threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "On December 13th this change to context creation was committed, which broke\nPostGIS trunk compile against PostgreSQL 11 head. \nhttps://git.postgresql.org/gitweb/?p=postgresql.git;a=commit;h=9fa6f00b1308d\nd10da4eca2f31ccbfc7b35bb461\n\nTicketed in PostGIS here:\nhttps://trac.osgeo.org/post... |
[
{
"msg_contents": "(cross-posting admin and hackers)\n\nHello,\n\nI wonder if I'm alone in my wish to have a way for estimating how much\nmaintenance work memory would suffice to allocate for a session when\ncreating an index and avoid spilling to disk?\n\nRecently I had to re-create some indexes on a 9.6 serve... |
[
{
"msg_contents": "AFAICT, base backups running on the replication protocol are always\nreported as \"idle\" in pg_stat_activity. This seems to have been an\noversight in the \"include walsender backends in pg_stat_activity\" in 10,\nwhich does include it for walsenders in general, just not for the ones\nsendin... |
[
{
"msg_contents": "Hi.\n\nI have a patch that rearranges the code around partition tuple-routing,\nsuch that allocation of per-partition objects (ResultRelInfo,\nTupleConversionMap, etc.) is delayed until a given partition is actually\ninserted into (i.e., a tuple is routed to it). I can see good win for\nnon-... |
[
{
"msg_contents": "Hi,\n\nWhile doing some wraparound debugging, I saw the hint regarding upcoming\nwraparound did not include the problem of having a stale replication\nslot (which I'm actually using to force wraparound issues).\n\nI remember a few discussions where a stale replication slot was actually\nthe c... |
[
{
"msg_contents": "Few day ago a faced a problem: Pl/PgSQL procedure works slower when running\nin parallel threads. I found the correlation between number of assignments\nin procedure code and performance. I decided to write the simple benchmark\nprocedures and perform some test on PostgreSQL 9.6.5 database in... |
[
{
"msg_contents": "Hello, hackers!\n\nThis is the fourth version of the patch for pgbench. As in the previous \none, most of the changes were made thanks to the comments of Fabien \nCoelho. Here I propose not to abort the client in pgbench in case of \nvarious errors including unsuccessful meta commands. As in ... |
[
{
"msg_contents": "Hello, hackers!\n\nThis is the fourth version of the patch for pgbench. As in the previous \none, most of the changes were made thanks to the comments of Fabien \nCoelho. Here I propose not to abort the client in pgbench in case of \nvarious errors including unsuccessful meta commands. As in ... |
[
{
"msg_contents": "ISTM that some of the portal-related memory context naming is a bit\nantiquated and at odds with current terminology. In this patch, I\npropose to rename PortalMemory to TopPortalContext and rename\nPortal->heap to Portal->portalContext, and then clean up some\nsurrounding APIs.\n\n-- \nPete... |
[
{
"msg_contents": "I'm looking at ways to get plpgsql expression evaluation to go faster,\nand one thing I'm noticing is the rather large overhead of going through\nExecEvalParamExtern and plpgsql_param_fetch to get to the useful work\n(exec_eval_datum). We've ameliorated that for DTYPE_VAR variables\nby keepi... |
[
{
"msg_contents": "Hi,\n\nIn [1] I'd discovered a only mildly related bug while reading code to\nmake sure my fix [2] et al was correct.\n\nQuoting a couple messages by myself:\n> Staring at the vacuumlazy hunk I think I might have found a related bug:\n> heap_update_tuple() just copies the old xmax to the new ... |
[
{
"msg_contents": "Here is a patch to improves how PL/Python deals with very large values\nof SPI_processed. The previous code converts anything that does not fit\ninto a C long into a Python float. But Python long has unlimited\nprecision, so we should be using that instead. And in Python 3, int and\nlong a... |
[
{
"msg_contents": "Hi hackers,\n\nThis is Haisheng Yuan from Greenplum Database.\n\nWe had some query in production showing that planner favors seqscan over\nbitmapscan, and the execution of seqscan is 5x slower than using\nbitmapscan, but the cost of bitmapscan is 2x the cost of seqscan. The\nstatistics were u... |
[
{
"msg_contents": "I saw this just now:\n\nTRAP: FailedAssertion(\"!(TransactionIdPrecedesOrEquals(safeXid, \nsnap->xmin))\", File: \"snapbuild.c\", Line: 580)\n\nwhile running 50 cascading instances on a single machine.\n\nselect version():\nPostgreSQL 11devel_HEAD_20171219_2158_7d3583ad9ae5 on \nx86_64-pc-lin... |
[
{
"msg_contents": "Hi hackers,\n\nI just saw some trailing garbage in my log file emanating from a\nparallel worker when my query happened to be a BUFFERALIGNed length\n(in this case 64 characters). Did commit 4c728f382970 forget to make\nsure the null terminator is copied? See attached proposed fix.\n\n-- \n... |
[
{
"msg_contents": "Hi hackers,\n\nFollowing up the recent discussion on zipfian distribution I was trying\nto reproduce some YCSB-like workloads. As this paper [1] describes, YCSB\nuses zipfian distribution to generate keys in order simulate intensive\nload on small number of records as it happens in real world... |
[
{
"msg_contents": "I've recently come across a query that produces different plans depending\non whether it's parameterized or not. The parameterized query takes ~50ms\nwhile the non-parameterized query takes ~4s. The issue seems to be that the\nquery contains a STABLE function (uid_langvar) whose parameter is ... |
[
{
"msg_contents": "I wonder if Postgres community is interested in supporting time travel \nqueries in PostgreSQL (something like AS OF queries in Oracle: \nhttps://docs.oracle.com/cd/B14117_01/appdev.101/b10795/adfns_fl.htm).\nAs far as I know something similar is now developed for MariaDB.\n\nIt seems to me t... |
[
{
"msg_contents": "Folks,\n\nIt's possible to arrange for schemas to precede pg_catalog and\ninformation_schema in a search_path setting, and when that's done,\nhilarity can ensue, especially when someone has created functions with\nidentical signatures but non-identical behavior. People who do that\nshould pr... |
[
{
"msg_contents": "Hello All\n\nAnyone know if there is a patch that changes the PostgreSQL cost model in\nsome way?\nI'm working with an tuning technique, based in hybrid storage environments\n(HDDs and SSDs).\n\nI need to know if exist any patches that allow postgresql to differentiate,\nfor example, I/O cost... |
[
{
"msg_contents": "I happened to notice that while writing this in plpgsql\nwill consider a parallel plan:\n\n\tselect count(*) into s from tenk1 where ten = x;\n\nwriting this will not:\n\n\ts := count(*) from tenk1 where ten = x;\n\nIs that intentional? Seems to me these cases ought to be\ntreated the same.\... |
[
{
"msg_contents": "PGCon 2018 will be on 29 May - 1 June 2018 at University of Ottawa.\n\n* 29-30 May (Tue-Wed) tutorials\n* 30 May (Wed) The Unconference\n* 31 May - 1 June (Thu-Fri) talks - the main part of the conference\n\nSee http://www.pgcon.org/2018/\n\nWe are now accepting proposals for the main part of... |
[
{
"msg_contents": "Hi,\n\nwith below query I am getting unexpected output. here !<space>= is behaving\nas =\nis this expected behaviour?\n\npostgres=# create table tbl (col_a int, col_b int);\nCREATE TABLE\npostgres=# insert into tbl values (1,2);\nINSERT 0 1\npostgres=# insert into tbl values (2,1);\nINSERT 0 ... |
[
{
"msg_contents": "Hi all\n\nThe pg_indent on www is outdated, and doesn't understand --version .\n\nhttps://www.postgresql.org/ftp/dev/\n\nIt might make sense to add a README there that mentions it's obsolete and\nnow lives at https://git.postgresql.org/git/pg_bsd_indent.git (or just \"see\nsrc/tools/pgindent/... |
[
{
"msg_contents": "Add parallel-aware hash joins.\n\nIntroduce parallel-aware hash joins that appear in EXPLAIN plans as Parallel\nHash Join with Parallel Hash. While hash joins could already appear in\nparallel queries, they were previously always parallel-oblivious and had a\npartial subplan only on the oute... |
[
{
"msg_contents": "Fix permissions check on pg_stat_get_wal_senders\n\nCommit 25fff40798fc4ac11a241bfd9ab0c45c085e2212 introduced the\npossibility for the pg_read_all_stats to have access to all pg_stat_*\nviews.\nIn the discussion, the pg_stat_replication and pg_stat_wal_receiver\nviews were also considered to... |
[
{
"msg_contents": "Hi Hackers,\n\nWith the multi host connection string feature, it is possible to specify\nmultiple\nhosts in primary_conninfo that is used in streaming replication to make sure\nthat WAL streaming is not affected.\n\nCurrently there is no information that is available in the standby node to\nf... |
[
{
"msg_contents": "Greetings all,\n\nSince a few different people asked about it (and replies have been\nposted to those threads), we figured it made sense to also post here\nabout it. PGLister's (finally) been reviewed by a couple of people and\nhopefully we didn't miss anything vital, but we've opened up the... |
[
{
"msg_contents": "Hi Robert, Todd, All,\n\nI think both I and commit e9baa5e9fa147e are confused.\n\nMantid started to fail with the parallel hash join commit, with the following\nassert failure [1]:\n#2 0x00000000008698e7 in ExceptionalCondition (conditionName=conditionName@entry=0x8efee8 \"!(CurrentTransact... |
[
{
"msg_contents": "Get rid of copy_partition_key\n\nThat function currently exists to avoid leaking memory in\nCacheMemoryContext in case of trouble while the partition key is being\nbuilt, but there's a better way: allocate everything in a memcxt that\ngoes away if the current (sub)transaction fails, and once ... |
[
{
"msg_contents": "-Hackers,\n\n\nAs I continue to test Logical Replication and its native capabilities I \nhave found the following:\n\n\nIt appears it is not possible to do this:\n\npublisher->public.foo replicates to subscriber->private.foo\n\nThis one seems like a relatively simple fix (theoretically) and I... |
[
{
"msg_contents": "Hi hackers,\n\nWhile debugging a build farm assertion failure after commit 18042840,\nand with the assumption that the problem is timing/scheduling\nsensitive, I tried hammering the problem workload on a few different\nmachines and noticed that my slow 2-core test machine fairly regularly\ngo... |
[
{
"msg_contents": "I've been using a base backup script that takes special care to\nhave pg_control be the last file it grabs. And I see that\nbasebackup.c takes similar care:\n\nhttps://git.postgresql.org/gitweb/?p=postgresql.git;a=blobdiff;f=src/backend/replication/basebackup.c;h=81203c9f5ac9dbf38da09e1ff55b2... |
[
{
"msg_contents": "Hi,\n\nAttached a patch for fixing a typo in autoprewarm.c. I'm sorry if I'm mistaken.\n\ns/withs/with/\n\nRegards,\n\n--\nMasahiko Sawada\nNIPPON TELEGRAPH AND TELEPHONE CORPORATION\nNTT Open Source Software Center",
"msg_date": "Fri, 22 Dec 2017 14:30:56 +0900",
"msg_from": "Masahik... |
[
{
"msg_contents": "Hi,\n\nI have an issue that COPY from a FIFO, which has no writers, could not be\ncanceled, because COPY invokes AllocateFile() -> fopen() -> blocking open().\n\n```\n[postgres@s1 ~]$ mkfifo /tmp/test0\n[postgres@s1 ~]$ /usr/local/pgsql/bin/psql test\npsql (11devel)\nType \"help\" for help.\n... |
[
{
"msg_contents": "try_partial_hashjoin_path() passes constant true to for the parallel_hash\nargument of initial_cost_hashjoin(). Shouldn't it instead pass the\nparallel_hash argument that it receives?\n\nThis is related to commit 1804284042e659e7d16904e7bbb0ad546394b6a3.\n\n-- \nAntonin Houska\nCybertec Schön... |
[
{
"msg_contents": "Hi, hackers!\n\n\nI want to propose the patch that allows to define custom signals and \ntheir handlers on extension side. It is based on ProcSignal module, \nnamely it defines the fixed set (number is specified by constant) of \ncustom signals that could be reserved on postgres initializatio... |
[
{
"msg_contents": "While my experiments with pthreads version of Postgres I find out that I \ncan not create more than 100k backends even at the system with 4Tb of RAM.\nI do not want to discuss now the idea of creating so large number of \nbackends - yes, most of the real production systems are using pgbouncer... |
[
{
"msg_contents": "Few observations in Parallel Append commit (ab727167)\n\n1.\n+++ b/src/include/nodes/execnodes.h\n@@ -21,6 +21,7 @@\n #include \"lib/pairingheap.h\"\n #include \"nodes/params.h\"\n #include \"nodes/plannodes.h\"\n+#include \"storage/spin.h\"\n..\n\nThere doesn't seem to be any need for includ... |
[
{
"msg_contents": "Hello,\n\nI'm giving this patch its own thread for mental sanity, but this is\nessentially what already posted in [1], plus some doc fixes. This patch\ndepends on the main \"local partitioned indexes\" in that thread, last\nversion of which is at [2].\n\nI also added a mechanism to set the c... |
[
{
"msg_contents": "Hi all,\n\nAttached is a patch series that implements two features to the logical\nreplication - ability to define a memory limit for the reorderbuffer\n(responsible for building the decoded transactions), and ability to\nstream large in-progress transactions (exceeding the memory limit).\n\n... |
[
{
"msg_contents": "As I mentioned in the commit message for the Parallel Append commit\n(ab72716778128fb63d54ac256adf7fe6820a1185), it's kind of sad that this\ndoesn't work with UNION ALL queries, which are an obvious candidate\nfor such parallelization. It turns out that it actually does work to\na limited de... |
[
{
"msg_contents": "Hello,\n\n\n\nThe company I’m working for develops a product which uses Centos 6/7\n(different versions of the product) and also uses Vertica and PostgreSQL.\n\nDuring the course of the development of the latest version of our product,\nwe ran into problems that lead us to contact Vertica’s R... |
[
{
"msg_contents": "Hi all,\n\nTriggering \"tuple concurrently updated\" from heapam.c, which is an elog(),\nis not that difficult for some DDL commands. For example with ALTER ROLE,\njust create a role:\npsql --no-psqlrc -c 'create role popo'\n\nAnd then run that in two sessions and the error will show up, trig... |
[
{
"msg_contents": "c3d09b3bd23f5f6 fixed it so concurrent CIC would not deadlock (or at least\nnot as reliably as before) by dropping its own snapshot before waiting for\nall the other ones to go away.\n\nWith commit 8aa3e47510b969354ea02a, concurrent CREATE INDEX CONCURRENTLY on\ndifferent tables in the same d... |
[
{
"msg_contents": "In Ethiopia the year have 13 months and but in Gregorian calendar the year\nhave 12 months,\nThe Ethiopian society's are want to use his Ethiopian calendar year,i have\nthe algorithm,\nbut the way i add to the postgresql source code as EXTENSION? or to the\nbackend(to kernel)?\n13th month hav... |
[
{
"msg_contents": "Hi hackers,\n\nWhile testing parallel hash join today, I saw a couple of errors like this:\n\n2017-12-26 23:34:37.402 NZDT [13082] ERROR: cannot create temporary\nsubdirectory \"base/pgsql_tmp/pgsql_tmp13080.0.sharedfileset\": File\nexists\n\nThere is a thinko in PathNameCreateTemporaryDir()... |
[
{
"msg_contents": "yes it works on my machine,it coverts from Gregorian calendar year to\nEthiopian calendar year.\nSwift ,C# & Java languages locally we have app to convert it, but i can't\nget it from them,it is as a commercial,Also ORACLE DB localize it.Oracle DB\nalso use for commercial,\nI want to Localize... |
[
{
"msg_contents": "Hi,\n\nAs of today, in the existing code in HeapTupleSatisfiesVacuum, if a\nrow is inserted via an aborted transaction, then it's deemed to be\ndead immediately and can be cleaned up for re-use by HOT or vacuum,\netc.\n\nWith logical decoding, there might arise a case that such a row, if it\n... |
[
{
"msg_contents": "Hello, hackers!\n\nIntroduction\n------------\n\nI'm going to implement a patch which will store Ispell dictionaries in a shared memory.\n\nThere is an extension shared_ispell [1], developed by Tomas Vondra. But it is a bad candidate for including into contrib.\nBecause it should know a lot o... |
[
{
"msg_contents": "Hi All,\n\nI have added support in Postgres TAP test framework to backup a data\ndirectory with tablespace mapping. Also added support to move the backup\ndirectory contents to standby node, because current option to init the\nstandby from backup does not support copying softlinks, which is n... |
[
{
"msg_contents": "(2017/04/08 10:28), Robert Haas wrote:\n> On Wed, Mar 22, 2017 at 6:20 AM, Etsuro Fujita\n> <fujita.etsuro@lab.ntt.co.jp> wrote:\n>> On 2017/02/22 19:57, Rushabh Lathia wrote:\n>>> Marked this as Ready for Committer.\n>>\n>> I noticed that this item in the CF app was incorrectly marked as\nC... |
[
{
"msg_contents": "Here is a small patch for fixing $Subject: s/it's/its/.\n\nBest regards,\nEtsuro Fujita",
"msg_date": "Wed, 27 Dec 2017 21:35:54 +0900",
"msg_from": "Etsuro Fujita <fujita.etsuro@lab.ntt.co.jp>",
"msg_from_op": true,
"msg_subject": "Comment typo in postgres_fdw.c"
},
{
... |
[
{
"msg_contents": "Hi,\n\nThis is my first post on the list. My name is Antonio. I am a CS grad\nstudent and my field of study is about databases and information retrieval.\nTo get some practical knowledge, I've been studying Postgresql codebase for\na while.\n\nNow I would like to contribute with some code and... |
[
{
"msg_contents": "Those with long memories will recall that for some time now I've been\narguing that plpgsql should be changed to treat all composite-type\nvariables (both \"record\" and named composite types) via its DTYPE_REC\ncode paths, instead of the current situation where it handles variables of\nnamed... |
[
{
"msg_contents": "Hi,\n\nThis is my first post on the list. My name is Antonio. I am a CS grad\nstudent and my field of study is about databases and information retrieval.\nTo get some practical knowledge, I've been studying Postgresql codebase for\na while.\n\nNow I would like to contribute with some code and... |
[
{
"msg_contents": "Attached are patches for two performance-improvement ideas that came\nto me while working on\nhttps://www.postgresql.org/message-id/8962.1514399547@sss.pgh.pa.us\nThe three patches are logically independent and could be committed in\nany order. But they touch some overlapping code, so as pre... |
[
{
"msg_contents": "I want to revive a patch I sent couple years ago to the performance list,\nas I have seen the issue pop up repeatedly since then.\n\nOriginal thread is here:\nhttps://www.postgresql.org/message-id/CAK_s-G2tc8r0N3AqPr8fW5QsRQMbZNurgAQ%3D_ME1aaf4vOmnnA%40mail.gmail.com\n\nThe problem is that i... |
[
{
"msg_contents": "Hello.\n\nIt would be useful if we have interactive sessions for TAP\ntests. My first attempt was apparently unstable one that was\nusing psql. \n\nhttps://www.postgresql.org/message-id/20170720.152533.252230418.horiguchi.kyotaro@lab.ntt.co.jp\n\nFinally the test for the patch in the thread m... |
[
{
"msg_contents": "Hi,\n\nAttached is a patch to implement a feature to get the current function\nname by GET DIAGNOSTICS in PL/pgSQL function.\n\nCurrentyly, we can get call stack by GET DIAGNOSTICS PG_CONTEXT, but\nwe cannot get the function name directly. One of our clients wanted\nthis feature for debugging... |
[
{
"msg_contents": "Hi,\n\nI tried to ask a question in pgsql-general@ but got no response.\n\nI have postgres 9.4 standby with archive-based replication (via restore_command).\nPostgreSQL 9.4.12 on x86_64-unknown-linux-gnu, compiled by gcc (Debian 4.9.2-10) 4.9.2, 64-bit\n\nOne day I noticed strange behavior in... |
[
{
"msg_contents": "Hi,\n\nAttached is a patch to fix a very trivial issue of the documentation.\n\nThe documentation of PL/pgSQL provides sample codes of Oracle-compatible\ninstr functions. However, the behaviour is a little differet.\nOracle's instr raises an error when the forth argument value is less than\nz... |
[
{
"msg_contents": "Hi,\n\nThe current behavior of session_replication_role = replica with TRUNCATE\nis not the same of with the other commands.\nIt does not check FKs for INSERT/UPDATE/DELETE but it does for TRUNCATE,\nso one cannot execute TRUNCATE on a table when it is possible to DELETE\nfrom table without W... |
[
{
"msg_contents": "Hi,\n\nThis patch implements support for TRUNCATE statements\nin logical replication. The work has mainly done by Simon Riggs then\nfinished by me. Tests are written by me.\n\nTRUNCATE is treated as a form of DELETE for the purpose of deciding\nwhether to publish, or not.\n\nThe \"TRUNCATE be... |
[
{
"msg_contents": "Hi,\nReading\nhttps://stackoverflow.com/questions/48022753/why-does-array-ndimsarray-produce-null#48022980\nconfused me much - why array_ndims never returns zero indeed?..\nselect char_length('') returns zero and according to\nhttps://www.postgresql.org/docs/current/static/functions-string.ht... |
[
{
"msg_contents": "In https://postgr.es/m/1514541656508-0.post@n3.nabble.com\nit's reported that \"SELECT pg_ls_dir('c:')\" works to allow\ndisplay of the root directory on drive C. If true, this\nwould be a violation of the principle that the core file\naccess functions only let you get at PG-related director... |
[
{
"msg_contents": "New rangejoin patch attached.\n\nI had previously attempted to make this work well for multiple range\njoin keys, but this patch implements single rangejoin keys only, and\nthe rest of the rangejoin clauses are effectively just rechecks. I\nbelieve it can be made effective for multiple rangej... |
[
{
"msg_contents": "This patch enables FOR EACH ROW triggers on partitioned tables.\n\nAs presented, this patch is sufficient to discuss the semantics that we\nwant for triggers on partitioned tables, which is the most pressing\nquestion here ISTM.\n\nHowever, this is incomplete: it doesn't create triggers when ... |
[
{
"msg_contents": "This is the patch series for UNIQUE / PRIMARY KEY indexes on partitioned\ntables. This is on top of the patch in\nhttps://postgr.es/m/20171229175930.3aew7lzwd5w6m2x6@alvherre.pgsql\nbut I included it here as 0001 for simplicity. (Don't review that patch\nin this thread please). This is ess... |
[
{
"msg_contents": "I said a couple of times in recent threads that it wouldn't be too hard\nto implement $SUBJECT given the other patches I've been working on.\nAttached is proof of the pudding. This needs to be applied on top of\nthe patches in\nhttps://postgr.es/m/23537.1514589378@sss.pgh.pa.us\nand\nhttps:/... |
[
{
"msg_contents": "Comments in ReserveXLogInsertLocation() says\n\"* This is the performance critical part of XLogInsert that must be serialized\n * across backends. The rest can happen mostly in parallel. Try to keep this\n * section as short as possible, insertpos_lck can be heavily contended on a\n * busy... |
[
{
"msg_contents": "We are having a discussion on the jdbc project about dealing with 24:00:00.\n\nhttps://github.com/pgjdbc/pgjdbc/pull/992#issuecomment-354507612\n\nDave Cramer\n\nWe are having a discussion on the jdbc project about dealing with 24:00:00.https://github.com/pgjdbc/pgjdbc/pull/992#issuecomment-3... |
[
{
"msg_contents": "While I've been fooling around with plpgsql, I've been paying close\nattention to code coverage reports to make sure that the regression tests\nexercise all that new code. It started to bug me that there were some\nserious gaps in the test coverage for existing code in pl_exec.c.\nOne thing ... |
[
{
"msg_contents": "Hello All,\n\nOn a per-session basis, we cache the last heap block used for inserts and\ntry to use the same block for subsequent inserts. We don't do that for\nindexes because the target block in the index is determined by the overall\nstructure of the index and the index key being inserted ... |
[
{
"msg_contents": "Hi,\n\nAttached is patch which adds ability to do fast-forwarding while\ndecoding. That means wal is consumed as fast as possible and changes are\nnot given to output plugin for sending. The implementation is less\ninvasive than I originally though it would be. Most of it is just\nadditional ... |
[
{
"msg_contents": "Often when looking through pg_stat_statements, it would be nice to have\nsome sample values for the constants and parameters. This patch\nimplements that by taking the values from the first execution of the\nnormalized query.\n\nTo keep things reasonable, there is a limit on how big the para... |
[
{
"msg_contents": "Hello,\n\nSo far, the relation's block and file sizes have been defined statically at\nserver build time.\nThis is also the case for the WAL block size.\n\nThis means that it is not possible to share the same Postgresql binary if\nusing different block and file sizes for the WAL and the relat... |
[
{
"msg_contents": "Hello, I`d like to show my implementation of SLRU file protection with \nchecksums.\n\nIt only has effect if checksums on database are enabled.\nDetection of a checksum failure during a read normally causes PostgreSQL \nto report an error. Setting ignore_slru_checksum_failure to on causes \nt... |
[
{
"msg_contents": "Greetings,\n\nThis patch adds a new default role called 'pg_access_server_files' which\nallows an administrator to GRANT to a non-superuser role the ability to\naccess server-side files through PostgreSQL (as the user the database is\nrunning as). By itself, having this role allows a non-sup... |
[
{
"msg_contents": "This patch enables foreign key constraints to and from partitioned\ntables.\n\nNaturally, FKs that reference a partitioned table require unique\nconstraints, and therefore they shares the restrictions of those: in my\nproposed patch, it is only possible if the partition keys are part of\nthe ... |
[
{
"msg_contents": "I find myself entirely unimpressed with the results of commit dbb3d6f01.\n\nIn the first place, even among the compilers that claim to understand\nthat directive at all, there is a noticeable tendency to issue warnings.\nI find the following in recent buildfarm \"make\" logs:\n\n baiji | ... |
[
{
"msg_contents": "I am ready to update the copyright notice on HEAD for 2018. Any\nobjections? This shouldn't affect any pending patches since the\ncopyright text is normally isolated at the top of the file.\n\n-- \n Bruce Momjian <bruce@momjian.us> http://momjian.us\n EnterpriseDB ... |
[
{
"msg_contents": "Hi,\n\nUpcoming versions of the linux kernel (and apparently also windows and\nothers), will include new feature that apparently has been implemented\nwith haste to work around an intel hardware bug.\n\nhttps://lwn.net/SubscriberLink/741878/eaff7b24627c41a2/\n\nThe fix, split userland / kerne... |
[
{
"msg_contents": "I've been poking around in the PHJ code trying to identify the reason\nwhy there are still so many buildfarm failures. I've not nailed it\ndown yet, but one thing I did notice is that there's an entirely\nundocumented assumption that offsetof(HashMemoryChunkData, data) is\nmaxalign'ed. If i... |
[
{
"msg_contents": "Now that the January fest has nominally started, we need somebody\nto act as CF manager. Any volunteers?\n\n(If someone already did volunteer and I missed it, my apologies.)\n\n\t\t\tregards, tom lane\n\n",
"msg_date": "Tue, 02 Jan 2018 19:54:47 -0500",
"msg_from": "Tom Lane <tgl@sss... |
[
{
"msg_contents": "Hi,\n\nIn order to run some cardinality estimation experiments I'm hoping to\naccess the raw values for selections predicates from within the\ncalc_joinrel_size_estimate method (in costsize.c). If possible, I would\nvery much appreciate the help.\n\nBest, Walter\n\nHi,In order to run some car... |
[
{
"msg_contents": "Hi,\n\nAnimal caiman just revived and promptly failed:\nhttps://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=caiman&dt=2018-01-03%2003%3A13%3A54\n\ncopy_fetch.c:159:1: error: conflicting types for ‘copy_file_range’\n copy_file_range(const char *path, off_t begin, off_t end, bool trunc)\n ^... |
[
{
"msg_contents": "Buildfarm member caiman just came back online after some time AWOL, and\nit's not happy. The problem evidently is that pg_rewind's copy_file_range\nfunction conflicts with a recently added Linux system call:\n\nhttp://man7.org/linux/man-pages/man2/copy_file_range.2.html\n\nLooks like we'd be... |
[
{
"msg_contents": "Hi All,\n\nToday while trying to understand the code for ALTER TABLE in\nPostgreSQL (basically the table rewrite part), I noticed that we are\nswitching to a per-tuple memory context even when table rewrite is not\nrequired. For e.g.. consider the case where we do ADD CONSTRAINTS (NOT\nNULL o... |
[
{
"msg_contents": "https://caub.github.io/pg-tsquery/\n\nhttps://caub.github.io/pg-tsquery/",
"msg_date": "Wed, 3 Jan 2018 10:05:29 +0100",
"msg_from": "Cyril Auburtin <cyril.auburtin@gmail.com>",
"msg_from_op": true,
"msg_subject": "tsquery pre-parser"
}
] |
[
{
"msg_contents": "This small and simple standalone patch extracted from the SQL/JSON work\nwould allow the user to supply a string with a time zone specified as\nhh:mm thus:\n\n\n SELECT to_timestamp('2011-12-18 11:38 -05:20', 'YYYY-MM-DD HH12:MI\n TZH:TZM');\n to_timestamp \n -----... |
[
{
"msg_contents": "Hi,\n\nIn order to run some cardinality estimation experiments I'm hoping to\naccess the raw values for selections predicates from within the\ncalc_joinrel_size_estimate method (in costsize.c). For example, if the\nrestriction\n\nWHERE name_attr = \"example_str\"\n\nappears in the query, I'd ... |
[
{
"msg_contents": "Hello,\n\nThis is version 2 of the patch to make the file and block sizes for WAL and\nrelations, run-time configurable at initdb.\n\n\nSo far, the relation's block and file sizes have been defined statically at\nserver build time. This is also the case for the WAL block size. This means\ntha... |
[
{
"msg_contents": "Our database has a query that looks like this -- note the OR between a\nsimple equality qual and a LIKE qual:\n\n=> explain SELECT 1 AS one FROM \"redirect_routes\" WHERE\nredirect_routes.path = 'foobar' OR redirect_routes.path LIKE\n'foobar/%';\n QUERY PL... |
[
{
"msg_contents": "Greetings!\n\nThe January commitfest (the second-to-last of the PG11 release cycle,\nwhich likely means lots of big patches landing, since big patches\naren't allowed to be first seen in the last commitfest before feature\nfreeze... ;) has officially begun and we have a lot of work ahead of u... |
[
{
"msg_contents": "Hi Hackers,\n\nThis is a proposal that let libpq support 'prefer-read' option in\ntarget_session_attrs in pg_conn. The 'prefer-read' means the libpq will try\nto connect to a 'read-only' server firstly from the multiple server\naddresses. If failed to connect to the 'read-only' server then it... |
[
{
"msg_contents": "\nPostgreSQL Buildfarm client Release 6 is now available and can be\ndownloaded from\n<https://buildfarm.postgresql.org/downloads/releases/build-farm-6.tgz>\n\nFeatures\n\n * use a persistent cache for configure, leading to a substantial\n performance improvement for that step, based on a... |
[
{
"msg_contents": "Hi all,\n\nUsing IMPORT FOREIGN SCHEMA I notice that the column descriptions of \nthe remote tables are not taken along to the imported schema. I am \nusgin postgres_fdw to 'import' a schema with foreign tables (but the \nsame omission happens with ordinary tables as well).\n\nThis is a bu... |
[
{
"msg_contents": "The following bug has been logged on the website:\n\nBug reference: 14999\nLogged by: Christian H.\nEmail address: office@tiptop-labs.com\nPostgreSQL version: 10.1\nOperating system: e.g. Debian Buster\nDescription: \n\nI have encountered a bug in PostgreSQL 10.1: ... |
[
{
"msg_contents": "Implement channel binding tls-server-end-point for SCRAM\n\nThis adds a second standard channel binding type for SCRAM. It is\nmainly intended for third-party clients that cannot implement\ntls-unique, for example JDBC.\n\nAuthor: Michael Paquier <michael.paquier@gmail.com>\n\nBranch\n------... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.