threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "This bug report:\nhttp://archives.postgresql.org/pgsql-bugs/2010-11/msg00139.php\nshows that this patch was ill-considered:\nhttp://archives.postgresql.org/pgsql-committers/2010-06/msg00013.php\nand this later attempt didn't fix it, because it still misbehaves in\nHEAD:\nhttp://archives.postg... |
[
{
"msg_contents": "I have been working on a plugin for GiST that has some unusual features:\n\n* The data type for both Node and Leaf keys is large (typically 4222 \nbytes on 32-bit; 4230 bytes on 64-bit).\n\n* Due to the large size the storage class is EXTENDED (main would only \ndegrade to EXTENDED in any c... |
[
{
"msg_contents": "Hi,\n\nWhen we lock on a Xid or VirtualXid, there's no way to obtain clear\ninformation on the reason for locking. Consider the following example:\n\nCREATE TABLE foo (a int);\n\nSession 1:\nBEGIN;\nSELECT 1;\n-- we now have a snapshot\n\nSession 2:\nCREATE INDEX CONCURRENTLY foo_a ON foo(a)... |
[
{
"msg_contents": "Patch attached.\n\nBest,\n\nDavid",
"msg_date": "Mon, 22 Nov 2010 15:36:38 -0800",
"msg_from": "\"David E. Wheeler\" <david@kineticode.com>",
"msg_from_op": true,
"msg_subject": "s/LABEL/VALUE/ for ENUMs"
},
{
"msg_contents": "\n\nOn 11/22/2010 06:36 PM, David E. Wheel... |
[
{
"msg_contents": "So I'm still trying to look into:\n\nhttp://archives.postgresql.org/pgsql-hackers/2010-11/msg00686.php\n\n\nwhich is still broken in 8.4-STABLE even after toms patch, to get better \ndebugging I was trying to get non-translated error messages in the log \nwhich seems to be completely impossib... |
[
{
"msg_contents": "I know there has been a lot of talk about replication getting built into\nPostgres and I know of many projects that aim to fill the role. However, I\nhave not seen much in the way of a serious attempt at multi-master write\nscaling. I understand the fundamental problem with write scaling acro... |
[
{
"msg_contents": "Remove useless whitespace at end of lines\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=fc946c39aeacdff7df60c83fca6582985e8546c8\n\nModified Files\n--------------\nREADME | 2 +-\nc... |
[
{
"msg_contents": "Hackers,\n\nI just thought of a good feature to put on the [E] list for TODO, if\npeople agree: a switch to pg_dump called \"--from-list\" which would take\na filename and dump the list of objects in that file.\n\nObjections to putting this on the TODO?\n\n-- \n ... |
[
{
"msg_contents": "Hi,\n\nhttp://archives.postgresql.org/pgsql-hackers/2010-11/msg01303.php\n\nWhen I did unusual operations (e.g., suspend bgwriter by gdb,\npgbench -i and issue txid_current many times) on the master\nin order to try to reproduce the above HS error, I encountered\nthe following assertion error... |
[
{
"msg_contents": "Per previous threats, I spent some time tonight running oprofile\n(using the directions Tom Lane was foolish enough to provide me back\nin May). I took testlibpq.c and hacked it up to just connect to the\nserver and then disconnect in a tight loop without doing anything\nuseful, hoping to me... |
[
{
"msg_contents": "Hi,\n\nI work on implementing binary protocol, during test \"select ?::int\" has\nbeen \nexecuted few times to compile it to server prepared statement, but after\nthis \nbackend returned field format = text, but sent field data in binary mode.\n\nThe parameter is long and is sent to in text m... |
[
{
"msg_contents": "Hello\n\nWe use a very rich log - for pgFouine processing. But sometime there\nare logger 1MB parameters. It's absolutely useless. These long values\ncan be replaced by\n\n\".... first n chars ... truncated original length: 223636, md5:\njhagjkafhskdjfhdsf\n\nRegards\n\nPavel Stehule\n",
... |
[
{
"msg_contents": "Someone offlist reported query slowness because we don't convert\nfunction calls with all-constant parameters to be a constants before we\nstart a sequential scan:\n\n\tEXPLAIN SELECT * FROM test WHERE \n\t\tx >= to_date('2001-01-01', 'YYYY-MM-DD') AND \n\t\tx <= to_date('2001-01-01', 'YYYY-M... |
[
{
"msg_contents": "I did some profiling of pgbench -j 36 -c 36 -T 500 banging on my\ntwo-core desktop box - with synchronous_commit turned off to keep the\nfsyncs from dominating the profile - and got these results:\n\n29634 4.7124 postgres base_yyparse\n27906 4.4377 postgres ... |
[
{
"msg_contents": "Hi everyone,\nI have a table containing file contents in bytea columns.\nThe functionality I am trying to achieve is having a result set\ncontaining such columns, iterating over them and streaming them while\nzipping them.\nThe problem is that I get ByteArrayInputStream from\nResultSet.getBin... |
[
{
"msg_contents": "Bug #5766 points out that we're still not there yet in terms of having\nsane behavior for locale-specific regex operations in Unicode encoding.\nThe reason it's not working is that regc_locale does this to expand\nthe set of characters that are considered to match [[:alnum:]] :\n\n\t/*\n\t * ... |
[
{
"msg_contents": "Document that a CHECKPOINT before taking a file system snapshot can\nreduce recovery time.\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=7276ab5888d85782d988fc297ad2e176c7ad1bca\n\nModified Files\n--------------\ndoc/src/sgml/b... |
[
{
"msg_contents": "bruce wrote:\n> Alvaro Herrera wrote:\n> > Excerpts from Bruce Momjian's message of nov 24 19:04:30 -0300 2010:\n> > > Bruce Momjian wrote:\n> > \n> > > > OK, we will not call inet_ntop() at all. I moved the CIDR part of\n> > > > adt/inet_net_ntop.c into adt/inet_cidr_ntop.c, and moved the r... |
[
{
"msg_contents": "I've just looked at pg_execute_from_file[1]. The idea here is to execute all\nthe SQL commands in a given file. My comments:\n\n* It applies well enough, and builds fine\n* It seems to work, and I've not come up with a way to make it break\n* It seems useful, and to follow the limited design ... |
[
{
"msg_contents": "Hello Dear All.\n\nHow strings are sorted when LC_COLLATE = ja_JP.UTF-8.\nI tried to read the documention on that,but there are just a few words,\nlike LC_COLLATE determines string sort order,\nIs there a specific reference about this?\nSo I can implement an equivalent string sort function in... |
[
{
"msg_contents": "Hello Dear All.\n\nHow strings are sorted when LC_COLLATE = ja_JP.UTF-8.\nI tried to read the documention on that,but there are just a few words,\nlike LC_COLLATE determines string sort order,\nIs there a specific reference about this?\nSo I can implement an equivalent string sort function in... |
[
{
"msg_contents": "Hi hackers,\n\nAttached is a patch that adds core functionality of SQL/MED. This\npatch provides:\n\n* new option HANDLER for FOREIGN DATA WRAPPER\n * CREATE/ALTER DDLs are supported\n * psql \\dew command shows handler option too\n * pg_dump can dump HANDLER option\n\n* new object type F... |
[
{
"msg_contents": "Hi, hackers,\n\nAttached is a patch that adds file_fdw, FDW which reads records from\nfiles on the server side, as a contrib module. This patch is based on\n\"SQL/MED core functionality\" patch.\n\n[SQL/MED - core functionality]\nhttp://archives.postgresql.org/pgsql-hackers/2010-11/msg01698.... |
[
{
"msg_contents": "Hi, hackers,\n\nAttached is a patch that adds postgresql_fdw, FDW which retrieves\ntuples from external PostgreSQL server, as a contrib module. This\npatch is based on \"SQL/MED core functionality\" patch.\n\n[SQL/MED - core functionality]\nhttp://archives.postgresql.org/pgsql-hackers/2010-1... |
[
{
"msg_contents": ">I've seen this as well be a performance issue, in particular with partitioned tables.\n>Out of habit I now write functions that always cache the value of the function in\n>a variable and use the variable in the actual query to avoid this particular \"gotcha\".\n\nsubquery may be used to cach... |
[
{
"msg_contents": "Win32 buildfarm members are red because of my inet_pton changes. I will\nlook into this in the next day, and also improve how we include C files\nfrom /port for libpq.\n\n-- \n Bruce Momjian <bruce@momjian.us> http://momjian.us\n EnterpriseDB http://ente... |
[
{
"msg_contents": "Hi,\n\nTrying to fix a regression test problem I've left for better days while\ndevelopping the extensions, some help is needed.\n\nThe pg_execute_from_file() function is designed to work with either a\nfilename as its sole argument, or the filename and a VARIADIC text list\nof arguments cont... |
[
{
"msg_contents": "Hi,\n\nSo I've been working on improving locks for foreign key checks, as\ndiscussed in a thread started by Joel Jacobson a while ago. I've posted\nabout this:\nhttp://www.commandprompt.com/blogs/alvaro_herrera/2010/11/fixing_foreign_key_deadlocks/\nhttp://www.commandprompt.com/blogs/alvaro_... |
[
{
"msg_contents": "Hello\n\nour customer showed a very significant memory leak in xml2.\n\ntry to\n\nselect xpath_number('<data>' || generate_series || '</data>','/data')\nfrom generate_series(1,500000);\n\nattention! take all memory very fast\n\nIt never release a memory allocated for context and doctree.\n\nR... |
[
{
"msg_contents": "Hello\n\nthere is difficult working with exception still. Isn't time to move forward?\n\nhttp://archives.postgresql.org/pgsql-general/2010-11/msg01129.php\n\nI am thinking, so we can little bit enhance a GET DIAGNOSTICS statement\n\ncondition item names are defined in SQL 92 - mainly: TABLE_N... |
[
{
"msg_contents": "I thought I could improve the way libpq/Makefile pulls in C files, but\nit turns out MSVC scrapes the OBJS lines from that file, so I had to\nrevert most of my cleanup, attached.\n\n-- \n Bruce Momjian <bruce@momjian.us> http://momjian.us\n EnterpriseDB h... |
[
{
"msg_contents": "Hi, all\n\nI created a pg_ctl patch to fix:\n* BUG #5103: \"pg_ctl -w (re)start\" fails with custom unix_socket_directory \nAllow pg_ctl to work properly with configuration files located outside the PGDATA directory\n\nI tested it under Windows XP sp3.\nAll of configuration files(postgresql.c... |
[
{
"msg_contents": "Hi, a very good business company, I suggest you go to see:\nmobiles3gs.com, and now all of products enjoy great discount, saving\ntime and money.\n\n h--)\n",
"msg_date": "Sat, 27 Nov 2010 22:21:28 +0800",
"msg_from": "Nikhil Sontakke <nikkhils@gmail.com>",
"msg_from_op": true,
... |
[
{
"msg_contents": "We've gotten a few inquiries about whether Postgres can use \"huge pages\"\nunder Linux. In principle that should be more efficient for large shmem\nregions, since fewer TLB entries are needed to support the address\nspace. I spent a bit of time today looking into what that would take.\nMy ... |
[
{
"msg_contents": "Hello,\n\nlast week someone raised in a german webforum the question why \n\"default_tablespace\" is not used for \"CREATE DATABASE\". After a brief \ndiscussion on irc (thanks RhodiumToad) I added a note to the \ndocumentation mentioning this point.\n\nPatch attached.\n\n\nBye\n\n-- \n\t\t\t... |
[
{
"msg_contents": "I ran some tests on my dual-Xeon machine to see what speedup using\nparallel gmake. After two builds to warm up the cache, I saw when\nalternating gmake and gmake -j2:\n\ngmake 0m57.207s\ngmake -j2 0m44.305s\ngmake 0m58.691s\ngmake -j2 0m33.979s\n\nAs you can ... |
[
{
"msg_contents": "I noticed that PL/Python uses a simple wrapper around malloc that does\nereport(FATAL) if malloc returns NULL. I find it a bit harsh, don't we\nnormally do ERROR if we run out of memory?\n\nAnd while looking at how PL/Perl does these things I find that one\nfailed malloc (in compile_plperl_fu... |
[
{
"msg_contents": "In recent discussions of the plan-tree representation for KNNGIST index\nscans, there seemed to be agreement that it'd be a good idea to explicitly\nrepresent the expected sort ordering of the output. While thinking about\nhow best to do that, it struck me that there's some pretty horrendous... |
[
{
"msg_contents": "Hi,\n\nI am working on patch adding xmlvalidate() functionality. LibXML 2.7.7 improved DTD, XSD, Relax-NG validation, so using that. I have idea of creating system table for holding DTDs, XSDs, Relax-NGs (similar as on ORACLE). \n\nIs that good idea? If so, how to implement that table? pg_att... |
[
{
"msg_contents": "Hi list,\n\nOften enough when developing PostgreSQL views and functions, I have\npasted the CREATE OR REPLACE commands into the wrong window/shell and\nran them there without realizing that I'm creating a function in the\nwrong database, instead of replacing. Currently psql does not provide\n... |
[
{
"msg_contents": "Well, it's been a productive holiday weekend. I've completed the\nswitch of the SSI implementation from one conflict pointer in and one\nconflict pointer out per transaction to a list of conflicts between\ntransactions. This eliminated all false positives in my dtester\nsuite. The only tes... |
[
{
"msg_contents": "Dear all,\n\nIn our company, we use both PostgreSQL and MySQL, our developers include me\nthink that the \"INSERT ... ON DUPLICATE UPDATE \" clause is a much more user\nfriendly function,so, would you please add this liked function in\nPostgreSQL, I know we can write PROCEDURE or RULE to sol... |
[
{
"msg_contents": "I have a hot_standby system and use it to bear the load of various reporting\nqueries that take 15-60 minutes each. In an effort to avoid long pauses in\nrecovery, I set a vacuum_defer_cleanup_age constituting roughly three hours of\nthe master's transactions. Even so, I kept seeing recover... |
[
{
"msg_contents": "Hi,\n\nhere's a trivial patch:\nChanged several occurrences of \",...\" to \", ...\" in SQL syntax to be\nconsistent with the rest.\n\ndiff --git a/doc/src/sgml/ref/alter_default_privileges.sgml b/doc/src/sgml/ref/alter_default_privileges.sgml\nindex e1aa293..c27466f 100644\n*** a/doc/src/sgm... |
[
{
"msg_contents": "Hi,\n\nI posted this email on the other postgres lists but did not get a reply. So as a last resort, I came here. I hope somebody can help.\n\nI am looking into the impact of large page sizes on the performance of commercial workloads e.g databases,webserver,virtual machines etc. I was wonde... |
[
{
"msg_contents": "PostgreSQL 9 is breaking for me on line 654 of ip.c. ip.c is checking\non presence SIOCGLIFCONF to determine if it's ok to use linux method\nof polling if addrs etc over ioctl, which is not enough. hpux provides\nthis method in similar fashion, but the structs are named different,\nand have ... |
[
{
"msg_contents": "Hi,\n\nfrequently i have accidents with DELETE/UPDATE commands. In fact, sometimes\nin the last 8 or 9 years (ok, a lot of times) I forget the entire WHERE\nclause or have a “not so perfectly“ WHERE clause, with an awful suprise.\nThere’s no words to figure the horror ever time i see that the... |
[
{
"msg_contents": "Last week, I posted a couple of possible designs for making the\nvisibility map crash-safe, which did not elicit much comment. Since\nthis is an important prerequisite to index-only scans, I'm trying\nagain.\n\nhttp://archives.postgresql.org/pgsql-hackers/2010-11/msg01474.php\nhttp://archive... |
[
{
"msg_contents": "Hello,\n\nHere is the proposal to add synonyms to PostgreSQL. Initial goal is to add synonyms\nfor relations (tables, views, sequences) and an infrastructure to allow synonyms\nfor other database objects in the future. \n\nA thread with discussion of an old proposal by Jonah Harris is here: \... |
[
{
"msg_contents": "In the current KNNGIST patch, the indexable ORDER BY clauses are\ntransmitted to the executor by cramming them in with the index qual\nconditions (the IndexScan plan node's indexqual list), from whence\nthey become part of the ScanKey array passed to the index AM.\nRobert complained that this... |
[
{
"msg_contents": "Fujii, Simon, Greg, etc.:\n\nHas anyone submitted or committed a patch to make Standby Promotion*\neasier, at this point? We discussed it earlier in the dev cycle, but I\ncan't find anything which has actually been submitted.\n\n-- \n -- Josh Berkus\n ... |
[
{
"msg_contents": "Hackers,\n\nSome of you might already be aware that this combination produces a\nfatal startup crash in PostgreSQL:\n\n1. Create an Ext3 or Ext4 partition and mount it with data=journal on a\nserver with linux kernel 2.6.30 or later.\n2. Initdb a PGDATA on that partition\n3. Start PostgreSQL ... |
[
{
"msg_contents": "I have played around a little more, and think I found a problem.\n\nIf given enough time, an unlogged table makes it to disk, and a restart wont clear the data. If I insert a bunch of stuff, commit, and quickly restart PG, it table is cleared. If I let it sit for a while, it stays.\n\nBased... |
[
{
"msg_contents": "So I just made a commit that touched four files in all six active\nbranches, and I see:\n\n$ git push \nCounting objects: 172, done.\nCompressing objects: 100% (89/89), done.\nWriting objects: 100% (89/89), 17.07 KiB, done.\nTotal 89 (delta 80), reused 0 (delta 0)\nTo ssh://git@gitmaster.po... |
[
{
"msg_contents": "Hello,\n\nHere is the proposal: My 1st step towards Intelligent, Integrated database. I am explaining the proposal with the use of example. \nExample: We will have a master table say CustMast and a transaction table say salesOrder table.\nView of CustMast:\nCustCode Number(5),\nCustName ... |
[
{
"msg_contents": "Hi All,\nI am new to postgres.\nI want to know how frequently postgres search for trigger file to switch\nover.\nCan this switch over time be reduced?\nPlz let me know where postgres poll for trigger file. I could find it out in\n\"backend/access/trans/xlog.c\" ? am i right?\n\n\n\n-- \nThank... |
[
{
"msg_contents": "Hello list,\n\nFK's cannot refer to rows in inheritance childs. With some changes in \nLockRows, together with removing the ONLY keyword in ri_trigger.c, it \nwas possible to refer to the rows in child relations. (WIP patch attached)\n\nThough it passes simple tests, it is far from complete. ... |
[
{
"msg_contents": "Hackers,\n\nWe would like to inform you all that our extension to PostgreSQL, that\nincludes hypothetical indexes (and soon index self-tuning), is\navailable through a sourgeforge project.\nThis was suggested at PgCon 2010 and we hope some of you may find it\nuseful, contribute and give us yo... |
[
{
"msg_contents": "I can't see any place that \"cachedFetchXid\" is ever invalidated.\nShouldn't it be invalidated before transaction ID wraparound?\n\nIt's difficult to construct a test case to show whether this is a\nproblem or not, but I couldn't find how this is solved in the code. My\nunderstanding is that... |
[
{
"msg_contents": "When I build HEAD using just \"make -s install\", everything works fine.\nBut when I add \"-j12\", it appears to cause problems. This problem\nappeared very recently. Output below.\n\nI can get it to work simply by running \"make -s -j12 install\" over and\nover, and eventually it succeeds.\n... |
[
{
"msg_contents": "Hi,\n\nI've written twitter_fdw, which makes \"twitter\" table to get twitter\nsearch result by SELECT statement. This is done against WIP SQL/MED\nversion PostgreSQL (git sha-1 is\nc5e027c32b8faf0e2d25b8deecb4d3c05399411c). To play with it, make &&\nmake install, run the install script as yo... |
[
{
"msg_contents": "IIRC we announced that we will EOL 8.1 in 2010 -- given that we are now\nin December, will it be done soon?\n\n(As a packager, I'll be happy to drop is ASAP, so that I can use those\nVMs for new OSes).\n\nRegards,\n\n-- \nDevrim GÜNDÜZ\nPostgreSQL Danışmanı/Consultant, Red Hat Certified Engin... |
[
{
"msg_contents": "In a sleepy email late last night on the crash-safe visibility map\nthread, I proposed introducing a new buffer state BM_UNTIDY. When a\npage is dirtied by a hint bit update, we mark it untidy but not dirty.\n Untidy buffers would be treated as dirty by the background writer\ncleaning scan, ... |
[
{
"msg_contents": "I found there are some author names in some of source files.\nWill we keep their names? And, will we add new author names\nwho wrote patches for those files?\n\nsrc/backend/postmaster/syslogger.c(15) : * Author: Andreas Pflug\nsrc/backend/utils/adt/genfile.c(9) : * Author: Andreas Pflug... |
[
{
"msg_contents": "When writing log, dbms should synchronously flush log to disk. I'm\nwondering, if it is possible that the logs are in disk cache, while the\ncontrol is returned to dbms again, so dbms thinks logs are persistent on\ndisk. In this case, if the disk fails, then there's incorrectness for dbms\nlo... |
[
{
"msg_contents": "Hi there,\n\ncreate table qq (b box);\nCREATE TABLE\nselect count(*), b from qq group by b;\nERROR: could not identify an equality operator for type box\nLINE 1: select count(*), b from qq group by b;\n\nbut following select works fine\n\n select ' (43.6038,48.8664536),(1.3620777,1.44327)':... |
[
{
"msg_contents": "Extensions Patch v15 Review\n===========================\n\nSubmission review\n-----------------\n\n* Is the patch in context diff format?\n\nYes.\n\n* Does it apply cleanly to the current git master?\n\nYes.\n\n* Does it include reasonable tests, necessary doc patches, etc?\n\n`make check` p... |
[
{
"msg_contents": "Hi,\n\nThere is a description about streaming replication in the doc:\n----------------------------------------------------------\n25.2.5. Streaming Replication\n\n:\n:\nIf you set up a WAL archive that's accessible from the\nstandby, wal_keep_segments is not required as the standby can alway... |
[
{
"msg_contents": "Hi all,\n\nI have gone through the code a bit more this time and I think that what\nI originally thought of as a task of executor in the beginning (as I\nread the documentation) seems to be split at multiple points. So i am\nasking a question based on what I have understood so far. Keeping it... |
[
{
"msg_contents": "While working on the KNNGIST documentation I noticed that it's possible\nto create queries that will try to use a KNN scan as the inside of a\nmergejoin, leading to a failure because GIST hasn't got mark/restore\nsupport. For example, in the current HEAD regression database:\n\nregression=# ... |
[
{
"msg_contents": "There has been a lot discussion about index-only scans and how to make the visibillity map crash safe. Then followed by a good discussion about hint bits.\n\nWhat seems to be the main concern is the added wal volume and it makes me wonder if there is a way in-between that looks more like hint... |
[
{
"msg_contents": "Dear Greg,\n\nI have updated the MERGE patch for two main problems.\n\n1. Support of system attributes in MERGE action expressions.\n In old version of MERGE, I use the top join as the only RTE in var name\nspace, during the transformation process in parser. It contains all the\ncommon att... |
[
{
"msg_contents": "Vaibhav Kaushal wrote:\n \n> I want to parallelize the search / scan (specifically hash) by\n> using multiple cores in the processors today.\n \nYou might want to review the efforts of Markus Wanner, who submitted\na series of patches intended to move things in that direction to a\nrecent Co... |
[
{
"msg_contents": "Is there a reason why there are two spaces after the first colon in\n\n \"Sort Method: %s %s: %ldkB\\n\"\n\nIs the second part (Disk/Memory) considered part of the sort method?\n\nThe original commit for that is d2a4a406.\n\n\n",
"msg_date": "Sat, 04 Dec 2010 18:21:55 +0200",
"ms... |
[
{
"msg_contents": "Hi, folks,\n\nI'm a newbie to postgres. I'm confused with xlog and clog.\nTo my initial understanding, xlog is the periodic checkpoint log for data,\nwhile clog is for commit log (very unclear for me ...)\n\nThanks!\n\nHi, folks,I'm a newbie to postgres. I'm confused with xlog and clog. To my... |
[
{
"msg_contents": "I found that postgres uses different semaphore system call on some different\noperating systems.\nFor example, I found that on linux, System V semaphore (semop etc.) is used\nto implement locking, while on Darwin, POSIX semaphore (sem_wait, sem_post\netc.) is used.\nlinux and Darwin support b... |
[
{
"msg_contents": "I am suggesting adding a function to libpq:\n\nPGresult *PQvexec(PGconn *conn, const char *fmt, ...);\n\nIt behaves similar to PQexec, but it allows for printf style varargs and\ndoes connection re-establishing if the connection fails (it can be\ndiscussed if this already to much magic, maybe... |
[
{
"msg_contents": "flyusa2010 fly wrote:\n \n> clog is for commit log (very unclear for me ...)\n \nIt is an important part of tracking tuple visibility.\n \nYou might want to start with this page:\n \nhttp://wiki.postgresql.org/wiki/Hint_Bits\n \n-Kevin\n\n",
"msg_date": "Sun, 05 Dec 2010 08:30:25 -0600",... |
[
{
"msg_contents": "I'm reviving the discussion on the subject topic because I just had\nan epiphany which makes it seem simple to implement. The concept of\nthis is that if you start a SERIALIZABLE READ ONLY transaction in an\nSSI environment when certain conditions are true, it doesn't need to\nacquire predic... |
[
{
"msg_contents": "Attached is a patch that allows CopyReadAttibutesText() and \nCopyReadAttributesCSV() to read arbitrary numbers of attributes. \nUnderflowing attributes are recorded as null, and space is made for \noverflowing attributes on a line.\n\nThis patch doesn't result in any user-visible behavior. T... |
[
{
"msg_contents": "Tom Lane wrote:\n> \"Kevin Grittner\" writes:\n>> I'm reviving the discussion on the subject topic because I just\n>> had an epiphany which makes it seem simple to implement. The\n>> concept of this is that if you start a SERIALIZABLE READ ONLY\n>> transaction in an SSI environment when cer... |
[
{
"msg_contents": "Thanks to all the work they have done with this incredible database, \npostgresql.\n\nI've achieved some success in changing collate operating system (linux) \nto generate sort of way of Brazil Portuguese hopes by adding the \nfollowing code in LC_COLLATE\n\nLC_COLLATE\ncopy \"iso14651_t1_ci\... |
[
{
"msg_contents": "\"Kevin Grittner\" wrote:\n> Andres Freund wrote:\n>> On Thursday 02 December 2010 00:48:53 Kevin Grittner wrote:\n>>\n>>> Is there any provision for one backend to cause a *different*\n>>> backend which is idle in a transaction to terminate cleanly when\n>>> it attempts to process its next... |
[
{
"msg_contents": "After an unfortunate two weeks where I was lost in Nyquil-land and not \npaying as much attention as I should have, I just finished a pass \nchecking over all the open 26 CommitFest items. Just under 50% of the \npatches that were open at the start of the CF are still hanging around, \nwith ... |
[
{
"msg_contents": "On Mon, Dec 6, 2010 at 3:42 PM, Fujii Masao <masao.fujii@gmail.com> wrote:\n> On Fri, Oct 15, 2010 at 9:41 PM, Fujii Masao <masao.fujii@gmail.com> wrote:\n>> The timeout doesn't oppose to 'wait-forever'. Even if you choose 'wait\n>> -forever' (i.e., you set allow_standalone_master to false), ... |
[
{
"msg_contents": "Hi,\n\nI found two typos in xlog.c. Could you apply the attached patch?\n\nRegards,\n\n-- \nFujii Masao\nNIPPON TELEGRAPH AND TELEPHONE CORPORATION\nNTT Open Source Software Center",
"msg_date": "Mon, 6 Dec 2010 17:01:16 +0900",
"msg_from": "Fujii Masao <masao.fujii@gmail.com>",
"... |
[
{
"msg_contents": "Hi,\n\nWalsender doesn't need the periodic wakeups anymore, thanks to\nthe latch feature. So wal_sender_delay is basically useless now.\nHow about dropping wal_sender_delay or increasing the default\nvalue?\n\nRegards,\n\n-- \nFujii Masao\nNIPPON TELEGRAPH AND TELEPHONE CORPORATION\nNTT Open ... |
[
{
"msg_contents": "Hi, folks,\n\nin src/template/darwin:\n\n# Select appropriate semaphore support. Darwin 6.0 (Mac OS X 10.2) and up\n# support System V semaphores; before that we have to use POSIX semaphores,\n# which are less good for our purposes because they eat a file descriptor\n# per backend per max_con... |
[
{
"msg_contents": "Hi,\n\nI want to signal the postmaster (User defined signal) once I created the\ntrigger file (for switch over).\nI want to reduce my switch over time further by signaling the postmaster\nonce the trigger file is created.\n\nIs this provision already there in the code?\nThere ate already SIGU... |
[
{
"msg_contents": "I have encountered a problem while restoring the database. There is a \ntable that contains XML data (BLOB), ~ 3 000 000 records, ~ 5.5Gb of \ndata. pg_restore has been running for a week without any considerable \nprogress. There are plenty of lines like these in the log:\n\npg_restore: proc... |
[
{
"msg_contents": "Hey hackers@,\n\nlibpq execution function works with OIDs. In some cases it is\nhighly recommended specify OIDs of parameters according to\nlibpq documentation.\nWhile developing a database application with libpq and if\napplication works with custom data types IMO reasonable to\nprovide deve... |
[
{
"msg_contents": "Changes:\n\n Sets the local $_TD via C instead of passing an extra argument.\n So functions no longer start with \"our $_TD; local $_TD = shift;\"\n\n Pre-extend stack for trigger arguments for slight performance gain.\n\nPasses installcheck.\n\nTim.",
"msg_date": "Tue, 7 Dec 201... |
[
{
"msg_contents": "After reviewing the two ongoing threads about fixing the wal_sync_method\nfiasco, I think there is general agreement on these two points:\n\n1. open_datasync shouldn't be the default choice\n2. O_DIRECT shouldn't be forcibly bundled in with O_DSYNC/O_SYNC\n\nWhat I suggest we do about the lat... |
[
{
"msg_contents": "Hi,\n\nno, no patch(es) yet. I'm going through plpython.c trying as best I can\nto improve things there. I'll have a patch (or patches) ready for the\nJanuary commitfest, but I thought I'd open up a discussion already to\nspare me having to redo features because the way I attacked the problem... |
[
{
"msg_contents": "Hi, folks,\n\nI trace the running postgres, and I found that there are some randoms writes\nin xlog files!\nTo my impression, log file writing is always sequential, how come random\nwrites happen?\n\nThanks in advance!\n\nHi, folks,I trace the running postgres, and I found that there are some... |
[
{
"msg_contents": "Hey hackers@,\n\nIs it guaranteed that name of array types in pg_type system\ncatalog will always be prefixed by underscore or this convention\ncan be changed in future ?\n\nThanks.\n\n-- \n// Dmitriy.\n\nHey hackers@,Is it guaranteed that name of array types in pg_type systemcatalog will alw... |
[
{
"msg_contents": "Hi,\n\nI found an interesting article which explains how to solve \"sudoku\" by\nusing SQL(unfortunately written in Japanese):\nhttp://codezine.jp/article/detail/1629?p=2\n\nIn the article example sudoku is:\nhttp://static.shoeisha.jp/cz/static/images/article/1629/problem4.gif\n\nTo solve the... |
[
{
"msg_contents": "Heikki pointed out to me that the btree delete record processing does\nnot respect vacuum_defer_cleanup_age. It should.\n\nAttached patch to implement that.\n\nLooking to commit in next few hours barring objections/suggestions, to\nboth HEAD and 9_0_STABLE, in time for next minor release.\n\n... |
[
{
"msg_contents": "Hi there,\n\nbelow is the problem, which I don't have when running in shell. The database is \nin UTF-8 encoding.\n\nCREATE OR REPLACE FUNCTION url_decode(Vkw varchar) RETURNS varchar AS $$\n use strict;\n use URI::Escape;\n return uri_unescape($_[0]); \n$$ LANGUAGE plperlu;\nCRE... |
[
{
"msg_contents": "Optimize commit_siblings in two ways to improve group commit.\nFirst, avoid scanning the whole ProcArray once we know there\nare at least commit_siblings active; second, skip the check\naltogether if commit_siblings = 0.\n\nGreg Smith\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.