threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "Hackers,\n\nSequence of events:\n\n1. PITR backup of server on timeline 2.\n\n2. Restored the backup to a new server, new-master.\n\n3. Restored the backup to another new server, new-replica.\n\n4. Started and promoted new-master (now on Timeline 3).\n\n5. Started new-replica, connecting over... |
[
{
"msg_contents": "A recent thread in pgsql-general shows yet another user who's befuddled by\nthe need to add a USING clause to an ALTER TABLE ... ALTER COLUMN TYPE\ncommand:\n\nhttp://www.postgresql.org/message-id/flat/CAD25zGDiaqCG=eqXp=byvZcdgvcquBh7KBpJpJQseSPOwfvhiw@mail.gmail.com\n\nSpecifically, it's no... |
[
{
"msg_contents": "We hope to have a chance to discuss this during the upcoming developer\nunconference in Ottawa. Here are some preliminary ideas to shed some\nlight on what we're trying to do.\n\n\nI've been trying to figure out a plan to enable native column stores\n(CS or \"colstore\") for Postgres. Motiv... |
[
{
"msg_contents": "Hi, all.\n\nI am newbie in hackers.\nI have an idea from my point of view as one user, I would like to propose the following.\n\n\nProgress bar for pg_dump / pg_restore\n=====================================\n\nMotivation\n----------\n\"pg_dump\" and \"pg_restore\" show nothing if users don't... |
[
{
"msg_contents": "Could we update our git hook to refuse a push of a new commit whose\ntimestamp is more than, say, 24 hours in the past? Our commit history\nhas some timestamps in it now that are over a month off, and it's\nreally easy to do, because when you rebase a commit, it keeps the old\ntimestamp. If... |
[
{
"msg_contents": "Commit 902d1cb3, made in 2008, established that the functions\nheap_formtuple(), heap_modifytuple(), and heap_deformtuple() were\ndeprecated. The commit also actually removed those routines, replacing\nthem with simple wrappers around their real replacements, which are\nspelled slightly diffe... |
[
{
"msg_contents": "Hi all,\n\nI've noticed that LIKE tablename INCLUDING INDEXES skips any indexes\nthat were duplicated.\n\ne.g.\n\nCREATE TABLE people (id int, alias text);\nCREATE INDEX idx_people_id_1 ON people (id);\nCREATE INDEX idx_people_id_2 ON people (id) WHERE id % 2 = 0;\nCREATE INDEX idx_people_ali... |
[
{
"msg_contents": "Hi\n\nI am working on scheduler extension for 9.5. It use bgworkers intensively\nfor any task. This is reason, why I need to decrease a log level - and I am\nthinking so parallel computing needs it due high number of created and\nfinished workers.\n\nIt should be fixed in 9.5 - because it is ... |
[
{
"msg_contents": "Hi,\n\nlooks like I omitted psql tab completion from the TABLESAMPLE patch. The \nattached patch adds it.\n\n-- \n Petr Jelinek http://www.2ndQuadrant.com/\n PostgreSQL Development, 24x7 Support, Training & Services\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hac... |
[
{
"msg_contents": "I think the docs for the LEAKPROOF option in create_function.sgml\nought to mention RLS as well as security barrier views. Also the\ncurrent text is no longer strictly correct in light of commit\ndcbf5948e12aa60b4d6ab65b6445897dfc971e01. Suggested rewording\nattached.\n\nRegards,\nDean\n\n\n\... |
[
{
"msg_contents": "Hi,\n\nThis is an attempt to revive the resetsysid tool submission that didn't \ngo anywhere last year [1].\n\nIn that thread there were mainly two complains, one that pg_resetxlog \nwould preferably be tool for more advanced stuff and similar utilities \nthat are not as dangerous should have... |
[
{
"msg_contents": "If you ever had trouble understanding a git manual page, this humorous\nrandom git man page generator is full of fun:\n\n\thttp://git-man-page-generator.lokaltog.net/\n\nTry a few \"generate\" runs to find one that seems logical. :-)\n\n-- \n Bruce Momjian <bruce@momjian.us> http://... |
[
{
"msg_contents": "I have run a script to count the number of \"<listitem>\" items in the\nmajor release notes of each major version of Postgres back to 7.4:\n\n\t7.4 280\n\t8.0 238\n\t8.1 187\n\t8.2 230\n\t8.3 237\n\t8.4 330\n\t9.0 252\n\t9.1 213\n\t9.2 250\n\t9.3 187\n\t9.4 21... |
[
{
"msg_contents": "Buildfarm member hamster has failed a pretty significant fraction of\nits recent runs in the BinInstallCheck step:\nhttp://buildfarm.postgresql.org/cgi-bin/show_history.pl?nm=hamster&br=HEAD\n\nSince other critters aren't equally distressed, it seems likely that\nthis is just an out-of-disk-s... |
[
{
"msg_contents": "This patch tracks memory usage (at the block level) for all memory\ncontexts. Individual palloc()s aren't tracked; only the new blocks\nallocated to the memory context with malloc().\n\nIt also adds a new function, MemoryContextMemAllocated() which can\neither retrieve the total allocated for... |
[
{
"msg_contents": "Hi,\n\nI am getting tired installing manually required extensions manually. I \nwas wondering if we might want to add option to CREATE SEQUENCE that \nwould allow automatic creation of the extensions required by the \nextension that is being installed by the user.\n\nI also wrote some prototy... |
[
{
"msg_contents": "I've encountered a query with 11 joins whose execution time (i.e., the time\nnot taken up by planning) is significantly faster with geqo on rather than\noff. This is surprising to me and seems like it might be a bug in the\nplanner, so I am posting it here rather than to -performance.\n\nThe ... |
[
{
"msg_contents": "Simon and I have been going over some ideas about how to make improvements\nto aggregate performance by cutting down on the duplicate work that's done\nwhen 2 aggregate functions are used where one knows how to satisfy all the\nrequirements of the other.\n\nTo cut a long story short, all our ... |
[
{
"msg_contents": "Hi all\n\nI frequently see users confused by one of our more common and less\nclear error messages:\n\n fe_sendauth: no password supplied\n\nWhat this really means is that the server requested a password for md5\nor cleartext authentication but no password was supplied to the client\nand i... |
[
{
"msg_contents": "While reviewing the seqam patches, I noticed that psql has tab\ncompletion for ALTER SEQUENCE, but not for CREATE SEQUENCE.\n\nThe attached trivial patch fixes that.\n-- \nVik Fearing +33 6 46 75 15 36\nhttp://2ndQuadrant.fr PostgreSQL : Expertise,... |
[
{
"msg_contents": "... has this comment (fmgr.h:506):\n\n /* These are for invocation of a function identified by OID with a\n * directly-computed parameter list. Note that neither arguments\n nor result\n * are allowed to be NULL. These are essentially FunctionLookup()\n followed\n * ... |
[
{
"msg_contents": "In XidInMVCCSnapshot, it will check xid from tuple if is in snapshot->subxip. It means tuple store subtransaction?\n\nBut in PushTransaction, I see TransactionState.subTransaction will assign currentSubTransactionId, currentSubTransactionId will reinitialize in StartTransaction. So I think tu... |
[
{
"msg_contents": "Hi all,\n\nI noticed while using gin function of pageinspect that there are some\ninconsistency data types.\nFor example, data type of GinMetaPageData.head, and tail is\nBlockNumber, i.g, uint32.\nBut in ginfunc.c, we get that value as int64.\n\nSo I think the output is odd a little. Is it i... |
[
{
"msg_contents": "> Date: Thu, 4 Jun 2015 13:55:52 +0530\n> From: Jeevan Chalke <jeevan.chalke@enterprisedb.com>\n> To: david.g.johnston@gmail.com\n> Cc: PostgreSQL Hackers <pgsql-hackers@postgresql.org>\n> Subject: Re: [PATCH] two-arg current_setting() with fallback\n> Message-ID: <CAM2+6=VsTg=pAEdq6ioKpoLx7y... |
[
{
"msg_contents": "Hi,\n\nCurrently the issue is easily reproducible. Steps to reproduce:\n* Set some aggressive values for auto-vacuuming.\n* Run a heavy database update/delete/insert queries. This leads to invoking\nauto-vacuuming in quick successions.\n* Change the system time to older for eg. 1995-01-01\n\n... |
[
{
"msg_contents": "Hi all;\n\nWe are running a multi-TB bioinformatics system on PostgreSQL and use a \ndenormalized schema in\nplaces with a lot of tsvectors aggregated together for centralized \nsearching. This is\nvery important to the performance of the system. These aggregate many \ndocuments (sometimes ... |
[
{
"msg_contents": "Hi all,\n\nI just bumped into this report regarding pg_rewind, that impacts as\nwell the version shipped in src/bin/pg_rewind:\nhttps://github.com/vmware/pg_rewind/issues/45\n\nIn short, the issue refers to the fact that if the source server\nfilemap includes xlogtemp files pg_rewind will sur... |
[
{
"msg_contents": "Hi,\n\nCurrently we observed that certain postgres child process, for eg.\nautovacuum worker, are not working as expected if there is a system time\nchange. So I wanted to know if postgres already supports system time\nchanges or not.\n\nPlease confirm if postgres already handles system time ... |
[
{
"msg_contents": "Hi\n\nWe saw a rather extreme performance problem in a cluster upgraded from\n9.1 to 9.3. It uses a largish number of child tables (partitions) and\nmany columns. Planning a simple UPDATE via the base table started\nusing a very large amount of memory and CPU time.\n\nMy colleague Rushabh L... |
[
{
"msg_contents": "Hi,\n\nI'm currently running some tests on a 3TB TPC-H data set, and I tripped \nover a pretty bad n_distinct underestimate, causing OOM in HashAgg \n(which somehow illustrates the importance of the memory-bounded hashagg \npatch Jeff Davis is working on).\n\nThe problem is Q18, particularly ... |
[
{
"msg_contents": "The current HEAD of postgres in the git repo is not building when using\n\"make world\". It's been like this for about a month or so that I've been\naware of. I didn't really need the world build so been making due without\nit. At PGCon now, though, so asked Bruce and he said this error was d... |
[
{
"msg_contents": "Hi\n\nI am trying to check your patch - but diffs that you sent to mailing list\nare inconsistent.\n\nPlease, can you send one complete patch?\n\nRegards\n\nPavel\n\nHiI am trying to check your patch - but diffs that you sent to mailing list are inconsistent.Please, can you send one complete ... |
[
{
"msg_contents": "I've written an extension that hooks ExecutorStart_hook and\nExecutorEnd_hook. The hooks are assigned in _PG_init() (and the previous\nones saved to static vars) and reset to the previous values in _PG_fini().\nMaybe also of interest is the extension library is set in postgresql.conf\nas a l... |
[
{
"msg_contents": "Hi all,\n\nThe TAP tests in src/test/ssl are using system() in combination with\nchmod, but perl has a command chmod integrated into it, and it would\nwork better on Windows as well.\nThe attached patch is aimed at fixing that.\n\nRegards,\n-- \nMichael\n\n\n\n-- \nSent via pgsql-hackers mail... |
[
{
"msg_contents": "Hi all,\n\nIn pg_regress.c, there is the following code:\nstatic void\nstop_postmaster(void)\n{\n if (postmaster_running)\n {\n /* We use pg_ctl to issue the kill and wait for stop */\n char buf[MAXPGPATH * 2];\n int ... |
[
{
"msg_contents": "Hi,\n\nI have observed that we are not tab-completing word PASSWORD in the\nfollowing\nsyntaxes:\n\n1.\nCREATE|ALTER ROLE|USER rolname\n\n2.\nCREATE|ALTER ROLE|USER rolname WITH\n\nPASSWORD is used many times and should be in the tab-complete list.\n\nWas there any reason we have deliberately... |
[
{
"msg_contents": "Hi,\n\nTo make slot usage in pg_receivexlog easier, should we add\n--create-slot-if-not-exists? That'd mean you could run the same command\nthe first and later invocation.\n\nGreetings,\n\nAndres Freund\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make chan... |
[
{
"msg_contents": "Change TAP test framework to not rely on having a chmod executable.\n\nThis might not work at all on Windows, and is not ever efficient.\n\nMichael Paquier\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/pg/commitdiff/ca3f43aa48a83013ea50aeee7cd193a5859c4587\n\nModifi... |
[
{
"msg_contents": "During the Developer Unconference at PGCon, we made this wiki page to\ntrack the MultiXact bugs that are outstanding:\n\nhttps://wiki.postgresql.org/wiki/MultiXact_Bugs\n\nLet's update this page as we find or fix things, and let's release\nwhen we've fixed enough stuff.\n\nEven if you are not... |
[
{
"msg_contents": "I think we're missing a dependency declaration from the backend to\nlibpgport in the makefiles somewhere. I just git updated on 9.1 on a\ntree that was already built, ran make, and got this:\n\n../../src/port/libpgport_srv.a(dirmod_srv.o): In function `rmtree':\n/pgsql/source/REL9_1_STABLE/s... |
[
{
"msg_contents": "Hi list\n\nOne of my databases failed to upgrade successfully and produced this error\nin the copying phase:\n\nerror while copying relation \"pg_catalog.pg_largeobject\"\n(\"/srv/ssd/PG_9.3_201306121/1/12023\" to \"/PG_9.4_201409291/1/12130\"): No\nsuch file or directory\n\nTurns out this ha... |
[
{
"msg_contents": "I came across the following bug this week:\n\nSession 0:\nbegin;\ncreate schema bug;\nalter default privileges in schema bug grant all on tables to postgres;\ncommit;\n\nSession 1:\nbegin;\nalter default privileges in schema bug grant all on tables to postgres;\n\nSession 2:\nalter default pr... |
[
{
"msg_contents": "Has anybody noticed the way castoroides is randomly failing?\n\n SELECT test_shm_mq_pipelined(16384, (select string_agg(chr(32+(random()*95)::int), '') from generate_series(1,270000)), 200, 3);\n! PANIC: stuck spinlock (100cb92f4) detected at atomics.c:30\n! server closed the connection une... |
[
{
"msg_contents": "I would like to enhance the postgres_fdw to allow more complete support for user-defined types. \n\nRight now, postgres_fdw already does a good job of passing user-defined type data back and forth, which is pretty nice. However, it will not pass functions or operators that use user-defined ty... |
[
{
"msg_contents": "Hi\n\nThe comment for get_relation_info should probably include serverid in\nthe list of rel members that it can update (see attached).\n\n-- \nThomas Munro\nhttp://www.enterprisedb.com\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your sub... |
[
{
"msg_contents": "Hi,\n\nWhen discussing multixacts around pgcon (including the developer\nmeeting) we decided that we want to add test code to make exercising the\ntruncation code easier.\n\nIn master the most logical place for that seems to be src/test/modules.\nUnfortunately 9.3/9.4 don't have that. Given t... |
[
{
"msg_contents": "Hi,\n\nAs discussed on list, over IM and in person at pgcon I want to make\nmultixact truncations be WAL logged to address various bugs.\n\nSince that's a comparatively large and invasive change I thought it'd be\na good idea to start a new thread instead of burying it in a already\nlong thre... |
[
{
"msg_contents": "Hi\n\nI often need a function for identification if current user is database\nowner or is superuser.\n\nIt can be pretty simply implemented in C level.\n\nDo you think it should be available in core?\n\nRegards\n\nPavel\n\nHiI often need a function for identification if current user is databa... |
[
{
"msg_contents": "I've noted that upgrading from PostgreSQL 9.3 to 9.5 I'm suddenly\nunable to specify a \"check\" rule in the Makefile that includes the\nPGXS one. The error is:\n\n $ make check\n rm -rf ''/tmp_install\n make -C '/home/postgresql-9.5/lib/pgxs/src/makefiles/../..' DESTDIR=''/tmp_install instal... |
[
{
"msg_contents": "When a PostgreSQL system wedges, or when it becomes dreadfully slow\nfor some reason, I often find myself relying on tools like strace,\ngdb, or perf to figure out what is happening. This doesn't tend to\ninstill customers with confidence; they would like (quite\nunderstandably) a process th... |
[
{
"msg_contents": "Hello,\n\nThere are two places in parse_func.c where memcmp() conditionally gets a \nNULL as its first argument, which invokes undefined behavior. I guess \ngcc -O2 will make some assumptions based on memcpy's __nonnull attribute.\n\nYou can see when exactly that happens by applying the two t... |
[
{
"msg_contents": "Hi all,\n\nSome grepping is showing up that a couple of newlines are missing in\npg_rewind, leading to unreadable log entries:\nlibpq_fetch.c: pg_log(PG_DEBUG, \"getting file chunks\");\nlogging.c: pg_log(PG_PROGRESS, \"%*s/%s kB (%d%%) copied\",\nfilemap.c: pg_fatal(\"could ... |
[
{
"msg_contents": "\r\n\r\nHello, \r\n\r\nI'm trying to figure out how a query and its result is passed between a coordinator and a datanode. I know there are many messages passed between them to finish a query.\r\n\r\n\r\nI did a test against the coordinator by adding a row to a table and the sql was, insert i... |
[
{
"msg_contents": "Empirically, I seem to be getting the _PG_init call for a module while\nthe active memory context lifetime is that of the function call which\nfirst needed to load the shared object.\n\nIs this the case ? Documented anywhere ?\nInitializing memory meant to be alive for the whole lifetime of a... |
[
{
"msg_contents": "If I use the btree_gin extension to build a gin index on a scalar value, it\ndoesn't work well with BETWEEN queries. It looks like it scans the whole\nindex, with the part of the index between the endpoints getting scanned\ntwice. It is basically executed as if \"col1 between x and y\" were... |
[
{
"msg_contents": "Those of you who have been following\nhttp://www.postgresql.org/message-id/flat/1d3bc192-970d-4b70-a5fe-38d2a9f762b3@me.com\nare aware that Red Hat shipped a rather broken version of openssl last\nweek. While waiting for them to fix it, I've been poking at the behavior,\nand have found out t... |
[
{
"msg_contents": "Hackers,\n\nAt one site, I have some duplicate row corruption in a staging database.\n This database was created by a binary backup from 9.3.5, which was\nrestored via PITR with a timestamp target to 9.3.5, so known-bad\nversions. The strange thing about the duplicate rows is that they were\... |
[
{
"msg_contents": "Chasing a problem identified by my Salesforce colleagues led me to the\nconclusion that my commit f3b5565dd (\"Use a safer method for determining\nwhether relcache init file is stale\") is rather borked. It causes\npg_trigger_tgrelid_tgname_index to be omitted from the relcache init file,\nb... |
[
{
"msg_contents": "Hi,\n\nWhile reviewing the foreign join pushdown core patch, I noticed that the\npatch doesn't perform an EvalPlanQual recheck properly. The example\nthat crashes the server will be shown below (it uses the postgres_fdw\npatch [1]). I think the reason for that is because the ForeignScan nod... |
[
{
"msg_contents": "Hi,\n\nINSERT ON CONFLICT DO UPDATE doesn't seem to work on the current partitioning\nmechanism. For example, in the following SQL commands, the last UPSERT command\nwould fail with an error. The error message is\n\n ERROR: duplicate key value violates unique constraint \"hoge_20150601_pk... |
[
{
"msg_contents": "Dear hackers,\n\nThe text below has also been published at:\nhttps://trustly.com/en/postgresql-bug-bounty/\n\nYou are most welcome to contact me at joel@trustly.com or by replying\nto this thread, if you have any suggestions on how to improve this\nbug bounty program, or if you have any other... |
[
{
"msg_contents": "Hello,\n\nI'm interested in adding GiST support for the UUID column type from\nthe uuid-ossp extension. This has been requested and attempted before:\n\n http://dba.stackexchange.com/questions/83604/optimizing-postgres-row-overlap-constraints-involving-uuids-and-gist\n http://www.postgr... |
[
{
"msg_contents": "The transforms patch seems to have forgotten to add\nTransformRelationId to object_classes[], much like the RLS patch\nforgot to add PolicyRelationId in the same place.\n\nFixing this is easy, but ISTM that we need to insert some sort of a\nguard to prevent people from continuing to forget th... |
[
{
"msg_contents": "Is there a way to implement triConsistent for only some of the strategy\nnumbers?\n\nIt looks like I would have to internally re-implement something like\nshimTriConsistentFn for each strategy number for which I don't want to\nimplement the ternary system in full. Am I missing a trick?\n\nTh... |
[
{
"msg_contents": "I worry that \"8.14.3. jsonb Containment and Existence\" is not\nsufficiently clear in explaining that jsonb containment is nested.\nI've seen anecdata suggesting that this is unclear to users. We do\nsay:\n\n\"\"\"\nThe general principle is that the contained object must match the\ncontainin... |
[
{
"msg_contents": "Hi all,\n\nAfter looking at the issues with the TAP test suite that hamster faced\na couple of days ago, which is what has been discussed on this thread:\nhttp://www.postgresql.org/message-id/13002.1434307025@sss.pgh.pa.us\n\nI have developed a patch to improve log capture of the TAP tests by... |
[
{
"msg_contents": "There was agreement at the PGCon dev meeting that we should put out a\n9.5alpha1 release as soon as possible, to encourage wider testing\n(and that it should be called an \"alpha\", because people aren't convinced\nit's up to beta quality yet). After a little back and forth, that release\nha... |
[
{
"msg_contents": "Some time ago we upgraded from v. 8.4.4 to 9.4.0 and we are seeing errors in some regression tests.\n\nLOCATION: exec_simple_query, postgres.c:887\nERROR: 40001: could not serialize access due to read/write dependencies among transactions\n\nDetail can be one of these:\n DETAIL: Reason c... |
[
{
"msg_contents": "BRIN index WAL is broken in HEAD.\n\nCommit 2c03216d831160bedd72d4, the Revamp the WAL record format, is the\nculprit.\n\nThe easiest way to see this is via streaming replication.\n\nOn master:\n\ncreate table foobar as select * from generate_series(1,10000);\ncreate index on foobar using bri... |
[
{
"msg_contents": "Hi all\n\nAttached is a patch to implement ALTER TABLE ... ALTER CONSTRAINT ...\nSET DEFERRABLE on UNIQUE or PRIMARY KEY constraints.\n\nCurrently only FOREIGN KEY constraints are supported. Others are rejected with:\n\nconstraint \\\"%s\\\" of relation \\\"%s\\\" is not a foreign key constra... |
[
{
"msg_contents": "I configured the dingo and binturong Solaris 10 animals to build 9.3\nsome time ago but apparently they always failed the configure phase.\nTurns out this is caused by thread_test's usage of sched_yield which is\nin librt on Solaris but which is not pulled in by anything on 9.3 and\nearlier o... |
[
{
"msg_contents": "Hi,\n\nCoverity is nitpickingly pointing out the following thing:\n--- a/src/bin/pg_upgrade/controldata.c\n+++ b/src/bin/pg_upgrade/controldata.c\n@@ -402,8 +402,7 @@ get_control_data(ClusterInfo *cluster, bool live_check)\n }\n }\n\n- if (output)\n- ... |
[
{
"msg_contents": "The following bug has been logged on the website:\n\nBug reference: 13473\nLogged by: Marco Nenciarini\nEmail address: marco.nenciarini@2ndquadrant.it\nPostgreSQL version: 9.4.4\nOperating system: all\nDescription: \n\n= Symptoms\n\nLet's have a simple master -> st... |
[
{
"msg_contents": "Hello hackers.\n\nI found a strange thing. I hope it's not on purpose.\n\nExample:\ngit clone git://git.postgresql.org/git/postgresql.git\ncd postgresql\ngit checkout -b remotes/origin/REL9_4_STABLE\ngit merge master\nMANY CONFLICTS\n\nWhy? \n\nThanks. \n-- \nUriy Zhuravlev\nPostgres Professi... |
[
{
"msg_contents": "I looked into bug #13471, which states that we gripe about multiple\noccurrences of the same variable in postgresql.conf + related files.\nNow, that had clearly been fixed some time ago:\n\nAuthor: Fujii Masao <fujii@postgresql.org>\nBranch: master [e3da0d4d1] 2014-08-06 14:49:43 +0900\nBranc... |
[
{
"msg_contents": "Under high load against f7bb7f0625771bc71869cda, I occasionally get:\n\nPANIC: XLogBeginInsert was not called\n\nIt seems to only come from vacuuming.\n\nHere is an example back-trace:\n\n#0 0x0000003dcb632625 in raise () from /lib64/libc.so.6\n#1 0x0000003dcb633e05 in abort () from /lib64... |
[
{
"msg_contents": "I spent much of today going through here:\n\nhttps://wiki.postgresql.org/wiki/PostgreSQL_9.5_Open_Items\n\nHere's what I did:\n\n* Committed patches for four of the items, hopefully resolving those items.\n* Moved three items from \"open\" to either \"resolved\" or a new section\n\"don't need... |
[
{
"msg_contents": "Sometime back on one of the PostgreSQL blog [1], there was\ndiscussion about the performance of drop/truncate table for\nlarge values of shared_buffers and it seems that as the value\nof shared_buffers increase the performance of drop/truncate\ntable becomes worse. I think those are not ofte... |
[
{
"msg_contents": "> On Thu, Jun 25, 2015 at 11:55 PM, Robert Haas <robertmhaas@gmail.com> wrote:\r\n> > On Thu, Jun 25, 2015 at 6:25 PM, Kouhei Kaigai <kaigai@ak.jp.nec.com> wrote:\r\n> >> I have a serious open item reported 1.5 month ago then reminded\r\n> >> several times has not been fixed up yet.\r\n> >>\r... |
[
{
"msg_contents": "I happened to notice these in a backend's memory context dump:\n\n ident parser context: 0 total in 0 blocks; 0 free (0 chunks); 0 used\n hba parser context: 7168 total in 3 blocks; 3920 free (1 chunks); 3248 used\n\nIs there a good reason why these weren't made children of the\nPostmasterC... |
[
{
"msg_contents": "I noticed that in EXEC_BACKEND builds (ie, Windows) the pg_file_settings\nview doesn't act as its author presumably intended. Specifically, it\nreads as empty until/unless the current session processes a SIGHUP event.\nThis is because before that happens, it's depending on having inherited\n... |
[
{
"msg_contents": "convert_string_datum() says:\n\n\t\t/*\n\t\t * Note: originally we guessed at a suitable output buffer size, and\n\t\t * only needed to call strxfrm twice if our guess was too small.\n\t\t * However, it seems that some versions of Solaris have buggy strxfrm\n\t\t * that can write past the spe... |
[
{
"msg_contents": "This patch implements version 1.2 of contrib module pg_trgm.\n\nThis supports the triconsistent function, introduced in version 9.4 of the\nserver, to make it faster to implement indexed queries where some keys are\ncommon and some are rare.\n\nI've included the paths to both upgrade and down... |
[
{
"msg_contents": "Main pgbench logic consists of single file pgbench.c which is 4036\nlines of code as of today. This is not a small number and I think it\nwould be nice if it is divided into smaller files because it will make\nit easier to maintain, add or change features of pgbench. I will come\nup with an ... |
[
{
"msg_contents": "Hi\n\nI am thinking about simplifying a deployment some multiversion PostgreSQL\nextensions, and scripts.\n\nWith current possibilities, we have to use DO statement, what is not\noptimal or possible in some use cases. The implementation of condition\nblock (possible nested) is very simple.\n\... |
[
{
"msg_contents": "Hi,\n\nHow about the attached that adjusts errorcode for the error related to\nchecking the flag bgw_flags in BackgroundWorkerInitializeConnection*()\nfunctions so that it matches the treatment in SanityCheckBackgroundWorker()?\n\ns/ERRCODE_PROGRAM_LIMIT_EXCEEDED/ERRCODE_INVALID_PARAMETER_VAL... |
[
{
"msg_contents": "Hi all,\n\nAs far as I know, there is currently no way to find which column is triggering\nan error on an INSERT or ALTER COLUMN statement. Example:\n\n # create table foo(bar varchar(4), baz varchar(2));\n CREATE TABLE\n # insert into foo values ('foo!', 'ok');\n INSERT 0 1\n ... |
[
{
"msg_contents": "I have been working on to analyze different ways to reduce\nthe contention around ProcArrayLock. I have evaluated mainly\n2 ideas, first one is to partition the ProcArrayLock (the basic idea\nis to allow multiple clients (equal to number of ProcArrayLock partitions)\nto perform ProcArrayEndT... |
[
{
"msg_contents": "Commits b181a919 and arguably c79b6413 added bugs to\nbttext_abbrev_convert() in the process of fixing some others. In the\nmaster branch, bttext_abbrev_convert() can leak memory when the C\nlocale happens to be used and we must detoast, which is unacceptable\nfor about the same reason that i... |
[
{
"msg_contents": "I have a 9.5alpha1 cluster which is locked up. All the user back ends seem\nto be waiting on semop, eventually on WALInsertLockAcquire.\n\nIs there a way to use gdb to figure out who holds the lock they are waiting\nfor?\n\nIt is compiled with both debug and cassert.\n\nI am hoping someone c... |
[
{
"msg_contents": "I was going to rebase my HashAgg patch, and got some conflicts related\nto the grouping sets patch. I could probably sort them out, but I think\nthat may be the tipping point where we want to break up nodeAgg.c into\nnodeSortedAgg.c and nodeHashAgg.c, and probably a common file as well.\n\nTh... |
[
{
"msg_contents": "ClogControlLock contention is high at commit time. This appears to be due\nto the fact that ClogControlLock is acquired in Exclusive mode prior to\nmarking commit, which then gets starved by backends running\nTransactionIdGetStatus().\n\nProposal for improving this is to acquire the ClogContr... |
[
{
"msg_contents": "Hi all,\n(Petr in CC)\n\nCoverity is complaining about the following pointer dereference in\ntablesample_init@tablesample.c:\n+ ExprState *argstate = ExecInitExpr(argexpr, (PlanState *)\nscanstate);\n+\n+ if (argstate == NULL)\n+ {\n+ ... |
[
{
"msg_contents": "Hi all,\n\nCoverity is pointing out that a couple of return value checks are\nmissing for JsonbIteratorNext in jsonfuncs.c. Patch is attached to\nappend (void) to them, as far as I am guessing we want skip the value\niterated on.\nRegards,\n-- \nMichael\n\n\n\n-- \nSent via pgsql-hackers mail... |
[
{
"msg_contents": "Hello Hackers,\n\nFollowing is a proposal for feature to calculate VACUUM progress.\n\nUse Case : Measuring progress of long running VACUUMs to help DBAs make\ninformed decision\nwhether to continue running VACUUM or abort it.\n\nDesign:\n\nA shared preload library to store progress informati... |
[
{
"msg_contents": "Hi all,\n\ntimestamp2tm is called close to 40 times in the backend source code,\nreturning -1 in case of failure. However, there are two places in\ndatetime.c where we do not check for its return value: GetCurrentDateTime\nand GetCurrentTimeUsec.\nThis does not really matter much in practice ... |
[
{
"msg_contents": "Hi all,\n\nIn the category of nitpicky-code-style-issues, FreeSpaceMapTruncateRel\nis doing a NULL-pointer check for something that has been dereferenced\non all the code paths leading to this check.\n(Yes, that's not interesting for common humans, Coverity sees things\nbased on correctness).... |
[
{
"msg_contents": "Barring objections, I'll create the REL9_5_STABLE branch and stamp HEAD\nas 9.6devel sometime this afternoon, maybe around 1800 UTC.\n\n\t\t\tregards, tom lane\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://www.postg... |
[
{
"msg_contents": "Improve 9.5 release notes.\n\n1) Add sgml comments referencing commits. This is useful to search for\n missing items etc.\n\n The comments containing the commit notes are an excerpt from:\n git log --date=short \\\n --pretty='format:%cd [%h] %<(8,trunc)%cN: %<(48,trunc)%s%n%n%w(,4... |
[
{
"msg_contents": "Hi,\n\nWAL-related tools, i.e., pg_archivecleanup, pg_resetxlog and\npg_xlogdump don't seem to properly handle .paritial WAL file.\nI think that we should fix at least pg_archivecleanup, otherwise,\nin the system using pg_archivecleanup to clean up old archived\nfiles, the archived .paritial ... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.