threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "On Mon, Nov 21, 2016 at 1:59 PM, Kouhei Kaigai <kaigai@ak.jp.nec.com> wrote:\n\n> Hello,\n>\n> The attached patch is a revised version of pass-down LIMIT to FDW/CSP.\n>\n> Below is the updates from the last version.\n>\n> 'ps_numTuples' of PlanState was declared as uint64, instead of long\n> ... |
[
{
"msg_contents": "There's 11 functions which are marked immutable, but are marked as\nparallel unsafe.\n\npostgres=# select proname from pg_proc where provolatile = 'i' and\nproparallel = 'u';\n proname\n-----------------------------\n _pg_expandarray\n _pg_keysequal\n _pg_truetypid\n _pg_truetypmod\... |
[
{
"msg_contents": "Hi all\n\nNoticed this while reading something unrelated\n\nextern PGDLLIMPORT pid_t PostmasterPid;\nextern bool IsPostmasterEnvironment;\nextern PGDLLIMPORT bool IsUnderPostmaster;\nextern bool IsBackgroundWorker;\nextern PGDLLIMPORT bool IsBinaryUpgrade;\n\nI don't see any sane reason for s... |
[
{
"msg_contents": "Hi,\n\nThe SSL test suite (src/test/ssl) is broken in the master since commit \n9a1d0af4ad2cbd419115b453d811c141b80d872b, which is Robert's refactoring \nof getting the server hostname for GSS, SSPI, and SSL in libpq.\n\nThe error we get in the test suite:\n\n# Running: psql -X -A -t -c SELEC... |
[
{
"msg_contents": "Hi\n\nHere is a tiny patch to fix a typo in execParallel.c.\n\n-- \nThomas Munro\nhttp://www.enterprisedb.com\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://www.postgresql.org/mailpref/pgsql-hackers",
"msg_date... |
[
{
"msg_contents": "Hi,\n\nwhile doing some benchmarking, I've once again got confused by the \ndefault settings for checkpoint_flush_after and bgwriter_flush_after. \nThe sample config says this:\n\n#checkpoint_flush_after = 0 # 0 disables,\n # default is 256kB on linux, 0 otherw... |
[
{
"msg_contents": "\nDear pgsql community,\n\nI've been using postgres for a long time. Recently I'm doing table\nsharding over a bunch of pgsql instances. I'm using multiple tablespaces\none per disk to utilize all the IO bandwidth. Things went on pretty\nwell, however there is a troublesome problem I have whe... |
[
{
"msg_contents": "Hi all,\n\nIn some cases the meta tuple in pg_class for a specific relation is\ninvalid, which will cause relcache leak, and then report warning:\nWARNING: relcache reference leak: relation \"pg_class\" not closed.\n\nThe diff file in the attachment can fix this problem.\n\n\n-- \nSent via pg... |
[
{
"msg_contents": "I've been trying to figure out an issue with operators not being pushed\ndown for user defined types, in this case \"hstore\". TL;DR:\n\nhstore=# explain (verbose) select * from hs_fdw where h -> 'a' = '1';\n QUERY PLAN\n-------------------------------------------... |
[
{
"msg_contents": "Hi,\n\nwhile working on my faster expression evaluation stuff I noticed that a\nlot of expression types that call functions don't call the necessary\nfunctions to make track_functions work.\n\nExecEvalFunc/ExecEvalOper (via ExecMakeFunctionResultNoSets) call\npgstat_init_function_usage/pgstat... |
[
{
"msg_contents": "Hi,\n\nthe query below triggers an assertion in TS_phrase_execute. Testing was\ndone on master at dbdfd11.\n\nregards,\nAndreas\n\n-- TRAP: FailedAssertion(\"!(curitem->qoperator.oper == 4)\", File: \"tsvector_op.c\", Line: 1432)\n\nselect 'moscow' @@\n ts_rewrite('moscow', 'moscow',\n... |
[
{
"msg_contents": "It suddenly struck me that the problem being complained of in bug #14434\nis that dependency.c's findDependentObjects() is handling extension\ndependencies incorrectly. It has this logic:\n\n /*\n * This object is part of the internal implementation of\n ... |
[
{
"msg_contents": "Folks,\n\nWhile updating some extensions, I noticed that pg_config --version\nproduces output that's...maybe not quite as useful as it might be, at\nleast to a machine, so I'd like to throw out some proposals to fix the\nsituation.\n\n Add a --version-numeric option to pg_config\n\nor\n ... |
[
{
"msg_contents": "We are proposing changing the JDBC version from\n9.4.xxxx to 42.x.x\n\nWe have two issues we are trying to address here.\n\n1) we do not want to be tied to the server release schedule. This has been\nsomewhat addressed already but has left us with the second issue.\n\n2) Avoid confusion as to... |
[
{
"msg_contents": "Buildfarm member skink failed a couple days ago:\nhttp://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=skink&dt=2016-11-25%2017%3A50%3A01\n\nI believe the interesting parts of the log are\n\n2016-11-25 18:29:03.285 UTC [583882e7.2a45:1] LOG: autovacuum: dropping orphan temp table \"(null)\... |
[
{
"msg_contents": "Hi hackers,\n\nHASH_CHUNK_SIZE is defined as 1024 * 32 = 0x8000. The size of the\nchunks that nodeHash.c passes to palloc is that +\noffsetof(HashJoinMemoryChunkData, data), which is 0x20 here. So we\nask aset.c for 0x8020 bytes. Sizes in that range are sent directly to\nmalloc, after addi... |
[
{
"msg_contents": "Hello.\n\nI noticed that the psql completion code for \"ALTER TABLE x ALTER\n[COLUMN] x DROP\" is wrong. It works as the following\n\n=# alter table x alter x drop <tab>\n[nothing suggested]\n=# alter table x table x alter x drop <tab>\nDEFAULT NOT NULL \n\nThe attached patch fixes it.\n\n... |
[
{
"msg_contents": "Attached fixes reference in a comment to a non-existent function:\n\ns/GetRelationInfo/get_relation_info/g\n\nThanks,\nAmit\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://www.postgresql.org/mailpref/pgsql-hackers",... |
[
{
"msg_contents": "Hi hackers!\n\nHere is a patch with more concurrency-friendly vacuum of GIN.\n\n===Short problem statement===\nCurrently GIN vacuum during cleaning posting tree can lock this tree\nfor a long time without real need.\n\n===Problem statement===\nVacuum of posting tree (function ginVacuumPosting... |
[
{
"msg_contents": "This patch adds two very simple psql commands: \\quit_if and \\quit_unless.\n\nEach takes an optional string parameter and evaluates it for truthiness via\nParseVariableBool().\n\nIf a true-ish value is passed to \\quit_if, psql will behave as if the user\nhad input \\quit.\n\n\\quit_unless w... |
[
{
"msg_contents": "We think we have discovered a bug in the physical replication slots\nfunctionality in PostgreSQL 9.5.\nWe've seen the behavior across Operating Systems (CentOS-7 and openSUSE\nLEAP 42.1), filesystems (ext4 and xfs), and versions (9.5.3 and 9.5.4). All\nwere on x86_64.\n\nWe notice that if we ... |
[
{
"msg_contents": "With current limits, the most bgwriter can do (with 8k pages) is 1000 \npages * 100 times/sec = 780MB/s. It's not hard to exceed that with \nmodern hardware. Should we increase the limit on bgwriter_lru_maxpages?\n-- \nJim Nasby, Data Architect, Blue Treble Consulting, Austin TX\nExperts in A... |
[
{
"msg_contents": "contrib/btree_gist declares type gbtreekey32 without any particular\nalignment spec, which means it defaults to 4-byte (\"integer\") alignment.\nHowever, btree_interval.c supposes that it can map a pair of intervals\nonto that storage. Since an interval contains a float8, that means we\nend ... |
[
{
"msg_contents": "There's a complaint in bug #14438 about poor estimation of join size\nfor a semijoin whose inner side is empty. I think the root of it is that,\nhaving no statistics for the empty table, eqjoinsel_semi distrusts its\nestimate of the number of distinct values on the inner side, and falls\nbac... |
[
{
"msg_contents": "The following looks like a thinko, which fixed in attached:\n\n- Oid proparallel = func_parallel(...\n+ char proparallel = func_parallel(...\n\nThanks,\nAmit\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make ch... |
[
{
"msg_contents": "New (9.4) XactLockTableWait() sets the reason for the wait, so that if\nwe emit a log message then it will produce a log message to say its\nnot waiting on a transaction, its waiting on a lock.\n\nNice. log_lock_waits works great.\n\nNew (9.6) wait_event code is not correctly set. In ProcSlee... |
[
{
"msg_contents": "src/backend/access/heap/README.tuplock says we do this...\n\nLockTuple()\nXactLockTableWait()\nmark tuple as locked by me\nUnlockTuple()\n\nonly problem is we don't... because EvalPlanQualFetch() does this\n\nXactLockTableWait()\nLockTuple()\n\nIf README.tuplock's reasons for the stated lock ... |
[
{
"msg_contents": "Hello,\n\nWhile reading the documentation, I noticed an tiny inconsistency at the \nend of:\n\n \thttps://www.postgresql.org/docs/devel/static/app-psql.html\n\n testdb=> SELECT first, second, first > 2 AS gt2 FROM my_table;\n first | second | ge2\n -------+--------+-----\n 1 | one ... |
[
{
"msg_contents": "Ok, here's my second attempt at refactoring random number generation. \nPrevious attempt crashed and burned, see \nhttps://www.postgresql.org/message-id/E1bw3g3-0003st-6M@gemulon.postgresql.org. \nThis addresses the issues pointed out in that thread.\n\nThe goals are:\n\n* Have a pg_strong_ra... |
[
{
"msg_contents": "Greetings,\n\nNot sure if anyone else has been seeing these, but I'm getting a bit\ntired of them. Neither is a live bug, but they also seem pretty simple\nto fix. The attached patch makes both of these warnings go away. At\nleast for my common build, these are the only warnings that are t... |
[
{
"msg_contents": "Hi hackers,\n\nAttached is a patch which adds diff= directives to .gitattributes for C,\nPerl and (X|SG)ML files. This makes word diffs and the function\nindicator in the diff chunk header and more useful.\n\n\n\n\n-- \n\"A disappointingly low fraction of the human race is,\n at any given ti... |
[
{
"msg_contents": "The following sentence in the ALTER TABLE documentation is not entirely\naccurate:\n\n\"All the actions except RENAME, SET TABLESPACE and SET SCHEMA can be\ncombined into a list of multiple alterations to apply in parallel.\"\n\nSET TABLESPACE (in the ALTER TABLE form) can be combined with ot... |
[
{
"msg_contents": "The following sentence in the ALTER TABLE documentation is not entirely\naccurate:\n\n\"All the actions except RENAME, SET TABLESPACE and SET SCHEMA can be\ncombined into a list of multiple alterations to apply in parallel.\"\n\nSET TABLESPACE (in the ALTER TABLE form) can be combined with ot... |
[
{
"msg_contents": "Hello,\r\n\r\n Please execute me if I am using the wrong mailing list, but I ask the question in pgsql-admin, looks like no one know the answer.\r\n\r\n\r\nwe upgraded our pg db to 9.6, as we know, pg9.6 doesn't need full table scan in vacuum freeze.\r\n\r\nhttp://rhaas.blogspot.com/201... |
[
{
"msg_contents": "Hi,\n\nthere is a missing \"EXEC\" in ecpg.sgml in the list of transaction management commands.\n\nAttached a trivial patch to fix this.\n\nBest regards\nTobias\n\n\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://ww... |
[
{
"msg_contents": "Hi\n\nSmall doc patch to clarify how much of the query text is show in pg_stat_statements\nand a link to the relevant GUC.\n\n\nRegards\n\nIan Barwick\n\n-- \n Ian Barwick http://www.2ndQuadrant.com/\n PostgreSQL Development, 24x7 Support, Training & Services\n\n\n\n\n--... |
[
{
"msg_contents": "Hi,\n\nIt appears openssl has removed the public definition of EVP_CIPHER_CTX\nleading to pgcrypto failing with:\n\n/home/andres/src/postgresql/contrib/pgcrypto/openssl.c:253:17: error: field ‘evp_ctx’ has incomplete type\n EVP_CIPHER_CTX evp_ctx;\n ^~~~~~~\n/home/andres/src/... |
[
{
"msg_contents": "I've found myself wondering \"where is my xlog\" after running\npg_switch_xlog() in 10.0.\n\nRenaming pg_xlog to pg_wal created inconsistency between tools, function\nnames and directory name on disk.\n\nShould we also:\n\n- rename pg_switch_xlog and friends to pg_switch_wal?\n- rename pg_rec... |
[
{
"msg_contents": "Hello\n\nHow should I mark a function which calls CURRENT_DATE? Parallel safe or parallel restricted?\n\npg_proc shows that now() is marked as restricted, but transaction_timestamp() is marked as safe.\n\nThe manual (https://www.postgresql.org/docs/9.6/static/functions-datetime.html) says tha... |
[
{
"msg_contents": "Use latch instead of select() in walreceiver\n\nReplace use of poll()/select() by WaitLatchOrSocket(), which is more\nportable and flexible.\n\nAlso change walreceiver to use its procLatch instead of a custom latch.\n\nFrom: Petr Jelinek <petr@2ndquadrant.com>\n\nBranch\n------\nmaster\n\nDet... |
[
{
"msg_contents": "Permit dump/reload of not-too-large >1GB tuples\n\nOur documentation states that our maximum field size is 1 GB, and that\nour maximum row size of 1.6 TB. However, while this might be attainable\nin theory with enough contortions, it is not workable in practice; for\nstarters, pg_dump fails ... |
[
{
"msg_contents": "Add max_parallel_workers GUC.\n\nIncrease the default value of the existing max_worker_processes GUC\nfrom 8 to 16, and add a new max_parallel_workers GUC with a maximum\nof 8. This way, even if the maximum amount of parallel query is\nhappening, there is still room for background workers th... |
[
{
"msg_contents": "Introduce dynamic shared memory areas.\n\nProgrammers discovered decades ago that it was useful to have a simple\ninterface for allocating and freeing memory, which is why malloc() and\nfree() were invented. Unfortunately, those handy tools don't work\nwith dynamic shared memory segments bec... |
[
{
"msg_contents": "Current HaveNFreeProcs() iterates through the entire freeProcs list \n(while holding ProcStructLock) just to determine if there's a small \nnumber (superuser_reserved_connections) of free slots available. For the \ncommon case, presumably it'd be faster to put the n<=0 test inside the \nloop ... |
[
{
"msg_contents": "Hi,\n\nthe new hash index code on 11003eb failed an assertion yesterday:\n\n TRAP: FailedAssertion(\"!(bucket == obucket)\", File: \"hashpage.c\", Line: 1037)\n\nStatement was\n\n update public.hash_i4_heap set seqno = public.hash_i4_heap.random;\n\nIt can be reproduced with the data di... |
[
{
"msg_contents": "Hi all.\n\nI just bumped into the following post:\nhttps://blogs.windows.com/buildingapps/2016/12/02/symlinks-windows-10/#IAHXCu04bDmrrhxi.97\n\nBut... That's still Windows-specific as it uses a non-POSIX routine\nCreateSymbolicLink():\nhttps://msdn.microsoft.com/en-us/library/windows/desktop... |
[
{
"msg_contents": "I'm trying to get a new build of 9.6.1 working on Debian\nstable and I'm seeing some odd behavior where things work\ncorrectly if I run psql when logged in as user 'postgres',\nbut if I'm logged-in as user 'doom' (my usual login), I don't\nseem to have any select privileges. Even this fails\... |
[
{
"msg_contents": "Hi all,\n\nIs there any reason why the max values of\nautovacuum_vacuum/analyze_scale_factor are 100.0? These max values are\ndefined since when the parameters has been introduced but I think that\n1.0 is enough.\n\nPatch attached.\n\nRegards,\n\n--\nMasahiko Sawada\nNIPPON TELEGRAPH AND TELE... |
[
{
"msg_contents": "Hi\n\nI found some crazy queries in one customer application. These queries are\nstupid, but it was surprise for me so there are not some simple optimization\n\ncreate table foo(a int);\ninsert into foo select generate_series(1,100000);\nanalyze foo;\nexplain select * from foo where a <> a;\n... |
[
{
"msg_contents": "\nWindows XP has been past end of life for quite some time. Nevertheless I \nhave kept my instance running with three buildfarm members: frogmouth, \ncurrawong and brolga. Howeever, a recent commit (apparently fa2fa99, but \nI'm not 100% sure) started causing a compiler segmentation fault on ... |
[
{
"msg_contents": "max_connections is a frequent point of contention between users and \ndevelopers. Users want to set it high so they don't have to deal with \nYet More Software (pgpool or pgBouncer); PG developers freak out because \nbackends are pretty heavyweight, there's some very hot code that's \nsensiti... |
[
{
"msg_contents": "Hi,\n\non master as of a0ae54d, there's a 1 in 10e6 chance sqlsmith catches\ngather_readnext reading beyond the gatherstate->readers array with\nreaders[gatherstate->readnext]. Sample backtrace below.\n\nAs readnext is never explicitly initialized, I think what happens is\nthat a rescan gets... |
[
{
"msg_contents": "I looked into the issue reported in bug #14448,\nhttps://www.postgresql.org/message-id/20161205143037.4377.60754%40wrigleys.postgresql.org\n\nThe core of it seems to be that expandRTE() will report the type and\ntypmod of a column of a VALUES construct as being exprType() and\nexprTypmod() of... |
[
{
"msg_contents": "Add support for restrictive RLS policies\n\nWe have had support for restrictive RLS policies since 9.5, but they\nwere only available through extensions which use the appropriate hooks.\nThis adds support into the grammer, catalog, psql and pg_dump for\nrestrictive RLS policies, thus reducing... |
[
{
"msg_contents": "Hi Everyone,\n\nTL;DR: Making things faster. Architectural evalation.\n\nas some of you might be aware I've been working on making execution of\nlarger queries in postgresl faster. While working on \"batched execution\"\nI came to the conclusion that, while necessary, isn't currently showing\... |
[
{
"msg_contents": "Hi all.\n\nAny chance to get tzdata 2016j in supported branches?\n\n--\nMay the force be with you…\nhttps://simply.name\n\n\nHi all.Any chance to get tzdata 2016j in supported branches?\n--May the force be with you…https://simply.name",
"msg_date": "Tue, 6 Dec 2016 18:38:11 +0300",
"m... |
[
{
"msg_contents": "Hello,\n\n\nWorking in a DSS environment, we often need to truncate table partitions regarding a WHERE condition and have to:\n\n- query the dictionnary to identify thoses partitions,\n\n- build SQL statements,\n\n- truncate all partitions covered by the WHERE condition\n\n- eventually delete... |
[
{
"msg_contents": "Just saw another report of what's probably systemd killing off Postgres'\nSysV semaphores, as we've discussed previously at, eg,\nhttps://www.postgresql.org/message-id/flat/57828C31.5060409%40gmail.com\nSince the systemd people are generally impervious to suggestions that\nthey might be mista... |
[
{
"msg_contents": "Hi,\n\nthe following query crashes master as of 4212cb7.\n\nselect ts_rewrite(\n tsquery_phrase(\n tsquery $$'sanct' & 'peter'$$,\n tsquery $$'5' <-> '6'$$,\n\t 42),\n tsquery $$'5' <-> '6'$$,\n plainto_tsquery('I') );\n\nBacktrace below.... |
[
{
"msg_contents": "Hi All,\n\nI have noticed that any SQL query executed immediately after attaching\nto a particular debugging target (pldebugger) either takes longer time\nfor execution or it hangs on Windows. I have checked it on\nPostgreSQL-9.5 and PostgreSQL-9.6 versions and have found that the\nissue only... |
[
{
"msg_contents": "Hello.\n\nI noticed, that `make installcheck` fails on my laptop with following\nerrors:\n\nhttp://afiskon.ru/s/98/6f94ce2cfa_regression.out.txt\nhttp://afiskon.ru/s/b3/d0da05597e_regression.diffs.txt\n\nMy first idea was to use `git bisect`. It turned out that this issue\nreproduces on commi... |
[
{
"msg_contents": "This is a design proposal for matrix data type which can be larger\nthan 1GB. Not only a new data type support, it also needs a platform\nenhancement because existing varlena has a hard limit (1GB).\nWe had a discussion about this topic on the developer unconference\nat Tokyo/Akihabara, the d... |
[
{
"msg_contents": "Implement table partitioning.\n\nTable partitioning is like table inheritance and reuses much of the\nexisting infrastructure, but there are some important differences.\nThe parent is called a partitioned table and is always empty; it may\nnot have indexes or non-inherited constraints, since ... |
[
{
"msg_contents": "Hi!\n\n(Copying this one to -hackers as well, for git coverage)\n\nWe were just notified by one of our hosting providers, MeetMe, that they\nare about to start physically moving two of our servers to a different\ndatacenter. Unfortunately, we were only given about an hours warning on\nthis, s... |
[
{
"msg_contents": "Dear PostgreSQL Hackers,\n\nI am working in PostgreSQL 9.4.* optimizer module. In costsize.c file and\nfinal_cost_hashjoin() function, the innerbucketsize is either:\n\na) calculated using a cached copy\n OR\nb) calculated afresh using sta... |
[
{
"msg_contents": "Hi!\n\nWe are now accepting talk proposals for PgConf.Russia 2017.\n\nThe Russian PostgreSQL conference will take place on 15-17 March 2017 in\nMoscow, at the Digital October venue, same as for PgConf.Russia 2015. The\naudience include a wide range of application and system developers,\ndatab... |
[
{
"msg_contents": "Hi.\n\nI noticed that there is a lot of repeating code like this:\n\n```\nif (strspn(str, \" \\t\\n\\r\\f\") == strlen(str))\n```\n\nI personally don't find it particularly readable, not mentioning that\ntraversing a string twice doesn't look as a good idea (you can check\nusing objdump that ... |
[
{
"msg_contents": "I have a setup where a 9.6.1 server uses postgres_fdw to connect to a 9.4.9\nhot standby server.\n\nI have a DML statement which triggers the error:\n\nERROR: XX000: outer pathkeys do not match mergeclauses\nLOCATION: create_mergejoin_plan, createplan.c:3722\n\nThe error first starts appear... |
[
{
"msg_contents": "Hi,\n\nI'm wondering if it's not time for $subject:\n- V0 causes confusion / weird crashes when PG_FUNCTION_INFO_V1 was\n forgotten\n- They have us keep weird hacks around just for the sake of testing V0\n- they actually cost performance, because we have to zero initialize Datums, even if\n ... |
[
{
"msg_contents": "Hello hackers,\n\nI met SIGSEGV when using `array_position()` with record type\narguments, so I've written a patch which corrects this problem. It\nseems that `array_position...()` sets wrong memory context for the\ncached function (in this case `record_eq()`) which is used to find a\nmatchin... |
[
{
"msg_contents": "There is this problem with the jsonb operators \"? text\" \"?| text[]\"\nand \"?& text[]\" that the question mark is typically used for prepared\nstatement parameters in the most used abstraction APIs in Java and\nPHP.\n\nThis really needs an alternative. Something like \"HAS text\", \"HAS\nA... |
[
{
"msg_contents": "Looking at src/port/pg_strong_random.c this would be a bug in autoconf\n\n#else\n /* The autoconf script should not have allowed this */\n#error no source of random numbers configured\n\nMy configure line is:\n\n./configure --prefix=/usr/local/pgsql/10 --enable-debug --with-python\n--... |
[
{
"msg_contents": "The patch attached (a one-liner) includes information about how to\nreproduce the issue and why it happens.\n\nWe applied it to our production servers (9.5) and has been working as\nexpected for a while. I've also checked that it can be applied cleanly to\ncurrent master.\n\nCould it make its... |
[
{
"msg_contents": "Hi\n\nLong time I am pushing a COPY RAW - without success.\n\nNow I propose functionally similar solution - reduced to only to psql\nconsole\n\nNow we have a statement \\g for execution query, \\gset for exec and store\nresult in memory and I propose \\gstore for storing result in file and\n\... |
[
{
"msg_contents": "Since:\n\ncommit f0e44751d7175fa3394da2c8f85e3ceb3cdbfe63\nAuthor: Robert Haas <rhaas@postgresql.org>\nDate: Wed Dec 7 13:17:43 2016 -0500\n\n Implement table partitioning.\n\nIf I use psql compiled from 10devel to connect to a 9.6.1 server, then \\d\nfails:\n\npsql (10devel-f0e4475, ser... |
[
{
"msg_contents": "Hi. This message is going to xen-devel (because that's where the\nosstest project is) and to pgsql-hackers (because I hope they may be\nable to advise about the scope of the PostgreSQL SERIALIZABLE\nconstraint problem).\n\nIn summary: PostgreSQL only provides transaction serialisability for\... |
[
{
"msg_contents": "I would like to propose a patch for parallelizing merge join path.\nThis idea is derived by analyzing TPCH results.\n\nI have done this analysis along with my colleagues Rafia sabih and Amit kaplia.\n\nCurrently we already have infrastructure for executing parallel join,\nso we don't need any... |
[
{
"msg_contents": "Hi,\n\nI recently found couple of issues with the way initial logical decoding\nsnapshot is made.\n\nFirst one is outright bug, which has to do with how we track running\ntransactions. What snapbuild basically does while doing initial snapshot\nis read the xl_running_xacts record, store the l... |
[
{
"msg_contents": "\njascana (mingw, 64 bit compiler, no openssl) is currently hung on \"make \ncheck\". After starting the autovacuum launcher there are 120 messages on \nits log about \"Could not acquire random number\". Then nothing.\n\n\nSo I suspect the problem here is commit \nfe0a0b5993dfe24e4b3bcf52fa64... |
[
{
"msg_contents": "Hi,\n\nI found a minor typo at\nhttps://www.postgresql.org/docs/9.6/static/catalog-pg-am.html.\n\npg_catalog.pg_am. amhandler is of type \"oid\" according to the\ndocumentation, but it's actually of type \"regproc\" in reality.\n\nCompare with e.g. pg_aggregate where the columns of type \"reg... |
[
{
"msg_contents": "Hi,\n\nI noticed that during an index-only scan the filter check is executed\nafter the visibility of a tuple is confirmed. This means a lot of\nunnecessary heap fetches, if the filter condition is highly selective.\n\nIn the application I'm maintaining, the visibility map is mostly\ndirty. I... |
[
{
"msg_contents": "Hello all,\nThis is to propose a patch for enabling parallel index-only scans. With the\npatch of parallel index scan around [1], adding the mechanism for\nparallelising index-only scans makes sense. Without this mechanism for the\nqueries preferring index-only scans, it is likely that at hig... |
[
{
"msg_contents": "Hello.\n\nCurrently doc/src/sgml/datatype.sgml states:\n\n```\n When <type>timestamp</> values are stored as eight-byte integers\n (currently the default), microsecond precision is available over \n the full range of values. When <type>timestamp</> values are\n stored as double pr... |
[
{
"msg_contents": "Add support for temporary replication slots\n\nThis allows creating temporary replication slots that are removed\nautomatically at the end of the session or on error.\n\nFrom: Petr Jelinek <petr.jelinek@2ndquadrant.com>\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/... |
[
{
"msg_contents": "On Tue, Oct 4, 2016 at 11:59 AM, Robert Haas <robertmhaas@gmail.com> wrote:\n> On Mon, Oct 3, 2016 at 8:43 PM, Michael Paquier\n> <michael.paquier@gmail.com> wrote:\n>> The rest looks good to me. Thanks for the feedback and the time!\n>\n> Thanks for the fixes. I committed this with an addit... |
[
{
"msg_contents": "Last week I noticed that the Wait Event/Locks system doesn't correctly\ndescribe waits for tuple locks because in some cases that happens in\ntwo stages.\n\nNow I notice that the Wait Event system doesn't handle waiting for\nrecovery conflicts at all, though it does access ProcArrayLock\nmult... |
[
{
"msg_contents": "Hi.\n\nThe first attached patch implements recursive processing of nested \nobjects and arrays in json[b]_populate_record[set](), \njson[b]_to_record[set](). See regression tests for examples.\n\nIt also fixes the following errors/inconsistencies caused by lost \nquoting of string json values... |
[
{
"msg_contents": "The attached patch addresses the complaint raised in\nhttps://www.postgresql.org/message-id/CAHt_Luuao4gd6De61GryK=2ff-MTgHzjqffdjz02uSdVqYmKKQ@mail.gmail.com\n\nnamely, that if you have an index on, say, integer columns i and j,\nthen the planner will figure out that it can use an indexscan ... |
[
{
"msg_contents": "Hi,\n\nI propose to append an error message when index name and table name are \nsame.\n\n\nexample:\n\npostgres@postgres=# create table t (a int not null, constraint t primary \nkey (a));\nERROR: relation \"t\" already exists\n\n\nEnd users will confusing pretty, because if users meet this... |
[
{
"msg_contents": "Hi\n\nI don't remember well, there was maybe similar ToDo.\n\nYesterday I got a incident on high load system when I executed DROP INDEX\ncmd.\n\nThis statement waited on a finish of long transaction, but it stopped any\nother statements.\n\nCan be nice when waiting on lock statement doesn't b... |
[
{
"msg_contents": "Hi Hackers,\n\nAttached is a new builtin function that exposes the CATALOG_VERSION_NO \nconstant under the pg_catversion() function, e.g.\n\ntest=# SELECT pg_catversion();\n pg_catversion\n---------------\n 201612121\n(1 row)\n\nAlthough it mostly useful during the development cycle to ... |
[
{
"msg_contents": "Based on discussion in\nhttps://www.postgresql.org/message-id/CAE1wr-w%3DLE1cK5uG_rmAh-VBxc4_Bnw-gAE3qSqL-%3DtWwvLvjQ%40mail.gmail.com\n.\n\nTested via regression tests.\nTo be applied in master only and to be included in 10.0.\n\n\n\n-- \nVladimir Rusinov\nStorage SRE, Google Ireland\n\nGoog... |
[
{
"msg_contents": "Hi hackers,\n\nI have a workload using SSI that causes a lot of tuple predicate to be\npromoted to page locks. This causes a lot of spurious serialisability\nerrors, since the promotion happens at once three tuples on a page are\nlocked, and the affected tables have 30-90 tuples per page.\n\... |
[
{
"msg_contents": "They are considered bad practice in many style guides and many editors\nconfigured to stip them on every save.\n\nSuch editors will produce spurious diffs when editing the documentation.\n\nTherefore, I propose this patch.\n\n-- \nVladimir Rusinov\nStorage SRE, Google Ireland\n\nGoogle Irelan... |
[
{
"msg_contents": "I've been looking into the problem reported here:\nhttps://www.postgresql.org/message-id/CAOfJSTwzQ7Fx6Yjeg9mFkMsM5OVKPoa=EgkHceGdkr1TWg8vkA@mail.gmail.com\n\nI can reproduce similar misbehavior with this test case:\n\ncreate table updates as\nselect (-log(random()) * 10)::int as driver_id,\n... |
[
{
"msg_contents": "Hi all\n\nI recently had a real world case of a user confused by the\n(non)relationship between txid_current()'s output and that of the xid\ntype, specifically pg_stat_replication.backend_xmin .\n\nI quote:\n\n\"\n> What should we look for to determine normal? I thought maybe it would\n> comp... |
[
{
"msg_contents": "Folks,\n\nI'm having some trouble understanding what's going on here. When I \\i\nthe file in 55caaaeba877eac1feb6481fb413fa04ae9046ac without starting\na transaction explicitly, it produces the expected results. When I \\i\nit after a BEGIN, not so much.\n\nWhat's going on?\n\nBest,\nDavid... |
[
{
"msg_contents": "Hi\n\nI am sending set of patches - for simpler testing these patches are\nindependent in this moment.\n\nThese patches are replacement of my previous patches in this area: COPY RAW\nand fileref variables.\n\n1. parametrized queries support - the psql variables can be passed as query\nparamet... |
[
{
"msg_contents": "I had a system where the recovery.conf file was renamed \"out of the way\" at\nsome point, and then the system was promoted. This is obviously operator\nerror, but it seems like something we should handle.\n\nWhat happens now is that the non-existance of recovery.conf is a FATAL\nerror. I won... |
[
{
"msg_contents": "I've started work on a patch to make pg_basebackup use the temporary slots\nfeature that has been committed (thanks Petr!!). The reason for this is to\navoid the cases where a burst of traffic on the master during the backup\ncan cause the receive log part of the basebackup to fall far enoug... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.