threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "for anyone interested....\n\nSolaris 2.5.1 with --enable-thread-safety\n\nconfigure:25848: gcc -o conftest -O2 -Wall -Wmissing-prototypes \n-Wpointer-arith -Wdeclaration-after-statement -Wendif-labels \n-fno-strict-aliasing -fwrapv -pthreads -pthreads -D_REENTRANT \n-D_THREAD_SAFE -D_POSI... |
[
{
"msg_contents": "Sorry about the HTML post, but I wanted to include the error form.\n\n \n\nWhen trying to perform an upgrade with PostgreSQL 8.3.5 over 8.3.4, the\ninstaller insists that PostgreSQL has failed to start.\n\n \n\nHowever, I can see the processes in memory (it has indeed started) and\neven attac... |
[
{
"msg_contents": "Hi,\n\nI just released new RPM sets, which is based on Nov 18 2008 11:00 PM\nEEST CVS snapshot.\n\nThese packages *do* require a dump/reload, even from previous 8.4\npackages, because of a catversion update.\n\nWe have more than 400 testers using these sets.\n\nAs usual, please find detailed ... |
[
{
"msg_contents": "This patch adds the ability to retrieve async notifications using dblink,\nvia the addition of the function dblink_get_notify.\n\nIt is written against cvs head, includes documentation and regression\ntesting. It compiles, tests and works well.\n\nI would be interested in some feedback on the... |
[
{
"msg_contents": "Hello\n\nI migrated database cluster from 8.1 to 8.3, and I had problem with\ntsvector domain. I imported tsearch2 module into template1 and then I\nstarted import of complete dump - but public.tsvectors wasn't be known\n- and I got message\n\nERROR: type \"tsvector\" is only a shell\nLINE 5... |
[
{
"msg_contents": "The patch mentioned in the subject leaves this compiler warning:\n\ntsvector_op.c: In function ‘insertStatEntry’:\ntsvector_op.c:815: warning: ‘res’ may be used uninitialized in this function\n\nI took a quick look but it is not immediately obvious whether the\ncompiler might actually have a ... |
[
{
"msg_contents": "Log Message:\n-----------\nRethink the way FSM truncation works. Instead of WAL-logging FSM\ntruncations in FSM code, call FreeSpaceMapTruncateRel from smgr_redo. To\nmake that cleaner from modularity point of view, move the WAL-logging one\nlevel up to RelationTruncate, and move RelationTrun... |
[
{
"msg_contents": "Simple select give wrong result when it uses the Bitmap Heap Scan path.\n\n\npostgres=# CREATE OR REPLACE FUNCTION my_exec_im_test_func(i integer)\nRETURNS integer AS $$\n BEGIN\n RETURN i + 1;\n END;\n$$ LANGUAGE plpgsql;\nCREATE FUNCTION\n\npostgres=# set enable... |
[
{
"msg_contents": "\nI haven't looked into what's causing it yet but...\n\npostgres=# SELECT res, * FROM (\n SELECT 'foo'||i AS test, i AS res \n FROM (VALUES (1)) AS x(i) \n UNION ALL \n SELECT 'foo'||i, i \n FROM (VALUES (2)) AS x(i)\n ) AS x;\nserver closed the c... |
[
{
"msg_contents": "Hi there,\n\nwe got no comments on our proposal about changing GIN interface to \nsupport multicolumn feature of GIN, see \nhttp://archives.postgresql.org/message-id/490B3752.3010800@sigaev.ru\nfor details, so here are some numbers demonstrating benefits and some\nissues. Any help and feedbac... |
[
{
"msg_contents": "Folks,\n\nSince it's too late to look at this for 8.4, can the following go on the \nTODO list?\n\nReferential Integrity\n\n[] Allow creation of FKs targeting unique expression indexes on the \nreferenced table. Syntax: REFERENCES <reftable> ( ( column expression ) )\n\nReason: current FK ru... |
[
{
"msg_contents": "We are now in discussion about toast table upgrading. I try to collect ideas and\nfigure out how it should work and where are problems.\n\nOverview:\n---------\n\nA few weeks ago we made a decision to use convert on read. We already made a \ndecision how to solve problem with overflow data af... |
[
{
"msg_contents": "\nSo based on Graeme Job's T-SQL hack over at thedailywtf.com I adapted the\nT-SQL code to Postgres and got this. Thought others might find it amusing.\n\nWITH RECURSIVE Z(IX, IY, CX, CY, X, Y, I) AS (\n SELECT IX, IY, X::float, Y::float, X::float, Y::float, 0\n ... |
[
{
"msg_contents": "I wrote:\n> All,\n>\n> This is my first patch review for PostgreSQL. I did submit a patch last\n> commit fest (Boyer-Moore) so I feel I should review one this commit fest.\n> I'm quite new to PostgreSQL so please don't rely on me totally. I'll do my\n> best. Heikki is also reviewing this patc... |
[
{
"msg_contents": "Hello,\n\nHEAD is failed to be built on win32 mingw.\nIt requires manual 'mkdir man7' or so.\n\nAre there any changes in build process?\nOr am I missing something required to build?\n\n----\n$ make\n/bin/sh.exe: man7/.timestamp: No such file or directory\n\n$ mkdir doc/man7 <- HERE... |
[
{
"msg_contents": "Hi!\n\nI want to use the fnmatch() function in libpq, to support wildcard\ncertificate matching. This function is part of the standard\n(http://www.opengroup.org/onlinepubs/000095399/functions/fnmatch.html),\nbut obviously not implemented on all platforms (naturally, it's missing\non win32, b... |
[
{
"msg_contents": "Hi,\n\nIn the current Synch Rep patch, the standby cannot catch up with the\nprimary which has a bigger timeline. So, whenever making the standby\ncatch up, a fresh base backup is required. This is obviously undesirable,\nand I'd like to get rid of this restriction.\n\nPostgres itself can rec... |
[
{
"msg_contents": "Log Message:\n-----------\nTABLE command\n\nModified Files:\n--------------\n pgsql/doc/src/sgml/ref:\n create_table_as.sgml (r1.39 -> r1.40)\n (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/doc/src/sgml/ref/create_table_as.sgml?r1=1.39&r2=1.40)\n select.sgml (r1.109 ... |
[
{
"msg_contents": "I get the following warnings in HEAD\n\npl_exec.c: In function ‘exec_stmt_raise’:\npl_exec.c:2538: warning: format not a string literal and no format arguments\npl_exec.c:2538: warning: format not a string literal and no format arguments\n\n\nAttached patch seems right to me - objections?\n\n... |
[
{
"msg_contents": "Log Message:\n-----------\nSilence compiler warning about ignored return value. Our comment already\nclearly stated that we are aware that we're ignoring it.\n\nModified Files:\n--------------\n pgsql/src/test/regress:\n pg_regress.c (r1.49 -> r1.50)\n (http://anoncvs.postgre... |
[
{
"msg_contents": "Okay, I have had a chance to run some timing benchmarks.\nHere are my results for the parallel pg_restore patch:\n\nKen\n--------------------------------------------------\nServer settings:\n\n max_connections = 100 # (change requires restart)\n shared_buffers = 256MB ... |
[
{
"msg_contents": "\nSounds somewhat evil, I know, but I was wondering if it was even\nremotely possible with the current design?\n\nThe reason: we are contemplating using pg_standy to create a\nwarm-standby. It would be a bonus if we would run read-only queries\nagainst this DB to take some of the load off or ... |
[
{
"msg_contents": "I think we should avoid the #define's like below which uses a local\nvariable. I guess the same #define is used elsewhere in the code as well. If\nthe code is rearranged or the variable name is changed, the code would\nbreak.\n\nThe use of malloc should also be avoided (unless the memory subs... |
[
{
"msg_contents": "The attached patch changes pg_settings.enumval to be an array of text\ninstead of just a string, per previous discussion and the open items list.\n\nComments?\n\n//Magnus",
"msg_date": "Fri, 21 Nov 2008 13:21:03 +0100",
"msg_from": "Magnus Hagander <magnus@hagander.net>",
"msg_fro... |
[
{
"msg_contents": "I was writing a simple standalone function compiled with pgxs - been a\nwhile since I did that.\n\nIncluding \"postgres.h\" and \"fmgr.h\", it didn't work - the backend would\ncomplain that the function did not exist in the shared library.\nIncluding \"postgres.h\" and \"funcapi.h\" *did* wor... |
[
{
"msg_contents": "Buildfarm member pika showed an interesting transient failure yesterday:\nhttp://www.pgbuildfarm.org/cgi-bin/show_log.pl?nm=pika&dt=2008-11-21%2012:13:06\n\n./pg_regress --inputdir=. --dlpath=. --multibyte=SQL_ASCII --load-language=plpgsql --no-locale --temp-install=./tmp_check --top-builddi... |
[
{
"msg_contents": "Hi,\n\nI wonder how portable designated initializers are. As far as I can tell\nthey were only defined in C99. Can we use them in our source? If not,\nis there a way to do this in C89?\n\nI mean something like this:\n\ntypedef struct foo {\n char type;\n union {\n ... |
[
{
"msg_contents": "Hello,\nsearched documentation, FAQ and mailing list archives\n(mailing list archive search is volumous :-) )\n\nbut could not find an answer:\n\nI would like to be able to update\nseveral rows to different values at the same time\n\nIn oracle this used to be called Array update or \n'collect... |
[
{
"msg_contents": "Hi,\n\nIs $SUBJECT possible? If not, do you think it would be worth\nimplementing this?\n\nIf this parameter is not used appropriately, it may fill up logs --\nthat's why I am asking this question.\n\nRegards,\n-- \nDevrim GÜNDÜZ, RHCE\ndevrim~gunduz.org, devrim~PostgreSQL.org, devrim.gunduz~... |
[
{
"msg_contents": "Following test case gives a warning of snapshot not destroyed at commit\ntime.\n\nCREATE TABLE test (a int);\nINSERT INTO test VALUES (1);\nBEGIN;\nDECLARE c CURSOR FOR SELECT * FROM test FOR update;\nSAVEPOINT A;\nFETCH -2 FROM c;\nROLLBACK TO SAVEPOINT A;\nCOMMIT;\n\nShould we call FreeQuer... |
[
{
"msg_contents": "Hi,\n\nrecently, I wrote a really complex SELECT statement which consists of about \n20 relations using NATURAL JOIN method and every single relation consists 50 \ncolumns. It looks like:\nPREPARE ugly_stmt AS\nSELECT * FROM t1 NATURAL JOIN t2 NATURAL JOIN t3 ... NATURAL JOIN t20 WHERE \nid =... |
[
{
"msg_contents": "Log Message:\n-----------\nAdd support for matching wildcard server certificates to the new SSL code.\n\nThis uses the function fnmatch() which is not available on all platforms\n(notably Windows), so import the implementation from NetBSD into src/port.\n\nModified Files:\n--------------\n ... |
[
{
"msg_contents": "Hi Stehule,\n\nI have looked at the patch and it looks great. Here are my observation\n\nCompilation\n----------------\n1 - Patch applied successfully on CVS-HEAD\n2 - No compilation error found\n\nCode\n-------\n1 - Style of code is very close to the existing PG code.\n2 - Comments look OK t... |
[
{
"msg_contents": "I hope this is the right place to report a bug/issue in the official\ndocumentation...\n\nIn\nhttp://www.postgresql.org/docs/current/interactive/plpgsql-control-structures.html\nthere is an example for a function, cs_refresh_mviews(), chapter 38.6.4.\nWithin this function there are a \"PERFOR... |
[
{
"msg_contents": "It was brought to my attention that DISCARD ALL\ndoes not release advisory locks:\n\n http://pgfoundry.org/tracker/index.php?func=detail&aid=1010499&group_id=1000258&atid=983\n\nThus connection managers / poolers need to do that manually.\n\nThis obviously means DISCARD ALL needs fixing. Ap... |
[
{
"msg_contents": "While hacking on parser/gram.y just now I noticed in passing that the \nautomatically generated ecpg parser had 402 shift/reduce conflicts. \n(Don't panic, the parser in CVS is fine.) If you don't pay very close \nattention, it is easy to miss this. Considering also that we frequently \nhav... |
[
{
"msg_contents": "Exporting PGINTERVALSTYLE causes errors of the following kind when \nconnecting to an older server version:\n\npsql: FATAL: unrecognized configuration parameter \"intervalstyle\"\n\nShould the processing of this variable be made conditional upon the \nserver version?\n\n(What is the use case... |
[
{
"msg_contents": "Hi,\n\nIn current pg_standby, the presence of the trigger file causes\nrecovery to end whether or not the next WAL file is available.\nThereby, some transactions in the available WAL files will be\nlost. So, we cannot use this trigger file to promote the standby\nto the primary.\n\nI'd like t... |
[
{
"msg_contents": "I've been chewing on the problem described here:\nhttp://archives.postgresql.org/pgsql-general/2008-11/msg01220.php\n\nIt's not particularly easy to fix without making annoyingly large\nchanges to the API for PQExpBuffer. The best idea I have come up\nwith so far goes like this:\n\n* Upon fa... |
[
{
"msg_contents": "(I'm still downloading Fedora-10)\n\nEven though Tom is subscribed to the list below, I thought I should\nforward this e-mail to this list:\n\n-------- Forwarded Message --------\n> From: Itamar - IspBrasil <itamar@ispbrasil.com.br>\n> Reply-To: Development discussions related to Fedora\n> <f... |
[
{
"msg_contents": "Dear PostgreSQL developers,\n\n#### 1\n\nI stumbled upon an obscure bug (or undesirable feature:-) in the schema \nmetadata accessible through the information_schema, and possibly \npg_catalog as well. As it was mixed in a bug in some in my code, it was \nhard for me to identify it.\n\nThe is... |
[
{
"msg_contents": "Hi,\n\nI'm very new to hacking postgresql but am using on a very big site (\nhttp://ojp.nationalrail.co.uk). One of the issues that we have is moving\ndata from a live database to a reports one. I've hacked an extra option to\npg_dump to delete from tables rather than dropping them.\n\nOnce I... |
[
{
"msg_contents": "It appears that the mingw faction of the buildfarm doesn't like this\npatch:\nhttp://archives.postgresql.org/pgsql-committers/2008-11/msg00289.php\n\nThe failures look like the regression tests are being run with\nintervalstyle = iso (the default) rather than postgres_verbose\nas they ought t... |
[
{
"msg_contents": "Today I noticed I cannot login to cvs.pgfoundry.org anymore since the\nIP address has been changed am asked password which seems to be\nchanged. So I cannot use CVS any more. Does anybody why this happens\nand how to fix it?\n--\nTatsuo Ishii\nSRA OSS, Inc. Japan\n",
"msg_date": "Wed, 26 ... |
[
{
"msg_contents": "Hello\n\nI added two new columns to pg_proc. I have a problem because access\nvia SearchSysCache doesn't work\n\n /* Search syscache by name only */\n catlist = SearchSysCacheList(PROCNAMEARGSNSP, 1,\n\nCStringGetDatum(funcname),\n ... |
[
{
"msg_contents": "i get lost with this one... i thought there were two patches that get\nmerged into one, but i don't find nor the merged version nor the\nactualized version of any of one...\n\n-- \nAtentamente,\nJaime Casanova\nSoporte y capacitación de PostgreSQL\nAsesoría y desarrollo de sistemas\nGuayaquil... |
[
{
"msg_contents": "-----BEGIN PGP SIGNED MESSAGE-----\nHash: SHA1\n\n\n\n- --On Wednesday, November 26, 2008 14:00:59 -0800 \"Joshua D. Drake\" \n<jd@commandprompt.com> wrote:\n\n\n> Since were chatting :P. My vote would be to move everything back to port\n> 22 and force key based auth only.\n\nHow does that wo... |
[
{
"msg_contents": "This patch:\nhttp://archives.postgresql.org/pgsql-committers/2008-02/msg00356.php\nwhich was alleged to prevent memory leaks, seems to have introduced\nsome. The problem is that shutdown_MultiFuncCall() now performs a\n\tMemoryContextSwitchTo(flinfo->fn_mcxt);\nwhich means that an exiting SR... |
[
{
"msg_contents": "Can someone remind me why we have --enable-thread-safety? As opposed \nto it being default and having --disable-thread-safety.\n\n/Magnus\n\n",
"msg_date": "Thu, 27 Nov 2008 07:59:39 +0100",
"msg_from": "Magnus Hagander <magnus@hagander.net>",
"msg_from_op": true,
"msg_subjec... |
[
{
"msg_contents": "Hi all:\n\nI begin an \"pg external table\" in pgfoundry\n(http://pgfoundry.org/projects/pgexternaltable/ ), you can find the doc\n(brief and detail design)from that URL.\n\nI am newbie to postgresql, and this code still have many defects and low\nefficiency. I am very happy to get advice fr... |
[
{
"msg_contents": "test\n\n \n\nmaosen\n\n\n\n\n\n\n\n\n\n\n \ntest\n \nmaosen",
"msg_date": "Thu, 27 Nov 2008 16:00:13 +0800",
"msg_from": "\"maosen.zhang\" <maosen.zhang@alibaba-inc.com>",
"msg_from_op": true,
"msg_subject": "test"
}
] |
[
{
"msg_contents": "Please find enclosed a revised version of pg_readahead and a patch to\ninvoke pg_readahead.\nChanges from the previous one is as follows:\n\nPg_readahead now does not return any prefetched point. It simply\nprefetches all the datapages\nrefered from WAL records in a given WAL segment, except... |
[
{
"msg_contents": "Query is:\n\n\"explain analyze select sum(A), count(*) from tab1 left outer join \ntab_outer on id=out_id where id=10;\"\n\noutput:\n\n Aggregate (cost=31.91..31.92 rows=1 width=4) (actual \ntime=14.185..14.185 rows=1 loops=1)\n -> Nested Loop Left Join (cost=0.00..17.11 rows=2959 wid... |
[
{
"msg_contents": "There was a small bug in the new compute_tsvector_stats function in CVS \nHEAD, causing a crash with this surprisingly simple test case:\n\npostgres=# CREATE TABLE tstest (ts tsvector);\nCREATE TABLE\npostgres=# INSERT INTO tstest values ('foobar');\nINSERT 0 1\npostgres=# ANALYZE tstest;\nse... |
[
{
"msg_contents": "Hello,\n\nI chaged the patch accoding to the comment against v3 and fixed\nsome bugs. Since I failed to post the patch because of the excess\nof mail size, I would attach it to Wiki from this time.\nhttp://wiki.postgresql.org/wiki/NTT%27s_Development_Projects#Patch_set\n\nList of updates\n---... |
[
{
"msg_contents": "Hi,\n\nThe immediate shutdown (pg_ctl -m i stop) might not be able to\nkill the startup process during archive recovery. It's because\nthe startup process calls system() which ignores SIGQUIT for\nexecuting the restore_command. So, only the startup process\nmight survive the immediate shutdow... |
[
{
"msg_contents": "Following test end up with the server crash into 8.4 cvs Head.\n\nuname -a\nLinux localhost.localdomain 2.6.18-53.el5 #1 SMP Wed Oct 10 16:34:19 EDT\n2007 x86_64 x86_64 x86_64 GNU/Linux\n\nTestcase with ISN contrib module:\n=========================\n\nCREATE OR REPLACE function isbn_issn_pro... |
[
{
"msg_contents": "The problem reported by Rushabh Lathia boils down to the fact that\ncontrib/isn intends to define a datatype that is represented \"just like\nint8\", but it supposes that int8 must be pass by reference. There is\nnot anything wrong with the C code; the problem is the CREATE TYPE\ncommand in ... |
[
{
"msg_contents": "I'm trying to actually run the experiments Tom suggested running EXPLAIN on\nthe DBT3 DSS queries for various default_stats_target sizes. It's a bit of a\nheadache because DBT3 seems to be unmaintained these days so lots of things\ndon't really work.\n\nI ran into an error I've never seen bef... |
[
{
"msg_contents": "\n-----BEGIN PGP SIGNED MESSAGE-----\nHash: RIPEMD160\nNotDashEscaped: You need GnuPG to verify this message\n\n\nLooks like inheritance causes a statement-level trigger to fire on\nthe last evaluated table in the inheritance chain. Is this the\ndesired behavior? If so, is there any way to pr... |
[
{
"msg_contents": "\nThe following bug has been logged online:\n\nBug reference: 4553\nLogged by: Andrew Gierth\nEmail address: andrew@tao11.riddles.org.uk\nPostgreSQL version: 8.3-8.4\nOperating system: all\nDescription: HOLD cursors not materializing results fully\nDetails: \n\n(te... |
[
{
"msg_contents": "Log Message:\n-----------\nRemove inappropriate memory context switch in shutdown_MultiFuncCall().\nThis was a thinko introduced in a patch from last February; it results\nin memory leakage if an SRF is shut down before the actual end of query,\nbecause subsequent code will be running in a lo... |
[
{
"msg_contents": "I wrote:\n> I was also reading over the standard tonight. I've discovered that the\n> OFFSET in LEAD() and LAG() is optional. It should default to 1 if it is\n> not present. Oracle seems to support this.\n> \n> SQL2008 says:\n> > If <lead or lag function> is specified, then:\n> > i) Let VE1 b... |
[
{
"msg_contents": "While fooling with the recent contrib problem\nhttp://archives.postgresql.org/pgsql-committers/2008-11/msg00346.php\nI noticed another way that tablefunc.c is being cavalier about the\nstate when it returns control: the crosstab() function returns with\nan internal SPI call still active! Thi... |
[
{
"msg_contents": "Could anyone with a MinGW system please run the ecpg regression suite including\ntcp checks for the current CVS HEAD for me?\n\nJust run \"make checktcp\" in src/interfaces/ecpg and afterwards send me the file\n.../ecpg/test/results/connect-test1.stderr. There is a special expected file\nfor ... |
[
{
"msg_contents": "\nBreaking down of patch into sections works very well for review. Should\nallow us to get different reviewers on different parts of the code -\nreview wranglers please take note: Dave, Josh.\n\nCan you confirm that all the docs on the Wiki page are up to date? There\nare a few minor discrepa... |
[
{
"msg_contents": "I like strict in general, but this doesn't seem logical:\n\npostgres=# SELECT to_timestamp('29-12-2005 01:2:03', 'DD-MM-YYYY \nHH24:MI:SS'); -- works\n to_timestamp\n------------------------\n 2005-12-29 01:02:03+02\n(1 row)\n\npostgres=# SELECT to_timestamp('29-12-2005 01:02:3', 'DD-M... |
[
{
"msg_contents": "Can anyone please help in getting a solution...\nI have two table, 1 contains all details of a person with person id as\nprimary key.\nthe other table has person id and parent id both referring to person table\nperson id.\nMy query is i have to get all the parents of a given person name.\nPle... |
[
{
"msg_contents": "Is it possible to reproduce on your system?\n\nI got the following result at the latest CVS HEAD independently from\nthe feature of SE-PostgreSQL. I could not get the result a week ago,\nbut could not find where cause the regression.\n\n-- (1) make a dummy file.\n[kaigai@saba base]$ dd if=/de... |
[
{
"msg_contents": "Here is an update version of contrib/pg_stat_statements.\n\nNew modifications from the last version are:\n1. New counters in struct Instrumentation.\n2. EXPLAIN ANALYZE VERBOSE shows buffer statistics in 'actual' section.\n3. Buffer statistics counsters are not reset to zero anymore.\n\n----\... |
[
{
"msg_contents": "While playing around with the \"GIN fast insert\" patch, I was puzzled why \nbuilding a GIN index seemed to take so long. The test case I use was simply:\n\nCREATE TABLE foo (bar tsvector);\nINSERT INTO foo SELECT to_tsvector('foo' || a) FROM generate_series(1, \n200000) a;\nCREATE INDEX foog... |
[
{
"msg_contents": "\nHaving just tried to restore a 64 bit BSD database to a 32 bit linux\nmachine (using PiTR), I have now realized the (with hindsight, obvious)\nerror in my ways.\n\nNow, I need to plan a way forward. From reading of other peoples similar\nproblems, I now realize that I need a system with ide... |
[
{
"msg_contents": "The skyline operator is a non-standard extension (even SQL:2008 does\nnot include it). Queries of that type can be expressed in standard\nSQL (although a bit clumsy and the execution is slow). Others have\nalso worked on extending PostgreSQL with the skyline operator, see:\n\nhttp://archive... |
[
{
"msg_contents": "I've noticed that pg_stat_all_tables returns NULL for idx_scan and\nidx_tup_fetch if there are no indexes present on a table.\n\nIs this actually intended, or is that something that should be fixed?\n\n//Magnus\n",
"msg_date": "Tue, 02 Dec 2008 14:38:34 +0100",
"msg_from": "Magnus Hag... |
[
{
"msg_contents": "Would it make sense to be able to configure maintenance_work_mem\nspecifically for the autovacuum processes? Given that there can be a\nnumber of them, it might be good to be able to have one default for all\n*other* processes, and a separate one from the ones kicked off by autovac?\n\n//Magn... |
[
{
"msg_contents": "Hi all,\n\ncvs head configured without --enable-debug hang in initdb while making \ncheck.\n\nwarthog doesn't exhibit it because it's configured with debug.\n\nwhen it hangs, postmaster takes 100% cpu doing nothing. initdb waits for \nit while creating template db.\n\nAccording to truss, the ... |
[
{
"msg_contents": "Since this whole in-place upgrade thing is going nowhere until there's \nalso a good solution for ye olde \"The database cluster was initialized \nwith CATALOG_VERSION_NO ...\" error, I spent some time today touring \nthrough what everybody else has done there and have some initial review \nc... |
[
{
"msg_contents": "The following test flashes snapshot leak warning and subsequently dumps\ncore. Though this looks very similar to other bug report, this is a\ndifferent issue.\n\n\npostgres=# BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE ;\nBEGIN\npostgres=# SAVEPOINT A;\nSAVEPOINT\npostgres=# SELECT count(*... |
[
{
"msg_contents": "=> select distinct catcode, catdesc from pg_get_keywords();\n catcode | catdesc\n---------+-----------------------\n C | Column name\n T | Type or function name\n R | Reserved\n U | Unreserved\n\nI find the descriptions of C and T quite confusing. For exam... |
[
{
"msg_contents": "While attacking this issue(*1), I found that tuplestore that is on the\nfile status has potential performance problem.\n\nThe performance problem introduced by Heikki's new approach was caused\nby BufFile's frequent flush out in such cases like you put a new row\ninto it and read middle row o... |
[
{
"msg_contents": "Log Message:\n-----------\nProperly unregister OpenSSL callbacks when libpq is done with\nit's connection. This is required for applications that unload\nthe libpq library (such as PHP) in which case we'd otherwise\nhave pointers to these functions when they no longer exist.\n\nThis needs a b... |
[
{
"msg_contents": "\nThe following query causes an assertion failure on CVS head:\n\nSELECT * FROM (SELECT 1 AS i) AS a WHERE NOT EXISTS (SELECT 1 WHERE 1 <> i);\n\nTRAP: FailedAssertion(\"!(!bms_is_empty(min_righthand))\", File: \"initsplan.c\", Line: 685)\n\nLooks like it's assuming there's at least one relat... |
[
{
"msg_contents": "As far as I have understood the following query\n SELECT DISTINCT foo\n FROM bar\n LIMIT baz\nis done by first sorting the input and then traversing the sorted data, \nensuring uniqueness of output and stopping when the LIMIT threshold is \nreached. Furthermore, a part of the sort proce... |
[
{
"msg_contents": "Hello.\n\nI can successfully ALTER a NON-EMPTY table (ct) which ROWTYPE is used as a\ncolumn type for another table (dt):\n\nCREATE TABLE ct (id INTEGER);\nCREATE TABLE dt (id INTEGER, c ct);\n\nINSERT INTO dt VALUES(1, '(666)');\nSELECT * FROM dt;\n-- (1, '(666)')\n\nALTER TABLE ct ADD COLUM... |
[
{
"msg_contents": "This is a fresh snapshot -- is also happened on Nov 28th snapshot, too:\n\n-bash-3.2$ pg_ctl start\nserver starting\n-bash-3.2$ LOG: authentication option not in name=value format: sameuser\n\nCONTEXT: line 70 of configuration file \"/var/lib/pgsql/data/pg_hba.conf\"\nFATAL: could not load... |
[
{
"msg_contents": "Hi all,\n\nI attached pg_upgrade.sh script patch which works now for 8.3->8.4. It is \ncontrib module in contrib/pg_upgrade directory. Just make/make install and it works.\n\nThere are two changes from previous 8.1->8.2.\n\n1) pg_largeobject is also upgraded\n2) added check for dropped column... |
[
{
"msg_contents": "Mark Cave-Ayland wrote:\n> On Thu, 2008-12-04 at 13:51 -0800, Kevin Neufeld wrote:\n> \n>> Wow, that's bad. I just updated to PostgreSQL 8.3.5 from 8.3.3 and I now get the same thing.\n>>\n>> test=# create temp table tmp as select st_makepoint(random(), random()) as the_geom from generate_se... |
[
{
"msg_contents": "On Thu, Nov 27, 2008 at 6:46 PM, Gregory Stark <stark@enterprisedb.com> wrote:\n>> ANALYZE with default_statistics_target set to 10 takes 13 s. With\n>> 100, 92 s. With 1000, 289 s.\n>\n> That is interesting. It would also be interesting to total up the time it\n> takes to run EXPLAIN (with... |
[
{
"msg_contents": "ISTM that the PD_ALL_VISIBLE flag and the visibility map bit can be set at\nthe end of pruning operation if we know that there are only tuples visible\nto all transactions left in the page. The way pruning is done, I think it\nwould be straight forward to get this information.\n\nThanks,\nPav... |
[
{
"msg_contents": "/*\n * We don't need to lock the page, as we're only looking at a single\nbit.\n */\n result = (map[mapByte] & (1 << mapBit)) ? true : false;\n\n\nIsn't this a dangerous assumption to make ? I am not so sure that even a bit\ncan be read atomically on all platforms. I think the only... |
[
{
"msg_contents": "Hi,\n\nSometimes people would like to call C++ code in the PostgreSQL\nbackend environment... for example, in user-defined functions,\ntriggers, access methods. And there is sometimes a need for\nC++ code to call back into PostgreSQL's C functions, such as\nthe SPI interface. Many useful so... |
[
{
"msg_contents": "One issue with packaged multiple-version installations (practiced by \nDebian, Solaris, and soon Fedora?) is that the mo files (translation \nfiles) should be in a fixed location like /usr/share/locale, which does \nnot allow PostgreSQL-version specific subdirectories.\n\nThe Debian packages ... |
[
{
"msg_contents": "\nThe following bug has been logged online:\n\nBug reference: 4566\nLogged by: Randy Isbell\nEmail address: jisbell@cisco.com\nPostgreSQL version: 8.3.4\nOperating system: FreeBSD 6.2\nDescription: pg_stop_backup() reports incorrect STOP WAL LOCATION\nDetails: \n\n... |
[
{
"msg_contents": "FYI, this is not limited to PostGIS GIST, it appears to be all GIST\n(gid is an integer column and I have loaded btree_gist):\n\npramsey=# create table ttt as select * from counties;\nSELECT\npramsey=# create index gid_bix on ttt using gist (gid);\nCREATE INDEX\npramsey=# select count(*) from... |
[
{
"msg_contents": "Who anyone be opposed to \"ssldir = path\" as a connection option? \nCurrently, there is no way to change the homedir method ~/.postgresql \n... or am I missing something? I am willing to supply a patch.\n\n-- \nAndrew Chernow\neSilo, LLC\nevery bit counts\nhttp://www.esilo.com/\n",
"msg... |
[
{
"msg_contents": "While waiting for a large restore to complete (and investigating why\nparts of it were so slow), I came across this scenario. This isn't\nquite the same as some previous discussion of hint bits, but I thought\nit was something that could probably be taken into account in future.\nThis also ma... |
[
{
"msg_contents": "--- repost to hackers as suggested by RhodiumToad ---\n\nhi,\ni tried to test new \"visibility map\" feature.\n\nto do so i:\n1. fetched postgresql sources from cvs\n2. compiled\n3. turned autovacuum off\n4. started pg\n5. ran this queries:\n - CREATE TABLE test_1 (i INT4);\n - CREATE TABLE... |
[
{
"msg_contents": "Sorry for top posting but we are getting a bit far afield from the\noriginal topic. I followed up the tests I did last night:\n\nhttp://archives.postgresql.org/pgsql-hackers/2008-12/msg00369.php\n\nI benchmarked 100 iterations of EXPLAIN on the query Greg Stark put\ntogether as a synthetic b... |
[
{
"msg_contents": "Looking for a way to limited a user to a specific set of queries. I don't think \nthis can be done right now ... or can it? Has this feature request surfaced in \nthe past?\n\nI currently need this as an extra security measure for a libpq client app (want \nto block arbitrary queries from m... |
[
{
"msg_contents": "Hi folks,\n\nWas recently poked and reminded that this patch may be of interest to\nthe community. It was mostly done as an academic exercise, just to see\nhow it works, and so it has a rather hackish feel. The patch adds the\nsequence owner, if available, to psql's \\d <sequence> output, a... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.