cpluspluscheck.jobsM6/test.cpp:3:\n./src/include/jit/llvmjit_emit.h:13:25: error: llvm-c/Core.h: No such file or directory\n\nand then there's a bunch of ens... |
[
{
"msg_contents": "In <6044.1548524131@sss.pgh.pa.us> I worried about how much planner\nstuff that patch might end up dragging into files that contain\nplanner support functions, and suggested that we could refactor the\nplanner's header files to reduce the inclusion footprint. Attached\nare some proposed patc... |
[
{
"msg_contents": "Hi,\n\nBecause of pg_pwrite()[1], openLogOff, a static global variable in xlog.c,\nseems taken over by a local variable startoffset and no longer used now.\n\nI write the attached patch that removes openLogOff. Both \"make check\" and\n\"make installcheck\" passed, and just after that, \"pg_c... |
[
{
"msg_contents": "Hi all,\n\nIn 2016, a thread was started about implementing INSERT INTO ... SET ... \nthat included a patch and was basically ready for inclusion in \nPostgreSQL. However, it seems as though it stagnated for some reason. \nDoes anybody remember this and is there perhaps someone who knows what... |
[
{
"msg_contents": "Hi,\n\nPeter Eisentraut wrote:\n> When you manage the WAL (or perhaps in the future relation files)\n> through PMDK, is there still a file system view of it somewhere, for\n> browsing, debugging, and for monitoring tools?\n\nFirst, I assume that our patchset is used with a filesystem that sup... |
[
{
"msg_contents": "Hello.\n\nhttps://www.postgresql.org/message-id/20190128.133143.115303042.horiguchi.kyotaro@lab.ntt.co.jp\n\n> C. Provide session-specific GUC variable (that overides the global one)\n> - Add new configuration file \"postgresql.conf.<PID>\" and\n> pg_reload_conf() let the session with... |
[
{
"msg_contents": "Dear all,\n\nPlease find attached the patch extending the sets of symbols allowed in\nltree labels. The patch introduces 2 variants of escaping symbols, via\nbackslashing separate symbols and via quoting the labels in whole for\nltree, lquery and ltxtquery datatypes.\n\n-- \nSY, Dmitry Belyav... |
[
{
"msg_contents": "Hi,\n\nSorry but I found that the patchset v2 had a bug in managing WAL segment\nfile offset. I fixed it and updated a patchset as v3 (attached).\n\nRegards,\nTakashi\n\n-- \nTakashi Menjo - NTT Software Innovation Center\n<menjo.takashi@lab.ntt.co.jp>",
"msg_date": "Wed, 30 Jan 2019 15:... |
[
{
"msg_contents": "Hello,\nWe are using Postgresql (postgresql-bdr94).\n\nPlease let us know whether Postgresql supports Open JDK?\n\nThanks\nPratheej\n\nHello,We are using Postgresql (postgresql-bdr94).Please let us know whether Postgresql supports Open JDK? ThanksPratheej",
"msg_date": "Wed, 30 Jan 2019 ... |
[
{
"msg_contents": "Hi all,\n\nI just got a run with CFLAGS with the following options:\n-Wunused-function -Wunused-variable -Wunused-const-variable\n -Wno-unused-result -Wunused-parameter -Wunused-macros\n\nAnd while this generates a lot of noise for callbacks and depending on\nthe environment used, this is als... |
[
{
"msg_contents": "Hi\n\nWhile trying out the progress report mechanism for btrees, I noticed\nthis strange chain of errors:\n\n2019-01-29 15:51:55.928 -03 [43789] ERROR: no se pudo crear el �ndice �nico �a_generate_series_idx�\n2019-01-29 15:51:55.928 -03 [43789] DETALLE: La llave (generate_series)=(... |
[
{
"msg_contents": "Hi hackers,\n\nOne of our customers was faced with the following problem:\nhe has setup physical primary-slave replication but for some reasons \nspecified very large (~12 hours)\nrecovery_min_apply_delay. I do not know precise reasons for such large \ngap between master and replica.\nBut ev... |
[
{
"msg_contents": "Hi,\n\nMy colleague Alexander Lakhin has noticed an assertion failure in\nreorderbuffer.c:1330. Here is a simple snippet reproducing it:\n\nSELECT 'init' FROM pg_create_logical_replication_slot('regression_slot', 'test_decoding');\n\ncreate table t(k int);\nbegin;\nsavepoint a;\nalter table t... |
[
{
"msg_contents": "Hi Nagaura-san,\r\n\r\nThank you for your review and advice.\r\n\r\n> This merit was very helpful for my use, so I want your proposal function in\r\n> postgres.\r\nThank you. \r\n\r\n> 1)\r\n> It would be better making the log format the same as the server log format,\r\n> I think.\r\n> Your ... |
[
{
"msg_contents": "Here's a bug tracking system that Nathan set up many years ago and\napparently has kept going unattended. It seems to me that it's\nsomething that we could base a semi-official bug tracking system on.\n\nhttps://granicus.if.org/pgbugs/\n\nThere are clearly things that need fixed, such as fai... |
[
{
"msg_contents": "Hi\n\nI have a value '400 days 30 sec':: interval\n\nI don't understand why cast\n\npostgres=# select '400days 30sec'::interval minute to second;\n+-------------------+\n| interval |\n+-------------------+\n| 400 days 00:00:30 |\n+-------------------+\n(1 row)\n\nproduce 400 days\n\n... |
[
{
"msg_contents": "Hi, hackers.\n\n It seems that heapam.c:3082 calls XLogRegisterData() with an argument\nallocated on stack, but following call to XLogInsert() happens after\nend of context for that variable.\n Issue spotted by clang's AddressSanitizer. Fix attached.\n\n--\nStas Kelvich\nPostgres Profession... |
[
{
"msg_contents": "At least on my FreeBSD 11 box, the current definition of\n$(with_temp_install) is not sufficient to ensure that the various .so\nfiles are loaded from tmp_install and not from the compiled rpath (which\nwill be the final install dir, which may of course contain old\nlibraries).\n\nLD_LIBRARY_... |
[
{
"msg_contents": "Hi Hackers,\n\nDuring the testing of new feature allowing group access mode for the data\ndirectory by specifying initdb --allow-group-access by one of my collegue,\nwe didn't observe any change in the data folder permissions. With/without\ngroup access the data folder have permissions to the... |
[
{
"msg_contents": "Sorry, I lost previous mail[1].\n\nOn Fri, 28 Dec 2018 at 20:36, Tsunakawa, Takayuki\n<tsunakawa(dot)takay(at)jp(dot)fujitsu(dot)com> wrote:\n> Although I may say the same thing as you, I think a natural idea would be to create a generic plan gradually. The starting simple question is \"why ... |
[
{
"msg_contents": "postgres@Server2[DEV][backup] $ psql -c 'select * from pg_class where relname\n= pg_toast_22345'\nERROR: column \"pg_toast_16387\" does not exist\nLINE 1: select * from pg_class where relname = pg_toast_22345\n\n\n\n\n-----\n--\nThanks,\nRajan.\n--\nSent from: http://www.postgresql-archive.o... |
[
{
"msg_contents": "I just read an article about the advantages of zheap over the heap that\nPostgreSQL currently uses and there were several posts on this list about\nit being more compact as well as more performant. I am hopeful that it\nmakes it into PostgreSQL 12.0, but it is unclear what the current status ... |
[
{
"msg_contents": "Hi,\n\nThis patch series is to add support for spgist quadtree @<(point,circle)\noperator. The first two patches are to refactor existing code before\nimplemention the new feature. The third commit is the actual implementation\nprovided with a set of simple unit tests.\n\nMatwey V. Kornilov (... |
[
{
"msg_contents": "Hi,\n\nWhile working on the patch to slotify trigger.c I got somewhat confused\nby the need to expand tuples in trigger.c:\n\nstatic HeapTuple\nGetTupleForTrigger(EState *estate,\n EPQState *epqstate,\n ResultRelInfo *relinfo,\n ItemPointe... |
[
{
"msg_contents": "Hi,\n\nIn https://postgr.es/m/20190201162404.onngi77f26baem4g%40alap3.anarazel.de\nI noticed that composite_to_json() deforms column-by-column. Given that\nit always processes all columns, that seems quite the waste of resources.\n\nIn some quick'n dirty dirty testing this gives a ~4% benefit... |
[
{
"msg_contents": "While testing an xidStopLimit corner case, I got this:\n\n3656710 2019-01-05 00:05:13.910 GMT LOG: automatic aggressive vacuum to prevent wraparound of table \"test.pg_toast.pg_toast_826\": index scans: 0\n3656710 2019-01-05 00:05:16.912 GMT LOG: could not truncate directory \"pg_xact\": ap... |
[
{
"msg_contents": "Please help to understand the following. Where the User(who is not the owner\nof a table) is able to ALTER DEFAULT PRIVILEGES and GRANT SELECT rights for\nall tables???? Is providing USAGE on schema is enough to do that? How is\nthis secure?\n\nlearning=> select current_user;\n current_user\n... |
[
{
"msg_contents": "Hello\n\nIn reply to https://postgr.es/m/alpine.DEB.2.21.1901102211350.27692@lancre\nwherein Fabien wrote:\n\n> I'm not very happy with the resulting syntax, but IMO the feature is useful.\n> My initial design was to copy PL/pgSQL \"into\" with some \"\\into\" orthogonal\n> to \\; and ;, but ... |
[
{
"msg_contents": "I've been poking at the problem discussed in a couple of recent threads\nof letting extensions in on the ability to create \"lossy index conditions\"\nfor complex operators/functions. The current design for that in\nindxpath.c is frankly just a pile of kluges of varying ages. In the\ninitia... |
[
{
"msg_contents": "pgcrypto bundles a copy of the imath library for arbitrary-precision integer\narithmetic in non-SSL builds. Upstream fixed buffer overflows through the\nyears, and commit 8b59672 brought those fixes into PostgreSQL. In master, I\nwould like to fully resynchronize with fresh imath 1.29. We'... |
[
{
"msg_contents": "I happened to notice while looking at code coverage reports that\n\n(1) the portions of create_unique_path() that report that the input\nrelation is already known unique are no longer reached in our\nregression tests;\n\n(2) the loop in relation_has_unique_index_for() that deals with\nan expr... |
[
{
"msg_contents": "Avoid creation of the free space map for small heap relations, take 2.\n\nPreviously, all heaps had FSMs. For very small tables, this means that the\nFSM took up more space than the heap did. This is wasteful, so now we\nrefrain from creating the FSM for heaps with 4 pages or fewer. If the la... |
[
{
"msg_contents": "In the name of god!\n\nHi,\n\nWhat happens if checkpoint haven't completed until the next checkpoint \ninterval or max_wal_size?\n\nthanks,\n\nregards, Mohammad Sherafat.\n\n\n",
"msg_date": "Mon, 4 Feb 2019 09:06:39 +0330",
"msg_from": "Mohammad Sherafat <mhdsherafat@chmail.ir>",
... |
[
{
"msg_contents": "Hi all,\n\nAs per $subject, CF 2019-01 is now closed for business. Here is the\nfinal score:\nCommitted: 58.\nMoved to next CF: 113.\nWithdrawn: 4.\nRejected: 3.\nReturned with Feedback: 52.\nTotal: 230.\n\nI have done a pass over all the remaining entries, updating them\naccording to their ... |
[
{
"msg_contents": "There are a few regexes in pg_dump's tap tests that are neglecting to\nescape the dot in \"schema.table\" type expressions. This could result\nin the test passing when it shouldn't. It seems worth putting that\nright.\n\nPatch attached.\n\n-- \n David Rowley http://www.2ndQ... |
[
{
"msg_contents": "I think the recovery parameters\n\n archive_cleanup_command\n promote_trigger_file\n recovery_end_command\n recovery_min_apply_delay\n\ncan be changed from PGC_POSTMASTER to PGC_SIGHUP without any further\ncomplications (unlike for example primary_conninfo, which is being\ndiscuss... |
[
{
"msg_contents": "Hello\n\nIf you do ALTER TABLE .. REPLICA IDENTITY to a partitioned table, the\ncommand operates on the parent table itself and does not propagate to\npartitions. Why is this? Maybe not recursing was the right call when\nwe only had regular inheritance (back in 9.4), but since partitioned\n... |
[
{
"msg_contents": "Hi,\n\ntrigger.c goes through some trouble to free the tuples returned by\ntrigger functions. There's plenty codepaths that look roughly like:\n\t\tif (oldtuple != newtuple && oldtuple != slottuple)\n\t\t\theap_freetuple(oldtuple);\n\t\tif (newtuple == NULL)\n\t\t{\n\t\t\tif (trigtuple != fdw... |
[
{
"msg_contents": "The docs in PG11 and master both state:\n\nWhen an UPDATE causes a row to move from one partition to another,\nthere is a chance that another concurrent UPDATE or DELETE misses this\nrow. Suppose session 1 is performing an UPDATE on a partition key, and\nmeanwhile a concurrent session 2 for w... |
[
{
"msg_contents": "Hello.\n\nThe following command complains with an internal\nerror. (allow_system_table_mods is on).\n\nalter table pg_attribute set (fillfactor = 90);\n> ERROR: AccessExclusiveLock required to add toast table.\n\nThe same happens for pg_class. This is because ATRewriteCatalogs\ntries to add ... |
[
{
"msg_contents": "I came across an edge case in how our foreign key implementation works\nthat I think is not SQL conforming. It has to do with how updates to\nvalues that \"look\" different but compare as equal are cascaded. A\nsimple case involves float -0 vs. 0, but relevant cases also arise with\ncitext ... |
[
{
"msg_contents": "While poking at bug #15613 (in which FDWs are failing to mark created\nPaths with correct outer-reference sets), I thought it'd be a good idea\nto add some asserts to Path creation saying that every Path should be\nparameterized by at least whatever the relation's required LATERAL\nreferences... |
[
{
"msg_contents": "Hi,\n\nThe first example below works while the second one is a syntax error \ndespite that the documentation \n(https://www.postgresql.org/docs/11/sql-createtableas.html) makes it \nseem like both should be valid. I do not see any reason to not support \nCTAS with both IF NOT EXISTS and EXECU... |
[
{
"msg_contents": "Hello,\n\nCurrently either the table level option `toast_tuple_target` or the compile\ntime default `TOAST_TUPLE_TARGET` is used to decide whether a new tuple\nshould be compressed or not. While this works reasonably well for most\nsituations, at times the user may not want to pay the overhea... |
[
{
"msg_contents": "In https://postgr.es/m/1676.1548726280@sss.pgh.pa.us Tom Lane wrote:\n\n> Sure: every errcode we have is unsafe to treat this way.\n> \n> The backend coding rule from day one has been that a thrown error requires\n> (sub)transaction cleanup to be done to make sure that things are back in a\n>... |
[
{
"msg_contents": "The errorhandling in be_tls_init(), and functions called from it, set the\nappropriate elevel by the isServerStart. ssl_protocol_version_to_openssl() is\nhowever erroring out unconditionally with ERROR on invalid TLS versions. The\nattached patch adds isServerStart handling to the TLS versi... |
[
{
"msg_contents": "Hi,\n\nWhile working on benchmarks for the syscache patch (negative entries and\nall of that), I've ran into a an issue in remove_from_unowned_list.\n\nIf you create a lot of relations in a single transaction (say, 100k) and\nthen abort the transaction (or if it fails for some reason, e.g. be... |
[
{
"msg_contents": "\nThese two files are used by TAP test suites but fall foul of the\ntightening of perl's searchpath rules. See for example\n<https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=bowerbird&dt=2019-02-07%2000%3A10%3A31>\n\nThe obvious solution is to perform the same surgery for the ssl and\n... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.