threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "\n> The bad news is that except in the stats_command_string cases, HEAD\n> is noticeably slower than 8.1 on the machine with slow gettimeofday.\n> In the single-transaction test this might be blamed on the addition\n> of statement_timestamp support (which requires a gettimeofday per\n> statem... |
[
{
"msg_contents": "We all know that PostgreSQL suffers performance problems when rows are\nupdated frequently prior to a vacuum. The most serious example can be seen\nby using PostgreSQL as a session handler for a busy we site. You may have\nthousands or millions of active sessions, each being updated per page ... |
[
{
"msg_contents": "\n\nOn Wed, 21 Jun 2006, Relyea, Mike wrote:\n\n> ExecutorState: 550339936 total in 123 blocks; 195003544 free (740135\n> chunks); 355336392 used\n> HashBatchContext: 293593176 total in 44 blocks; 3107384 free (80\n> chunks); 290485792 used\n> TIDBitmap: 2088960 total in 8 blocks; 924720 free... |
[
{
"msg_contents": "On CVS tip:\n\n$ postgres --single -Ddata postgres\n\nPostgreSQL stand-alone backend 8.2devel\nSegmentation fault (core dumped)\n\n$ gdb -c core.18119\n(gdb) bt\n#0 pgstat_report_activity (cmd_str=0x67d25e \"<IDLE>\") at pgstat.c:1484\n1484 beentry->st_changecount++;\n(gdb) bt\n#0... |
[
{
"msg_contents": "\n> >> Each time the record is updated, a new version is created, thus \n> >> lengthening the \"correct\" version search each time row is accessed,\n\n> >> until, of course, the next vacuum comes along and corrects the\nindex \n> >> to point to the latest version of the record.\n> >>\n> >> Is... |
[
{
"msg_contents": "I'm developing the summer of code project to create a xlog viewer.\nThe tool we want to create is a DBA tool used for inspect the xlog files,\nlooking for some operations, statistcs of database usage and status of\ntransactions.\n\nSome use cases:\n* Some user made a mistake and commited it t... |
[
{
"msg_contents": "[ redirecting to -hackers, as I see no need for this to be a core issue ]\n\nCharles Comiskey <comiskey@us.ibm.com> writes:\n> Hello,\n> I've recently looked through the PostgreSQL code and a couple of questions \n> surfaced. I was hoping someone here may be able to answer them. Two have \n... |
[
{
"msg_contents": "Log Message:\n-----------\nSome more deformed tuple structures consolidation works. Now even faster but still half way done.\n\nModified Files:\n--------------\n fd:\n algutils.c (r1.9 -> r1.10)\n (http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/fulldisjunction/fd/algutils.c.diff... |
[
{
"msg_contents": "There was a discussion some time back concerning the linking of the \npostgres backend. Some libraries where linked although they where not \nneeded, mainly because it was convenient. I had a problem with PL/Java \nsince a Sun JVM ships with their own version of libz.so (they call it \nlibzip... |
[
{
"msg_contents": "Csaba, \n\n> Regularly clustering is out of question as it would render \n> the system unusable for hours. There's no \"0 activity hour\" \n> we could use for such stuff. There's always something \n> happening, only the overall load is smaller at night...\n\nWe are planning to implement a btr... |
[
{
"msg_contents": "\n> back and forth the data between an archive table and the live \n> table, based on how active the groups are, I can't imagine \n> any other way of partitioning it. And that would also mean \n> some quite big load given the pretty high dynamics of the groups.\n\nYou said the activity comes ... |
[
{
"msg_contents": "Hi,\nI think there is a bug/misscalculation of some rare query i am using.\n\nSuppose we only query one specific relation R.\n\nR contains indices but not on all attributes or not on \nall ordered subset of keys.\n\nQuery example:\n(SELECT * FROM R\nWHERE a=3, b=6,. ...)\nUNION\n(SELECT * FRO... |
[
{
"msg_contents": "Verging on offtopic, but ...\n\nRegarding the best place for session data, memcached isn't really the \nanswer, for opposite reasons as to why it isn't so great to store it \nin the central DB for a bug web farm.\n\nFolks on the memcached lists propose this [ \"I keep all my session \ndata... |
[
{
"msg_contents": "> > This could be a solution... but then I'm not sure how well would do \n> > queries which need the first 10 records based on some criteria which\n\n> > does not include the group id. I guess limit queries across the\nunion \n> > of the partitions don't work too well for now, and we do \n> h... |
[
{
"msg_contents": "Hi for all\n Please,\n I have two doubts. Can somebody help me please??\n First doubt: I want know how many pages Postgres use when execute one query for Nest-loop-Join. I have used �explain analyze�, but it don�t give number of pages�\n Second doubt: Norm... |
[
{
"msg_contents": "I've updated my VC++ build env with latest CVS, and it no longer builds\nbecause of changes to GIST:\n\nsrc\\backend\\access\\gist\\gistutil.c(237) : error C2057: expected constant\nexpression\nsrc\\backend\\access\\gist\\gistutil.c(237) : error C2466: cannot allocate\nan array of constant si... |
[
{
"msg_contents": "bruce wrote:\n> Tom Lane wrote:\n> > Bruce Momjian <bruce@momjian.us> writes:\n> > > I think at some point we have to admit that _polling_ the tables, which\n> > > is what autovacuum does, just isn't going to work well, no matter how\n> > > much it is tweeked, and another approach should be c... |
[
{
"msg_contents": "HEAD built with msvc++ crashes on initdb. I'd appreciate any pointers as\nto where to start looking...\n\nUnhandled exception at 0x0046b9c7 in postgres.exe: 0xC0000005: Access\nviolation reading location 0x000006b3.\n\nBacktrace:\n>\tpostgres.exe!_bt_start_vacuum(RelationData * rel=0x0000067f... |
[
{
"msg_contents": "I originally suggested a methodology for preserving MVCC and everyone is\nconfusing it as update \"in place,\" this isnot what I intended.\n\nHow about a form of vacuum that targets a particular row? Is this\npossible? Would if have to be by transaction?\n",
"msg_date": "Sat, 24 Jun 2006 ... |
[
{
"msg_contents": "There is a well known command called CLUSTER which organizes table\nin specified index's order. It has a drawback, that new tuples added are\nnot in this order. Last night I had idea which could be interesting, I hope.\n\nThe idea is to make use of 'histogram_bounds' collected statistical da... |
[
{
"msg_contents": "Dawid, \n\n> Other idea than using histogram_bounds would be using the \n> position of key inside the index to determine the \"ideal\" \n> place of row inside the table and find the closest free spot \n> there. This would be of course much more precise and wouldn't \n> rely on statistic.\n\nT... |
[
{
"msg_contents": "bruce wrote:\n> Why three? I explained using only two heap tuples:\n> \n> \t[item1]...[tuple1]\n> \n> becomes on UPDATE:\n> ---------->\n> \t[item1]...[tuple1][tuple2]\n> ----->\n> \n> on another UPDATE, if tuple1 is no longer visible:\n> \n> -----... |
[
{
"msg_contents": "\nI did some research on this item. I downloaded the source code to WN from:\n\n\thttp://hopf.math.northwestern.edu/index.html\n\nI could only find the most recent version. wn-2.4.7. I then looked at\nits image.c file:\n\n\thttp://momjian.us/expire/image.c\n\nI looked at the last two functi... |
[
{
"msg_contents": "While looking at the recently-noticed problem that HashAggregate nodes\nstore more columns of the input than they need to, I couldn't help\nnoticing how much of the hashtable space goes into HeapTuple header\noverhead. A couple months ago we were able to get a useful improvement\nin sorting ... |
[
{
"msg_contents": " >> > The installer for 8.1 will let you pick UNICODE encoding\n >> during setup.\n >>\n >> What if he uses initdb directly?\n >\n >Then it should be a simple \"-E UNICODE\", just as on unix.\n >\n >//Magnus\n\nHello, I'm the guy that post this question on psql-es-ayuda.\nThanks to Alvaro for... |
[
{
"msg_contents": "\nCurrently analyze.c and tablecmds.c both have some very similar code to handle\ncopying columns from \"parent\" tables. As long as they were just copying\ncolumns and in the case of tablecmds.c copying check constraints that wasn't\nreally that bad an idea, the code is pretty simple.\n\nHow... |
[
{
"msg_contents": "on IRC somebody mentioned that it took >34h to greate a GIN index (on a\ntsvector) on a ~3 Million column table (wikipedia dump) with a\nreasonable speced box (AMD 3400+).\nAfter getting hold of a dump of said table (around 4,1GB in size) I\nmanaged to get the following timings:\n\ntest=# CRE... |
[
{
"msg_contents": "Log Message:\n-----------\nChange the row constructor syntax (ROW(...)) so that list elements foo.*\nwill be expanded to a list of their member fields, rather than creating\na nested rowtype field as formerly. (The old behavior is still available\nby omitting '.*'.) This syntax is not allow... |
[
{
"msg_contents": "Log Message:\n-----------\nDisallow changing/dropping default expression of a SERIAL column\n\nDhanaraj M\n\nModified Files:\n--------------\n pgsql/src/backend/catalog:\n dependency.c (r1.54 -> r1.55)\n (http://developer.postgresql.org/cvsweb.cgi/pgsql/src/backend/catalog/de... |
[
{
"msg_contents": "Log Message:\n-----------\nClamp last_anl_tuples to n_live_tuples, in case we vacuum a table without\nanalyzing, so that future analyze threshold calculations don't get confused.\nAlso, make sure we correctly track the decrease of live tuples cause by\ndeletes.\n\nPer report from Dylan Hansen... |
[
{
"msg_contents": "\nHi\n\nIs there anybody who knows about \"Turning off disk caching\" in solaris \nmachines.\nIf so, pl. reply back.\n\nThanks\nDhanaraj\n",
"msg_date": "Tue, 27 Jun 2006 16:07:48 +0530",
"msg_from": "Dhanaraj M <Dhanaraj.M@Sun.COM>",
"msg_from_op": true,
"msg_subject": "Turn... |
[
{
"msg_contents": "I have a PL/Java user that performs some lengthy operations. Eventually, \nhe get warnings like:\n\nWARNING: buffer refcount leak: [779] (rel=1663/16440/52989, blockNum=3, \nflags=0x27, refcount=1 2)\n\nI traced this to the function PrintBufferLeakWarning. AFAICS, it's only \ncalled from the ... |
[
{
"msg_contents": "Hi,\n\nI see a performance issue on win32. This problem is causes by the\nfollowing URL. \n\nhttp://support.microsoft.com/kb/823764/EN-US/\n\nOn win32, default SO_SNDBUF value is 8192 bytes. And libpq's buffer is\n8192 too.\n\npqcomm.c:117\n #define PQ_BUFFER_SIZE 8192\n\nsend() may take as ... |
[
{
"msg_contents": "> -----Original Message-----\n> From: pgsql-hackers-owner@postgresql.org \n> [mailto:pgsql-hackers-owner@postgresql.org] On Behalf Of \n> Martijn van Oosterhout\n> On Wed, Jun 28, 2006 at 12:23:13AM +0900, Yoshiyuki Asaba wrote:\n> > Hi,\n> > \n> > I see a performance issue on win32. This pro... |
[
{
"msg_contents": "There's an inconsistency between the handling of trailing whitespace in \nquery parameters in the client jdbc driver compared to the PL/Java SPI \nbased driver. According to Jean-Pierre, the former apparently trims the \ntrailing spaces before passing the query (see below). What is the \ncorr... |
[
{
"msg_contents": "I've been digging into why buildfarm member thrush has been dumping core\nconsistently during the regression tests since the posix_fadvise patch\nwent in. I've confirmed that posix_fadvise() itself will SIGSEGV in a\nstandalone test program, and found that this happens only if\n_FILE_OFFSET_... |
[
{
"msg_contents": "Log Message:\n-----------\nAdd GUC update_process_title to control whether 'ps' display is updated\nfor every command, default to on.\n\nModified Files:\n--------------\n pgsql/doc/src/sgml:\n config.sgml (r1.67 -> r1.68)\n (http://developer.postgresql.org/cvsweb.cgi/pgsql/do... |
[
{
"msg_contents": "-- \nliuchao\n\n-- liuchao",
"msg_date": "Wed, 28 Jun 2006 11:40:45 +0800",
"msg_from": "\"tju tju\" <lc666111@gmail.com>",
"msg_from_op": true,
"msg_subject": ""
}
] |
[
{
"msg_contents": "Bruce,\n\nDo we have anything in the pipeline that would result in page format changes \nfor 8.2? I'm wondering if it's worth reviving pg_upgrade, folks at work are \ninterested ...\n\n-- \nJosh Berkus\nPostgreSQL @ Sun\nSan Francisco\n",
"msg_date": "Tue, 27 Jun 2006 20:47:45 -0700",
... |
[
{
"msg_contents": "I'm building an app where I have several entities that are identified \nusing a UUID (i.e. a 128 bit quantity). My current implementation uses a \ncomposite primary key consisting of two int8 values. It's a bit \ncumbersome and I would much rather have a distinct type. An earlier \nimplementa... |
[
{
"msg_contents": "This is an extension of pgstattuple to query information from indexes.\nIt supports btree, hash and gist. Gin is not supported.\nIt scans only index pages and does not read corresponding heap tuples.\nTherefore, 'dead_tuple' means the number of tuples with LP_DELETE flag.\n\nAlso, I added an ... |
[
{
"msg_contents": "Hi,\n\nI need help finding out how to determine if two types are equality compatible\nand compare them.\n\nI am using the following call to check for equality between two values:\nDatumGetBool(\\\n FunctionCall2(&(fctx->tupleSetAttEQFunctions[attID]->eq_opr_finfo)\\\n , lvalue... |
[
{
"msg_contents": "I have an optimization I'd like to see which I think should be pretty\neasy for someone familiar with the planner code to implement. My\nsituation is this: I have an application using veil[1]. Essentially, I\nhave a schema \"private\" and another \"public\". Private contains regular\ntables, ... |
[
{
"msg_contents": "Buildfarm member platypus is showing a regression failure that I'm\nsurprised we have not seen before:\n\nhttp://www.pgbuildfarm.org/cgi-bin/show_log.pl?nm=platypus&dt=2006-06-28%2014:05:01\n\nBasically what this is showing is that when there is more than one\nreferencing table, the order in ... |
[
{
"msg_contents": "We have now experienced index corruption on two separate but identical\nslony clusters. In each case the slony subscriber failed after\nattempting to insert a duplicate record. In each case reindexing the\nsl_log_1 table on the provider fixed the problem.\n\nThe latest occurrence was on our... |
[
{
"msg_contents": "bruce wrote:\n> Greg Stark wrote:\n> > \n> > Bruce Momjian <bruce@momjian.us> writes:\n> > \n> > > PFC wrote:\n> > > > \n> > > > > My idea is that if an UPDATE places the new tuple on the same page as\n> > > > > the old tuple, it will not create new index entries for any indexes\n> > > > > wh... |
[
{
"msg_contents": "Hi All\n\nI am unable to compile the source code for windows.\nthe command in the INSTALL file nmake /f win32.mak\nis not working !!!\n\n\nCan any one help ???\n\n\nThanks in advance\n\nsandeep\n\nHi AllI am unable to compile the source code for windows.the command in the INSTALL file nmake... |
[
{
"msg_contents": "Hello,\n\nRecently an interview with Marten Mickos (mysql's ceo) appeared in\nrussian IT news (eg http://citcity.ru/12776/). I'd like ask some\n[similar] questions to PostgreSQL core developers and write an article\nin Russian. Actually, I'm sure that there is a great lack of PG news\nin our ... |
[
{
"msg_contents": ">>> Hi, I cant find any function, which tells me something like session\n>>> id. Is there something like that? I need it in my AM, because I need\n>>> to know, if something which I wrote in file was written in this\n>>> current session or previously.\n>>\n>> How about\n>> select procpid||' '|... |
[
{
"msg_contents": "Hannu Krosing wrote:\n> > > But we still have to think about similar cases (index entries pointing\n> > > inside CITC chains), unless we plan to disallow adding indexes to\n> > > tables.\n> > \n> > CREATE INDEX has to undo any chains where the new indexed columns change\n> > in the chain, and... |
[
{
"msg_contents": "Ooops: forgot to cc this to the list.\n\nOn Thu, 2006-06-29 at 19:27 -0400, Tom Lane wrote:\n> Are you *certain* this slave isn't running 8.0.2 or older? If you can\n> verify that, then I guess we need to look for another mechanism that\n> could cause the same kind of thing.\n\nCertain. We ... |
[
{
"msg_contents": "On Thu, 2006-06-29 at 19:59 -0400, Tom Lane wrote:\n\n> Ummm ... you did restart the server? \"select version();\" would be\n> the definitive test.\n\nCan't say I blame you for the skepticism but I have confirmed it again.\n\ntest=# select version();\n ... |
[
{
"msg_contents": "I have no messages since yesterday evening. Any problems with\nmailing list ?\n\n \tRegards,\n \t\tOleg\n_____________________________________________________________\nOleg Bartunov, Research Scientist, Head of AstroNet (www.astronet.ru),\nSternberg Astronomical Institute, Moscow University, ... |
[
{
"msg_contents": "I have two postgres servers, one on linux (fedora core 5), one on \nwindows, both are version 8.1.4.\n\nBoth databases are initialized with locale Croatian and win1250 encoding.\n\nrunning pg_controldata on windows returns this\n\nLC_COLLATE: Croatian_Croatia.1250\nLC_CTYPE: ... |
[
{
"msg_contents": "Here's how we solved the XID indexing problem at Skype. We took\nSlony-s xxid module and made it output 8-byte numbers by keeping\ntrack of wraparound count. Thus having stable relationship\nbetween values.\n\nIt would be good to have such functionality officially in PostgreSQL\nso that all... |
[
{
"msg_contents": "Log Message:\n-----------\nDo a pass of code review for the ALTER TABLE ADD INHERITS patch. Keep\nthe read lock we hold on the table's parent relation until commit.\nUpdate equalfuncs.c for the new field in AlterTableCmd. Various\nimprovements to comments, variable names, and error reporting.... |
[
{
"msg_contents": "I see that the just-applied fillfactor patch has changed the syntax\nfor CREATE TABLE to replace \n\nOptWithOids:\n WITH OIDS { $$ = MUST_HAVE_OIDS; }\n | WITHOUT OIDS { $$ = MUST_NOT_HAVE_OIDS; }\n | /*EMPTY*/ ... |
[
{
"msg_contents": "Hello hackers\n\nI think about solution multi values insert from TODO. There is one \"hack\" \nsolution .. transformation from VALUES (a), (b) into SELECT a UNION ALL \nSELECT b. Or there is next (more general) solution .. implementation of \ntable value constructor http://savage.net.au/SQL/s... |
[
{
"msg_contents": "\nI am seeing a strange failure on the new box Sun donated, when trying to \nbuild 7.4 for the buildfarm. This is what I get:\n\ngmake[3]: Entering directory `/export/home/oicu/bf/root/REL7_4_STABLE/pgsql.5463/src/backend/port'\nccache gcc -O2 -fno-strict-aliasing -g -Wall -Wmissing-prototype... |
[
{
"msg_contents": "Hi Hackers,\n\nI think about mapping NUMERIC(n) to fixed size integers corresponding to n:\n - n < 4 to int2\n - 5 <= n < 9 to int4\n - 10 <= n < 18 to int8\n - 19 <= n to original numeric\n\nNUMERIC is not so efficient datatypes compared to fixed size integers. \nIf the auto se... |
[
{
"msg_contents": "Hello\n\nIs it possible to use transactions with 'per sub table' locks? What I\nmean,if I partition a table and use transaction on that table with\nconstraint,will the database lock the master table (and all subtables),or\nonly sub table where the data is ?\n\nSincerely\n\nDragan Zubac\n\nHel... |
[
{
"msg_contents": "Is there a difference in PostgreSQL performance between these two\ndifferent strategies:\n\n\nif(!exec(\"update foo set bar='blahblah' where name = 'xx'\"))\n exec(\"insert into foo(name, bar) values('xx','blahblah'\");\nor\nexec(\"delete from foo where name = 'xx'\");\nexec(\"insert into ... |
[
{
"msg_contents": "\nIn nbtree.c there's a path that calls btvacuumscan to gather statistics if\nthere aren't already statistics. I'm not exactly clear how this code path is\nreached but that's not my question. There's a comment that \"there's no need to\ngo through all the vacuum-cycle-ID pushups\" in this cas... |
[
{
"msg_contents": "\nSometime in late June the buildfarm passed 50,000 builds reported on. \nHere are stats over the lifetime.\n\ncheers\n\nandrew\n\n\n ym | builds | reporting_members\n\n --------+--------+-------------------\n 2004-10 | 181 | 6\n 2004-11 | 1533 | 12\n ... |
[
{
"msg_contents": "Why are only select, insert, update, and delete supported for $X binds? \nWhy can't preparation be used as a global anti-injection facility?\n\nExample using the backend protocol for binds:\nPREPARE TRANSACTION $1;\nbind $1 ['text']\n-->syntax error at $1\n\nWhy am I able to prepare statement... |
[
{
"msg_contents": "(1) The docs claim that pg_get_viewdef() returns the \"CREATE VIEW\ncommand for view\", but that is clearly not the case:\n\npostgres=# create view v1 as select 1;\nCREATE VIEW\npostgres=# select pg_get_viewdef('v1'::regclass::oid);\n pg_get_viewdef \n----------------\n SELECT 1;\n(1 row)\n\n... |
[
{
"msg_contents": "I just noticed this warning:\n\ngcc -O -Wall -Wmissing-prototypes -Wpointer-arith -Winline \n-Wdeclaration-after-statement -Wendif-labels -fno-strict-aliasing -g \n-pthread -D_REENTRANT -D_THREAD_SAFE -D_POSIX_PTHREAD_SEMANTICS -fpic \n-DFRONTEND -I. -I../../../src/include -D_GNU_SOURCE -I/... |
[
{
"msg_contents": "\nJust wanted to make clear to Hackers that the gates are now open to\ninclude other parameters for CREATE INDEX, as originally requested here:\n\nhttp://archives.postgresql.org/pgsql-hackers/2005-09/msg00851.php\n\nThe new WITH (param=value...) syntax could easily be extended to include\na v... |
[
{
"msg_contents": "I need to build custom win32 binary package for PostgreSQL.\n\nI've downloaded source for PGinstaller but found them hard to\nunderstand - WiX toolkit and MSI is totally alien territory for me.\n\nThings I need to modify:\n\n1. Exclude all unneccessary extensions such as PostGIS\n2. Add some ... |
[
{
"msg_contents": "Hello, \n\nI've got problems building the client libraries. It seems that there this\nproblem is already known and dicussed this mailing list earlier:\n\n\n(snip)\nPatch applied to CVS HEAD and 8.1.X. Thanks.\n\nBorland CC also needed this change, so I modified your patch appropriately.\n\n-... |
[
{
"msg_contents": "I've recently migrated one of my databases to using veil. This involved\ncreating a 'private' schema and moving all tables to it. Functions\nremain in public, and secured views are created there which can be\naccessed by normal users.\n\nIn doing so, I found to my extreme displeasure that alt... |
[
{
"msg_contents": "\nI'm a bit confused about how scan keys work. Is there any simple way given a\nlist of Datums of the same type as the index tuple attributes to get all\nmatching index entries? This is for a non-system index.\n\nIt seems like the only place in the code where non-system index lookups are\ndon... |
[
{
"msg_contents": "test=# create schema private;\nCREATE SCHEMA\ntest=# create sequence private.seq;\nCREATE SEQUENCE\ntest=# create function bump() returns bigint language sql security definer as $$ select nextval('private.seq'); $$;\nCREATE FUNCTION\ntest=# revoke usage on schema private from pfrost;\nREVOKE\... |
[
{
"msg_contents": "I've worked on a prototype (attached to this email) of the SRF function and\nI can query the xlog files for some useful info.\nI know that the error codes are still incorrect and the tests are missing,\nbut this is only a proof of concept.\n\nExamples of usage:\nQuery for committed transactio... |
[
{
"msg_contents": "There are roughly three weeks left until the feature freeze on August 1.\nIf people are working on items, they should be announced before August\n1, and the patches submitted by August 1. If the patch is large, it\nshould be discussed now and an intermediate patch posted to the lists\nsoon.\... |
[
{
"msg_contents": "Hello,\n\nMy name is Beth Jen and I'm a Summer of Code student currently adding a\nhash-based implementation of DISTINCT to PostgreSQL. My prototype is near\ncompletion, and the current design is as follows:\n\nI've created a new exec node that uses hash table functions in execGrouping.\nThe... |
[
{
"msg_contents": ".\\src\\bin\\psql\\print.c(1912): warning C4090: 'function' : different\n'const' qualifiers\n.\\src\\bin\\psql\\print.c(1913): warning C4090: 'function' : different\n'const' qualifiers\n\nVariables are declared const, are you really supposed to pass that to\nfree()? Cast at free()? Remove con... |
[
{
"msg_contents": "Hi,\nHow do i get a char * type from a type Oid. i.e. getStringTypeFromOid(INT4OID) \nwill return \"int4\".\n10x.\n\n-- \nRegards,\n Tzahi.\n--\nTzahi Fadida\nBlog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info\nWARNING TO SPAMMERS: see at \nhttp://members.lycos.co.u... |
[
{
"msg_contents": "A while ago in connection with the 8.2 planning [1] there was some\ndiscussion of resource management and statement queuing [2].\n\nI am currently looking at implementing a resource management/queuing\nmodule for Greenplum - initially targeting Bizgres, but I'm thinking it\ncould be beneficia... |
[
{
"msg_contents": "\nPursuant to a conversation this evening I would like to a suggestion:\n\n BIRT pgsql-patches should be abolished in favour of something else that\n accomplishes the bandwidth-reduction aspect without the downsides.\n\nMy complaint is that -patches serves to\n\na) siphon off some of the most... |
[
{
"msg_contents": "While replying to the \"information_schema for all users\" thread in\npgsql-sql I noticed that a cursor returned from a SECURITY DEFINER\nfunction evalutes current_user as the user who executes FETCH, not\nas the user who defined the function that opened the cursor. Here\nare the question an... |
[
{
"msg_contents": "\nThe SQL spec has some detailed discussion of some strange null behaviours.\n\nSpecifically \"row(1,null) is null\" is false but \"row(1,null) is not null\" is\n*also* supposed to be false. Postgres currently gets this wrong. \"is [not]\nnull\" is apparently supposed to mean \"all the fields... |
[
{
"msg_contents": "Hello\n\nI have problem with expressions:\n\nselect (select row(ret_num,ret_den) from mp.x_numer(39,32)) = row(32,33);\nselect mp.x_numer(39,32)) = row(32,33);\n\nboth has error:\nERROR: operator does not exist: record = record\nLINE 1: ...lect row(ret_num,ret_den) from mp.x_numer(39,32)) = ... |
[
{
"msg_contents": "Folks,\n\nFor the code sprint, I'm starting off by removing the projects from \ncontrib which need to be removed by still have some usefulness. I'm not \nexactly sure what to do with adddepends, though. It seems unlike to \nlead an independant existance on pgFoundry; I'm inclined to just n... |
[
{
"msg_contents": "Now that the index options infrastructure is in, I am having a couple of\nsecond thoughts about the specific behavior that's been implemented,\nparticularly for btree fillfactor.\n\n1. The btree build code (nbtsort.c) is dependent on the assumption that\nthe fillfactor is at least 2/3rds. Th... |
[
{
"msg_contents": "Hi\n\nI've now setup a warm-standby machine by using wal archiving. The restore_command on the\nwarm-standby machine loops until the wal requested by postgres appears, instead of\nreturning 1. Additionally, restore_command check for two special flag-files \"abort\"\nand \"take_online\". If \"... |
[
{
"msg_contents": "Hi Bruce,\n\nIn order to fix a win32 build error with thread safety enabled, can you\nplease move src/interfaces/libpq/pthread-win32.h to\nsrc/include/port/win32/ ?\n\nI've updated snake to build with thread safety enabled so hopefully this\nwon't go unnoticed if broken in the future.\n\nThan... |
[
{
"msg_contents": "Folks,\n\nI was looking at migrating mSQL-interface to pgFoundry, but I'm not sure\nthere's any reason to do so. It was never finished, doesn't build, and\nit's not like I run across mSQL databases in the field. Does anyone?\n\nShall we just kill it?\n\nAlso, \"tips\" is an apache log conv... |
[
{
"msg_contents": "Hi,\nFirst, i use CTIDs to immensely speed up my function which is inherently slow \nbecause of the problem itself. \n\nI have a question about CTID invalidation when you open a read only cursor \nusing SPI. Why does it at all happens? Why is it so important to invalidate a \nctid of a read o... |
[
{
"msg_contents": "Dear Hackers,\n\nI would like to thank all of you for organizing, hosting, and attending \nthe 10th Anniversary PostgreSQL Conference last weekend. I was \nespecially interested in future PostgreSQL directions and that was \ndefinitely the conference's theme. It was great to meet the communit... |
[
{
"msg_contents": "Hi,\n\nWe came up with this patch in response to a problem reported to us by a\nclient. They had a query which took an unacceptably long time to respond\nto a cancel request (SIGINT). The client uses 8.1.4, so the patch is\nagainst that.\n\nTheir work_mem setting was rather large (1000000). W... |
[
{
"msg_contents": "Hi,All.\n\nThe problem has occurred in my customer.\npoor performance with Context Switch Storm occurred\nwith the following composition.\nUsually, CS is about 5000, WIPS=360.\nwhen CSStorm occurrence, CS is about 100000, WIPS=60 or less.\n(WIPS = number of web interactions per second)\n\nIt ... |
[
{
"msg_contents": "I am in the process of cleaning up the include files. Each include file\nshould compile on its own, and we should only include files we need. \nAlso, PostgreSQL include files should be in alphabetical order.\n\nI do this cleanup every few years.\n\n-- \n Bruce Momjian bruce@momjian.us\n ... |
[
{
"msg_contents": "Currently, we restrict btree index tuples to a size that ensures three of\nthem will fit on a page. The motivation for this is the following two\nconsiderations:\n\n1. In a non-rightmost page, we need to include a \"high key\", or page\nboundary key, that isn't one of the useful data keys.\n... |
[
{
"msg_contents": "On Marko Kreen's detailed suggestion, I've implemented a restartable\nrecovery mode for archive recovery (aka PITR). Restart points are known\nas recovery checkpoints and are normally taken every 100 checkpoints in\nthe log to ensure good recovery performance.\n\nAn additional mode\n\tstandby... |
[
{
"msg_contents": "Hi :)\n\npowered by the great talk for newbie hackers by Neil Conway & Gavin\nSherry [1] at the anniversary summit I sneaked into the Code Sprint\nand started working a bit on a Todo from Neil's Code Sprint Page:\n\n \"Add a psql option to have it submit SELECT queries via a cursor,\n to av... |
[
{
"msg_contents": "Few cleanups and couple of new things:\n\n - add SHA2 algorithm to older OpenSSL\n - add BIGNUM math to have public-key cryptography work\n on non-OpenSSL build.\n - gen_random_bytes() function\n\nThe status of SHA2 algoritms and public-key encryption\ncan now be changed to 'always availabl... |
[
{
"msg_contents": "Currently, functional indexes can be used by a query that explicitly\nmentions the exact phrasing of the functional index within the WHERE\nclause.\n\nIMHO it is feasible to extend the range of WHERE clauses for which\nfunctional indexes can be used by using implication, much in the same\nway... |
[
{
"msg_contents": "Hi,\n\nHannu Krossing asked me about his patch to ignore transactions running\nVACUUM LAZY in other vacuum transactions. I attach a version of the\npatch updated to the current sources.\n\nJust to remind what this is about: the point of the patch is to be able\nto run more than one VACUUM LA... |
[
{
"msg_contents": "I am trying to create an initialisation function that is called using\nthe preload_libraries option.\n\nThe purpose of this is to set up shared memory for Veil, independant of\npostgres' own shared memory. Simple init functions work fine, but as\nsoon as I place calls to ShemAlloc, or LWLoc... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.