threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "Hello.\n\nBefore the advent of procedures in PostgreSQL 11 that can manage \ntransactions, there could only be one transaction\nin one statement. Hence the end of the transaction also meant the end of \nthe statement. Apparently, this is why\nthe corresponding restriction is described differe... |
[
{
"msg_contents": "Attached patch harmonizes pg_bsd_indent's function parameter names, so\nthat they match the names used in corresponding function definitions.\n\nI have been putting this off because I wasn't sure that the policy\nshould be the same for pg_bsd_indent. Is there any reason to think\nthat this wi... |
[
{
"msg_contents": "Hi,\n\nDavid R and I were discussing vectorisation and microarchitectures and\nwhat you can expect the target microarchitecture to be these days, and\nit seemed like some of our choices are not really very\nforward-looking.\n\nDistros targeting x86-64 traditionally assumed the original AMD64 ... |
[
{
"msg_contents": "Hi All,\r\nFrist of all really appreciate the review of my patch. I've seperated the patch into two: patch_support_tls1.3 for tls1.3 support and patch_support_curves_group for a set of curves list. \r\n\r\n\r\n\r\nTLS1.3 support - patch_support_tls1.3\r\nI agree with Jelte that it's better to... |
[
{
"msg_contents": "Hi Jelte and Daniel,\r\n\r\n\r\nBased on my understanding currently there is no setting that controls the cipher choices used by TLS version 1.3 connections but the default value(HIGH:MEDIUM:+3DES:!aNULL) is used. So if I want to connect to Postgres (eg. Postgres 14) with different TLS versio... |
[
{
"msg_contents": "Hi all,\n\nWhile looking at ways to make pg_stat_statements more scalable and\ndynamically manageable (no more PGC_POSTMASTER for the max number of\nentries), which came out as using a dshash, Andres has mentioned me\noff-list (on twitter/X) that we'd better plug in it to the shmem\npgstats f... |
[
{
"msg_contents": "The purpose of MultiXactMemberFreezeThreshold() is to make the effective \nautovacuum_multixact_freeze_max_age smaller, if the multixact members \nSLRU is approaching wraparound. Per comment there:\n\n> * To prevent that, if more than a threshold portion of the members space is\n> * used, w... |
[
{
"msg_contents": "Hackers,\n\nAnother apparent inconsistency I’ve noticed in jsonpath queries is the treatment of the && and || operators: They can’t operate on scalar functions, only on other expressions. Some examples:\n\ndavid=# select jsonb_path_query('true', '$ && $');\nERROR: syntax error at or near \"&... |
[
{
"msg_contents": "Hi\n\nI am found strange switch:\n\n<--><-->switch (carg->mode)\n<--><-->{\n<--><--><-->case RAW_PARSE_PLPGSQL_EXPR:\n<--><--><--><-->errcontext(\"SQL expression \\\"%s\\\"\", query);\n<--><--><--><-->break;\n<--><--><-->case RAW_PARSE_PLPGSQL_ASSIGN1:\n<--><--><-->case RAW_PARSE_PLPGSQL_ASSI... |
[
{
"msg_contents": "IMHO there are a couple of opportunities for improving the predefined roles\ndocumentation [0]:\n\n* Several of the roles in the table do not have corresponding descriptions\n in the paragraphs below the table (e.g., pg_read_all_data,\n pg_write_all_data, pg_checkpoint, pg_maintain,\n pg_u... |
[
{
"msg_contents": "Hi hackers,\n\n\nI am using gcc version 11.3.0 to compile postgres source code. Gcc complains about the following line:\n\n\n```c\nstrncpy(sqlca->sqlstate, \"YE001\", sizeof(sqlca->sqlstate));\n```\n\n\nwith error as:\n\n\nmisc.c:529:17: error: ‘strncpy’ output truncated before terminating nu... |
[
{
"msg_contents": "Hello!\n\nThe src/backend/access/heap/README.tuplock says about HEAP_XMAX_INVALID bit\nthat \"Any tuple with this bit set does not have a valid value stored in XMAX.\"\n\nFound that FreezeMultiXactId() tries to process such an invalid multi xmax\nand may looks for an update xid in the pg_mult... |
[
{
"msg_contents": "I was performing tests around multixid wraparound, when I ran into this \nassertion:\n\n> TRAP: failed Assert(\"CritSectionCount == 0 || (context)->allowInCritSection\"), File: \"../src/backend/utils/mmgr/mcxt.c\", Line: 1353, PID: 920981\n> postgres: autovacuum worker template0(ExceptionalCo... |
[
{
"msg_contents": "Hello,\n\nI met an assertion failure, and identified the root of the problem, but no\nidea how to fix it.\n\nThe location of the problematic Assert() is at cost_memoize_rescan() to\ncheck 'hit_ratio' is between 0.0 and 1.0.\nThe 'calls' is provided by the caller, and 'ndistinct' is the result... |
[
{
"msg_contents": "Hi!\n\nWhile working on a making multix xact offsets 64-bit [0] I've discovered a\nminor issue. The\nthing is that type 'xid' is used in all macro, but it doesn't correct.\nAppropriate MultiXactId or\nMultiXactOffset should be used, actually.\n\nAnd the second thing, as Heikki Linnakangas po... |
[
{
"msg_contents": "Over at [1] Andres expressed enthusiasm for enabling TAP tests to call \nLibPQ directly via FFI, and there was some support from others as well. \nAttached is a very rough POC for just that.There are two perl modules, \none which wraps libpq (or almost all of it) in perl, and another which \n... |
[
{
"msg_contents": "Hackers,\n\nI noticed that neither `regex_like` nor `starts with`, the jsonpath operators, raise an error when the operand is not a string (or array of strings):\n\ndavid=# select jsonb_path_query('true', '$ like_regex \"^hi\"');\n jsonb_path_query \n------------------\n null\n(1 row)\n\ndavi... |
[
{
"msg_contents": "The blog post here (thank you depesz!):\n\nhttps://www.depesz.com/2024/06/11/how-much-speed-youre-leaving-at-the-table-if-you-use-default-locale/\n\nshowed an interesting result where the builtin provider is not quite as\nfast as \"C\" for queries like:\n\n SELECT * FROM a WHERE t = '...';\... |
[
{
"msg_contents": "Hi Team,\n\nGreetings of the day!!\n\nWe are planning to partition tables using pg_partman. Like we are planning\nfor their backup and restoration process.\n\nGot a few URLs where pg_dump had issues while restoring some data that was\nlost.\n\nkindly guide me the process or steps I need to fo... |
[
{
"msg_contents": "Separating this from the pytest thread:\n\nOn Sat, Jun 15, 2024 at 01:26:57PM -0400, Robert Haas wrote:\n> The one\n> thing I know about that *I* think is a pretty big problem about Perl\n> is that IPC::Run is not really maintained.\n\nI don't see in https://github.com/cpan-authors/IPC-Run/is... |
[
{
"msg_contents": "While my colleague is working on translating\ndoc/src/sgml/ref/create_role.sgml into Japanese, he has found\nfollowing sentence is hard to parse:\n\n The rules for which initial\n role membership options are enabled described below in the\n <literal>IN ROLE</literal>, <literal>ROLE</lit... |
[
{
"msg_contents": "Hi,\n\nassigned patch try to solve issue reported by Mor Lehr (Missing semicolon\nin anonymous plpgsql block does not raise syntax error).\n\nhttps://www.postgresql.org/message-id/CALyvM2bp_CXMH_Gyq87pmHJRuZDEhV40u9VP8rX=CAnEt2wUXg@mail.gmail.com\n\nby introducing a new extra error check. Wit... |
[
{
"msg_contents": "Hi,\n\nWhen connecting with a libpq based client, the TLS establishment ends up like\nthis in many configurations;\n\nC->S: TLSv1 393 Client Hello\nS->C: TLSv1.3 167 Hello Retry Request, Change Cipher Spec\nC->S: TLSv1.3 432 Change Cipher Spec, Client Hello\nS->C: TLSv1.3 1407 Server Hello, A... |
[
{
"msg_contents": "hi.\nsimlar to https://www.postgresql.org/docs/devel//xfunc-c.html#XFUNC-ADDIN-WAIT-EVENTS\n\n\nin https://www.postgresql.org/docs/devel//xfunc-c.html#XFUNC-ADDIN-INJECTION-POINTS\ndo we need to add a sentence like \" An injections points usage example\ncan be found in src/test/modules/inject... |
[
{
"msg_contents": "Hi PostgreSQL hackers,\r\n\r\nFor most access methods in PostgreSQL, the implementation of the access method itself and the implementation of its WAL replay logic are organized in separate source files. However, the HEAP access method is an exception. Both the access method and the WAL repl... |
[
{
"msg_contents": "Hi,\n\nWhile looking at the commit\nhttps://git.postgresql.org/gitweb/?p=postgresql.git;a=commitdiff;h=29d0a77fa6606f9c01ba17311fc452dabd3f793d,\nI noticed that get_old_cluster_logical_slot_infos gets called for even\ntemplate1 and template0 databases. Which means, pg_upgrade executes\nquerie... |
[
{
"msg_contents": "Hi.\n\nThere's the following inconsistency between try_mergejoin_path() and \ncreate_mergejoin_plan().\nWhen clause operator has no commutator, we can end up with mergejoin \npath.\nLater create_mergejoin_plan() will call get_switched_clauses(). This \nfunction can error out with\n\nERROR: c... |
[
{
"msg_contents": "Hackers,\n\nFYI, I wanted to try using PostgreSQL with LLVM on my Mac, but the backend repeatedly crashes during `make check`. I found the same issue in master and REL_16_STABLE. The crash message is:\n\nFATAL: fatal llvm error: Unsupported stack probing method\nserver closed the connection ... |
[
{
"msg_contents": "Hello,\n\nWhen currently trying to lock a virtual tuple the returned error\nwill be a misleading `could not read block 0`. This patch adds a\ncheck for the tuple table slot being virtual to produce a clearer\nerror.\n\nThis can be triggered by extensions returning virtual tuples.\nWhile this ... |
[
{
"msg_contents": "Hi,\n\nTo investigate an unrelated issue, I set up key logging in the backend (we\nprobably should build that in) and looked at the decrypted data. And noticed\nthat just after TLS setup finished the server sends three packets in a row:\n\nC->S: TLSv1.3: finished\nC->S: TLSv1.3: application ... |
[
{
"msg_contents": "Hi,\n\nAs I mentioned in my talk at 2024.pgconf.dev, I think that the biggest\nproblem with autovacuum as it exists today is that the cost delay is\nsometimes too low to keep up with the amount of vacuuming that needs\nto be done. I sketched a solution during the talk, but it was very\ncompli... |
[
{
"msg_contents": "AtCommit_Memory and friends have done $SUBJECT for at least a couple\nof decades, but in the wake of analyzing bug #18512 [1], I'm feeling\nlike that's a really bad idea. There is too much code running\naround the system that assumes that it's fine to leak stuff in\nCurrentMemoryContext. If... |
[
{
"msg_contents": "This commit added enable_group_by_reordering:\n\n\tcommit 0452b461bc4\n\tAuthor: Alexander Korotkov <akorotkov@postgresql.org>\n\tDate: Sun Jan 21 22:21:36 2024 +0200\n\t\n\t Explore alternative orderings of group-by pathkeys during optimization.\n\t\n\t When evaluating a query with a... |
[
{
"msg_contents": "Hi all,\n\nOn HEAD, xlog.c has the following comment, which has been on my own\nTODO list for a couple of weeks now:\n\n * TODO: With a bit of extra work we could just start with a pgstat file\n * associated with the checkpoint redo location we're starting from.\n\nPlease find a patch... |
[
{
"msg_contents": "Hi hackers,\n\nWhile reviewing another thread that proposes to include \"generated\ncolumns\" support for logical replication [1] I was looking for any\nexisting PostgreSQL documentation on this topic.\n\nBut, I found almost nothing about it at all -- I only saw one aside\nmention saying that... |
[
{
"msg_contents": "Hi,\r\nThanks a lot for the review.\r\nIndeed the original ssl_ecdh_curve is used to set a single value of curve name. If we want to change it to indicate a list of curve names, is there any rule for naming in Postgres? like ssl_curve_groups?\r\n\r\n\r\n\r\n \r\nOriginal... |
[
{
"msg_contents": "Under the topic of getting rid of thread-unsafe functions in the backend \n[0], here is a patch series to deal with strtok().\n\nOf course, strtok() is famously not thread-safe and can be replaced by \nstrtok_r(). But it also has the wrong semantics in some cases, because \nit considers adja... |
[
{
"msg_contents": "The CompilerWarnings task on Cirrus CI does not catch warnings in C++ \ncode. It tries to make warnings fatal by passing COPT='-Werror', but \nthat does not apply to C++ compilations.\n\nI suggest that we just add COPT to CXXFLAGS as well. I think passing \n-Werror is just about the only re... |
[
{
"msg_contents": "I have this patch series that fixes up the types of the new incremental \nJSON API a bit. Specifically, it uses \"const\" throughout so that the \ntop-level entry points such as pg_parse_json_incremental() can declare \ntheir arguments as const char * instead of just char *. This just \nwor... |
[
{
"msg_contents": "Hi everyone,\n\nI've noticed that truncating mapped catalogs causes the server to\ncrash due to an assertion failure. Here are the details:\n\nExecuting below commands:\n\n-- set allow_system_table_mods TO on;\n-- truncate table pg_type;\n\nResults into a server crash with below backtrace:\n\... |
[
{
"msg_contents": "Hi\n\nFurther to my previous report [1] about zlib detection not working with\nMeson on Windows, I found it's similarly or entirely broken for the\nmajority of other dependencies, none of which are tested on the buildfarm\nas far as I can see.\n\nFor convenience, I've put together a number of... |
[
{
"msg_contents": "Hi,\r\n\r\nPostgreSQL 17 Beta 2 is planned to be release on June 27, 2024. Please \r\ncontinue your hard work on closing out open items[1] ahead of the \r\nrelease and have the fixes targeted for the release committed by June \r\n22, 2024.\r\n\r\nThanks!\r\n\r\nJonathan\r\n\r\n[1] https://wik... |
[
{
"msg_contents": "Hi.\n\nNow when planner finds suitable pathkeys in \ngenerate_orderedappend_paths(), it uses them, even if explicit sort of \nthe cheapest child path could be more efficient.\n\nWe encountered this issue on partitioned table with two indexes, where \none is suitable for sorting, and another i... |
[
{
"msg_contents": "I noticed that the \"check\" variable, which is used for \"pg_upgrade\n--check\", is commented as follows:\n\n\tbool\t\tcheck;\t\t\t/* true -> ask user for permission to make\n\t\t\t\t\t\t\t\t * changes */\n\nThis comment was first added when pg_upgrade was introduced (commit\nc2e9b2f288), bu... |
[
{
"msg_contents": "Hello hackers,\n\nCurrently, obtaining the Access Control List (ACL) for a database object\nrequires querying specific pg_catalog tables directly, where the user\nneeds to know the name of the ACL column for the object.\n\nConsider:\n\n```\nCREATE USER test_user;\nCREATE USER test_owner;\nCRE... |
[
{
"msg_contents": "Hi,\n\nI tried running master under valgrind on 64-bit ARM (rpi5 running debian\n12.5), and I got some suspicous reports, all related to the radixtree\ncode used by tidstore. I'm used to valgrind on arm sometimes reporting\nharmless issues, but this seems like it might be an actual issue.\n\n... |
[
{
"msg_contents": "When doing performance hacking on pg_upgrade it's often important to see\nindividual runtimes to isolate changes. I've written versions of the attached\npatch numerous times, and I wouldn't be surprised if others have done the same.\nIs there any interest in adding something like the attache... |
[
{
"msg_contents": "The order of json related aggregate functions in the docs is currently \nlike this:\n\n[...]\njson_agg\njson_objectagg\njson_object_agg\njson_object_agg_strict\njson_object_agg_unique\njson_arrayagg\njson_object_agg_unique_strict\nmax\nmin\nrange_agg\nrange_intersect_agg\njson_agg_strict\n[..... |
[
{
"msg_contents": "While working on an idea from another thread [0], I noticed that each of\nmax_connections, max_worker_process, max_autovacuum_workers, and\nmax_wal_senders have a check hook that verifies the sum of those GUCs does\nnot exceed a certain value. Then, in InitializeMaxBackends(), we do the\nsam... |
[
{
"msg_contents": "https://postgr.es/m/20240512232923.aa.nmisch@google.com wrote:\n> Separable, nontrivial things not fixed in the attached patch stack:\n\n> - Trouble is possible, I bet, if the system crashes between the inplace-update\n> memcpy() and XLogInsert(). See the new XXX comment below the memcpy()... |
[
{
"msg_contents": "The pg_combinebackup --clone option currently doesn't work at all. Even \non systems where it should it be supported, you'll always get a \"file \ncloning not supported on this platform\" error.\n\nThe reason is this checking code in pg_combinebackup.c:\n\n#if (defined(HAVE_COPYFILE) && defi... |
[
{
"msg_contents": "\nHi, hackers\n\nWhen I read [1], I think the \"counted_by\" attribute may also be valuable for\nPostgreSQL.\n\nThe 'counted_by' attribute is used on flexible array members. The argument for\nthe attribute is the name of the field member in the same structure holding\nthe count of elements in... |
[
{
"msg_contents": "Hi,\n\nWhile running valgrind on 32-bit ARM (rpi5 with debian), I got this\nreally strange report:\n\n\n==25520== Use of uninitialised value of size 4\n==25520== at 0x94A550: wrapper_handler (pqsignal.c:108)\n==25520== by 0x4D7826F: ??? (sigrestorer.S:64)\n==25520== Uninitialised value... |
[
{
"msg_contents": "Hello Team,\nGood Day,\n\nI have been working on adding a CustomScanState object in the executor\nstate in my project. As part of CustomScanState, I execute queries and\nstore their results in the Tuplestorestate object. After storing all tuples\nin the Tuplestorestate, I retrieve each tuple ... |
[
{
"msg_contents": "Hi,\n\nKonstantin and I found an MVCC bug with:\n\n- a prepared transaction,\n- which has a subtransaction,\n- on a hot standby,\n- after starting the standby from a shutdown checkpoint.\n\nSee the test case in the attached patch to demonstrate this. The last \nquery in the new test returns i... |
[
{
"msg_contents": "Hi,\n\nI'm working to start a mentoring program where code contributors can\nbe mentored by current committers. Applications are now open:\nhttps://forms.gle/dgjmdxtHYXCSg6aB7\n\nNine committers have volunteered to mentor one person each; hence, the\nanticipated number of acceptances is less ... |
[
{
"msg_contents": "While working on skip scan, I stumbled upon a bug on HEAD. This is an\nissue in my commit 5bf748b8, \"Enhance nbtree ScalarArrayOp execution\".\nThe attached test case (repro_wrong_prim.sql) causes an assertion\nfailure on HEAD. Here's the stack trace:\n\nTRAP: failed Assert(\"so->keyData[ops... |
[
{
"msg_contents": "Hi,\n\nIf vacuum fails to remove a tuple with xmax older than\nVacuumCutoffs->OldestXmin and younger than\nGlobalVisState->maybe_needed, it will ERROR out when determining\nwhether or not to freeze the tuple with \"cannot freeze committed\nxmax\".\n\nIn back branches starting with 14, failing... |
[
{
"msg_contents": "FYI, looking at the release notes, I see 15 GUC variables added in this\nrelease, and two removed. That 15 number seemed unusually high so I\nthought I would report it.\n\n-- \n Bruce Momjian <bruce@momjian.us> https://momjian.us\n EDB https://e... |
[
{
"msg_contents": "\nHi,\n\nI relies on some compiler's check to reduce some simple coding issues, I\nuse clang 18.1.6 for now. however \"CFLAGS='-Wall -Werror ' ./configure\"\nwould fail, and if I run ' ./configure' directly, it is OK. I'm not sure\nwhy it happens. More details is below:\n\n(master)> echo $CC\... |
[
{
"msg_contents": "hi.\n-------------\n9.16.2.1.1. Boolean Predicate Check Expressions\nAs an extension to the SQL standard, a PostgreSQL path expression can\nbe a Boolean predicate, whereas the SQL standard allows predicates\nonly within filters. While SQL-standard path expressions return the\nrelevant element... |
[
{
"msg_contents": "Hi,\n\n\n\nRegarding the multicolumn B-Tree Index, I'm considering\n\nif we can enhance the EXPLAIN output. There have been requests\n\nfor this from our customer.\n\n\n\nAs the document says, we need to use it carefully.\n\n> The exact rule is that equality constraints on leading columns,\n\... |
[
{
"msg_contents": "Hi hackers,\nCc people involved in the original thread[1].\n\nI am starting a new thread to share and discuss the implementation of\nconflict detection and logging in logical replication, as well as the\ncollection of statistics related to these conflicts.\n\nIn the original conflict resoluti... |
[
{
"msg_contents": "In libpq, in the code of the lo_import function, for each piece of a file\nof 8KB in size, lo_write is called, which greatly slows down the work of\nlo_import because lo_write sends a request and waits for a response. The\nsize of 8KB is specified in define LO_BUFSIZE which changed from 1KB ... |
[
{
"msg_contents": "After initially trying to add trace support for\nStartupMessage/SSLRequest/GSSENCRequest[1] I realized there were many\nmore cases where PQtrace was not correctly tracing messages, or not\neven tracing them at all. These patches fix all of the issues that I\nwas able to find.\n\n0001 is some ... |
[
{
"msg_contents": "Dear Hackers,\n\nThis is a follow-up thread for pg_createsubscriber [1]. I started a new thread\nsince there is no activity around here.\n\n## Problem\n\nAssuming that there is a cascading replication like below:\n\nnode A --(logical replication)--> node B --(streaming replication)--> node C\... |
[
{
"msg_contents": "I recently got write access to the cfbot repo[1] and machine from\nThomas. And I deployed a few improvements this week. The most\nsignificant one is that it is now much easier to use GitHub as part of\nyour patch review workflow.\n\nOn the cfbot website[2] there's now a \"D\" (diff) link next... |
[
{
"msg_contents": "The release notes have this item:\n\n\tAllow specification of physical standbys that must be synchronized\n\tbefore they are visible to subscribers (Hou Zhijie, Shveta Malik)\n\n\tThe new server variable is standby_slot_names. \n\nIs standby_slot_names an accurate name for this GUC? It seems... |
[
{
"msg_contents": "While investigating a bug over on [1], I found that\nvacuum_set_xid_limits() is calculating freezeLimit in an unsafe way on\nat least Postgres 14 and 15.\n\n limit = *oldestXmin - freezemin;\n safeLimit = ReadNextTransactionId() - autovacuum_freeze_max_age;\n if (TransactionIdPrecede... |
[
{
"msg_contents": "Hello hackers,\n\nThis week dodo started failing on the 008_fsm_truncation test sporadically\ndue to pg_ctl timeout. For example, [1] (REL_14_STABLE):\n### Starting node \"standby\"\n# Running: pg_ctl -D \n/media/pi/250gb/proj/bf2/v17/buildroot/REL_14_STABLE/pgsql.build/src/test/recovery/tmp_... |
[
{
"msg_contents": "Hackers,\n\nThe treatment of timestamptz (and timetz) values with offsets that include seconds seems a bit inconsistent. One can create such timestamps through the input function:\n\ndavid=# select '2024-06-22T12:35:00+02:30:15'::timestamptz;\n timestamptz \n-----------------------... |
[
{
"msg_contents": "I was unable to parse a comment in src/backend/parser/gram.y around line 11364:\n\n/*\n * As func_expr but does not accept WINDOW functions directly (they\n * can still be contained in arguments for functions etc.)\n * Use this when window expressions are not allowed, so to disambiguate\n * t... |
[
{
"msg_contents": "Hi.\n\nIn src/include/access/xlogbackup.h, the field *name*\nhas one byte extra to store null-termination.\n\nBut, in the function *do_pg_backup_start*,\nI think that is a mistake in the line (8736):\n\nmemcpy(state->name, backupidstr, strlen(backupidstr));\n\nmemcpy with strlen does not copy... |
[
{
"msg_contents": "Is there a desire to have a division operator / that takes dividend\nand divisor of types interval, and results in a quotient of type\ndouble precision.\n\nThis would be helpful in calculating how many times the divisor\ninterval can fit into the dividend interval.\n\nTo complement this divis... |
[
{
"msg_contents": "Hello hackers,\n\nAs recent caiman failures ([1], [2], ...) show, plperl.sql is incompatible\nwith Perl 5.40. (The last successful test runs took place when cayman\nhad Perl 5.38.2 installed: [3].)\n\nFWIW, I've found an already-existing fix for the issue [4] and a note\ndescribing the change... |
[
{
"msg_contents": "hi.\nthe following two queries should return the same result?\n\nSELECT * FROM JSON_query (jsonb 'null', '$' returning jsonb);\nSELECT * FROM JSON_value (jsonb 'null', '$' returning jsonb);\n\nI've tried a patch to implement it.\n(i raised the issue at\nhttps://www.postgresql.org/message-id/C... |
[
{
"msg_contents": "InjectionPointRun() acquires InjectionPointLock, looks up the hash \nentry, and releases the lock:\n\n> \tLWLockAcquire(InjectionPointLock, LW_SHARED);\n> \tentry_by_name = (InjectionPointEntry *)\n> \t\thash_search(InjectionPointHash, name,\n> \t\t\t\t\tHASH_FIND, &found);\n> \tLWLockRelease... |
[
{
"msg_contents": "Hello hackers,\n\nThis patch is based on a suggestion from a separate thread [1]:\n\nOn Mon, Jun 24, 2024, at 01:46, Michael Paquier wrote:\n> Rather unrelated to this patch, still this patch makes the situation\n> more complicated in the docs, but wouldn't it be better to add ACL as\n> a ter... |
[
{
"msg_contents": "Hi,\n\nWhile doing some additional testing of (incremental) backups, I ran into\na couple regular failures. After pulling my hair for a couple days, I\nrealized the issue seems to affect regular backups, and incremental\nbackups (which I've been trying to test) are likely innocent.\n\nI'm usi... |
[
{
"msg_contents": "Hello,\n\nHope you are doing well.\n\nI've been playing a bit with the incremental backup feature which might\ncome as\npart of the 17 release, and I think I hit a possible bug in the WAL\nsummarizer\nprocess.\n\nThe issue that I face refers to the summarizer process getting into a hung\nstat... |
[
{
"msg_contents": "Xcode 16 beta was released on 2024-06-10 and ships with macOS SDK 15.0\n[1]. It appears PostgreSQL does not compile due to typedef redefiniton\nerrors in the regex library:\n\n/tmp/Xcode-beta.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/clang\n-Wall -Wmissing-prototypes ... |
[
{
"msg_contents": "Greetings,\n\nWhile testing pgjdbc I noticed the following\n\npgdb-1 | Will execute command on database postgres:\npgdb-1 | SELECT pg_drop_replication_slot(slot_name) FROM\npg_replication_slots WHERE slot_name = 'replica_one';\npgdb-1 | DROP USER IF EXISTS replica_one;\npgd... |
[
{
"msg_contents": "Prior to macOS SDK 15, there were include guards in\n$SDK_ROOT/usr/include/xlocale/_regex.h:\n\n #ifndef _REGEX_H_\n #include <_regex.h>\n #endif // _REGEX_H_\n #include <_xlocale.h>\n\nIn macOS SDK 15.5, these include guards are gone:\n\n #include <_regex.h>\n #include <_xlocale.h>\n\n... |
[
{
"msg_contents": "Hey!\n\nLots of SQL/JSON threads going about. This one is less about technical\ncorrectness and more about usability of the documentation. Though in\nwriting this I am finding some things that aren't quite clear. I'm going\nto come back with those on a follow-on post once I get a chance to... |
[
{
"msg_contents": "for json format, add a new option to let the explain json output also\ninclude the actual query string.\n\nit can make json usage more convenient.\nNow you only need to grab the json output, no need to\ncollect another explain statement and extract the actual query from\nthe explain statement... |
[
{
"msg_contents": "I’m a complete novice, although I’ve dipped my toes in Admin waters a\ncouple of times in my many years of using Linux.\n\nCan anyone recommend a good book on installing Postgres on multiple,\nconnected multiuser systems, tuning it, managing users, backups, updated,\netc.\n\nA cookbook/checkl... |
[
{
"msg_contents": "Hi,\n\nDuring backend initialisation, pgStat DSA is attached using\ndsa_attach_in_place with a NULL segment. The NULL segment means that\nthere's no callback to release the DSA when the process exits.\npgstat_detach_shmem only calls dsa_detach which, as mentioned in the\nfunction's comment, d... |
[
{
"msg_contents": "Hello,\n\nIt's possible I'm the only one who's been in this situation, but I've\nmultiple times found myself explaining to a user how column DEFAULT\nexpressions work: namely how the quoting on an expression following\nthe keyword DEFAULT controls whether or not the expression is\nevaluated a... |
[
{
"msg_contents": "Hi guys,\n\nThis is inspired by this TODO list\nhttps://wiki.postgresql.org/wiki/Todo#CLUSTER and by pg_repack and\npg_freeze projects.\nMy final goal is to create an extension that does direct data-file to\ndata-file transfer (no intermediate tables, no triggers) with no blocking\nat all in ... |
[
{
"msg_contents": "Hi,\n\nProblem #1: we're still using Ventura, but Cirrus has started doing this:\n\nOnly ghcr.io/cirruslabs/macos-runner:sonoma is allowed. Automatically upgraded.\n\nIt doesn't do it to cfbot, which runs macOS stuff on PGDG-hosted Mac\nMinis, but it does it to regular users who use free comp... |
[
{
"msg_contents": "psql (PostgreSQL) 17beta2 (Debian 17~beta2-1.pgdg+~20240625.1534.g23c5a0e)\n\nFailed to retrieve data from the server..\n\n\nretrieve information from database:\n\ncolumn \"daticulocale\" does not exist\nLINE 5: datconnlimit, daticulocale, daticurules, datcollversion,\n^\nHINT: Perhaps you me... |
[
{
"msg_contents": "Hi Team,\n\nCurrently we are using SHA-256 default for password_encryption in our postgresql deployments. Is there any active work being done for adding additional hashing options like PBKDF2, HKDF, SCRYPT or Argon2 password hashing functions, either of which is only accepted as a algorithms ... |
[
{
"msg_contents": "Hi All,\n\nIf pgindent encounters an error while applying pg_bsd_indent on a file, it\nreports an error to stderr but does not exit with non-zero status.\n\n$ src/tools/pgindent/pgindent .\nFailure in ./src/backend/optimizer/util/relnode.c: Error@2412: Stuff\nmissing from end of file\n\n$ ech... |
[
{
"msg_contents": "Hi,\n\nThe function 'libpqrcv_check_conninfo()' returns 'void', but the comment \nabove says that the function returns true or false.\nI've attached a patch to modify the comment.\n\nRegard,\nRintaro Ikeda",
"msg_date": "Wed, 26 Jun 2024 21:52:55 +0900",
"msg_from": "ikedarintarof <ik... |
[
{
"msg_contents": "Hi,\n\nWhile designing an improvement for the cost sort model, I discovered \nthat the query plan can vary if we slightly change the query text \nwithout pushing semantic differences. See the example below:\n\nCREATE TABLE test(x integer, y integer,z text);\nINSERT INTO test (x,y) SELECT x, 1... |
[
{
"msg_contents": "I initially ran into this while trying to reproduce the recent\nreports of trouble with LLVM 14 on ARM. However, it also reproduces\nwith LLVM 17 on x86_64, and I see no reason to think it's at all\narch-specific. I also reproduced it in back branches (only tried\nv14, but it's definitely n... |
[
{
"msg_contents": "Here is a patch for using gmtime_r() and localtime_r() instead of \ngmtime() and localtime(), for thread-safety.\n\nThere are a few affected calls in libpq and ecpg's libpgtypes, which are \nprobably effectively bugs, because those libraries already claim to be \nthread-safe.\n\nThere is one ... |
[
{
"msg_contents": "Attached is a POC patch that adds skip scan to nbtree. The patch\nteaches nbtree index scans to efficiently use a composite index on\n'(a, b)' for queries with a predicate such as \"WHERE b = 5\". This is\nfeasible in cases where the total number of distinct values in the\ncolumn 'a' is reaso... |
[
{
"msg_contents": "Hi,\n\nI have just tested PG17 beta1 with the E-Maj solution I maintain. The \nonly issue I found is the removal of the adminpack contrib.\n\nIn the emaj extension, which is the heart of the solution, and which is \nwritten in plpgsql, the code just uses the pg_file_unlink() function to \naut... |
[
{
"msg_contents": "I have a question regarding postgresql waiting for the client. I queried\nthe pg_stat_activity because I noticed a connection that had not been\nreleased for days!!! I saw that the wait_event was ClientRead and the query\nwas ROLLBACK. What the server is waiting for from the client? It is a\n... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.