threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "Hello list\n\nWhile testing an application with PostgreSQL 9.5, we experienced an issue\ninvolving aborted subtransactions and SELECT FOR UPDATE. In certain\nsituations, a locking query doesn't return rows that should be visible and\nalready locked by the current transaction.\n\nI bisected th... |
[
{
"msg_contents": "Hi,\n\nI found an useless entry in utf8_to_sjis.map\n\n> {0xc19c, 0x815f},\n\nwhich is apparently illegal as UTF-8 which postgresql\ndeliberately refuses. So it should be removed and the attached\npatch does that. 0x815f(SJIS) is also mapped from 0xefbcbc(U+FF3C\nFULLWIDTH REVERSE SOLIDUS) a... |
[
{
"msg_contents": "Hello.\n\nSome time ago two-phase state file format was changed to have variable size GID,\nbut several places that read that files were not updated to use new offsets. Problem\nexists in master and 9.6 and can be reproduced on prepared transactions with\nsavepoints. For example:\n\ncreate ta... |
[
{
"msg_contents": "PGDay Austin 2016 now accepting proposals for talks!\n\n\nJoin PostgreSQL users and advocates in Austin for a day of PostgreSQL at PGDay Austin 2016, on Saturday, November 12, 2016, at the Austin Galvanize campus<http://www.galvanize.com/campuses/austin-2nd-street-district/> (118 Nueces St. A... |
[
{
"msg_contents": "I was a bit surprised to discover the difference below in calling an SRF \nas part of a target list vs part of the from clause. The from clause \ngenerates a Function Scan, which (apparently blindly) builds a \ntuplestore. Is there a relatively easy way to either transform this type \nof quer... |
[
{
"msg_contents": "Hi,\n\nAn SQL (with only information_schema related JOINS) when triggered, runs\nwith max CPU (and never ends - killed after 2 days).\n- It runs similarly (very slow) on a replicated server that acts as a\nread-only slave.\n- Top shows only postgres as hitting max CPU (nothing else). When que... |
[
{
"msg_contents": "Hi all,\n\nLooking at the MSVC scripts for some stuff I have noticed the following thing:\n if ($options->{xml})\n {\n if (!($options->{xslt} && $options->{iconv}))\n {\n die \"XML requires both XSLT and ICONV\\n\";\n }\n }\nBut I don't understand the ... |
[
{
"msg_contents": "I happened to notice that if you type \"make\" in\nsrc/test/modules/test_pg_dump, you will get a \"make check\" action\nnot \"make all\". I hope this is just somebody being thoughtless\nabout Makefile ordering and not an intentional override of the\ndefault make target. If the latter, I beg... |
[
{
"msg_contents": "Hi,\n\nAttached patch add --disable-page-skipping option to vacuumed command for 9.6.\nIf by any chance the freeze map causes the serious data corruption bug\nthen the user will want to run pg_check_visible() and vacuum with\nDISABLE_PAGE_SKIPPING option to the multiple tables having problem.... |
[
{
"msg_contents": "Hi,\n\nI found these steps to a reliable crash (I know the patch is WIP but I \nassume you want to hear about these).\n\n(Running a single instance suffices)\n\n\npsql (10devel_logical_replication_20160901_1237_6f7c0ea32f80)\nType \"help\" for help.\n\n(PID 9809) [testdb] # CREATE PUBLICATIO... |
[
{
"msg_contents": "My apologies if this was already requested before...\n\nI think it would be fantastic if postgres had an \"explain the explain\" option:\nToday's explain tells us what loops and scans were used, and relative\ncosts, etc. It doesn't seem to tell *why* the planner elected to use\nwhat it did.\... |
[
{
"msg_contents": "People,\n\nI made a small modification in pg_dump to prevent parallel backup failures\ndue to exclusive lock requests made by other tasks.\n\nThe modification I made take shared locks for each parallel backup worker\nat the very beginning of the job. That way, any other job that attempts to\n... |
[
{
"msg_contents": "All,\n\nI have discovered a bug with the COPY command, specifically related to RLS.\n\nThe issue:\n\nWhen running COPY as superuser on a table that has RLS enabled, RLS is\nbypassed and therefore no issue exists. However, when performing a\nCOPY as a non-privileged user on the same table cau... |
[
{
"msg_contents": "While perusing the proposed PageIndexTupleOverwrite patch, I noticed\nthat PageIndexTupleDelete throws an error if size != MAXALIGN(size),\nwhere \"size\" is the ItemIdGetLength value. This seems wrong now that\nI look at it, because PageAddItem does not insist on tuple sizes being\nmaxalign... |
[
{
"msg_contents": "Hi,\n\nOn Debian unstable I just got a failure when running the regression\ntests:\nandres@alap4:~/build/postgres/dev-assert/vpath/src/bin/pg_rewind$ make check\nrm -rf '/home/andres/build/postgres/dev-assert/vpath'/tmp_install\n/bin/mkdir -p '/home/andres/build/postgres/dev-assert/vpath'/tmp... |
[
{
"msg_contents": "Hi,\n\nAm I missing something or is md.c:mdtruncate() leaking open files?\nThe relevant piece of code is:\n\n\t\tif (priorblocks > nblocks)\n\t\t{\n\t\t\t/*\n\t\t\t * This segment is no longer active (and has already been unlinked\n\t\t\t * from the mdfd_chain). We truncate the file, but do n... |
[
{
"msg_contents": "Hi,\n\nRegarding to the previous conversations [1], [2], here is a patch (with some\nimprovements from Nikita Glukhov) for the generic type subscription. It\nallows\nto define type-specific subscription logic for any data type and has\nimplementations for the array and jsonb types. There are ... |
[
{
"msg_contents": "Fix corruption of 2PC recovery with subxacts\n\nReading 2PC state files during recovery was borked, causing corruptions during\nrecovery. Effect limited to servers with 2PC, subtransactions and\nrecovery/replication.\n\nStas Kelvich, reviewed by Michael Paquier and Pavan Deolasee\n\nBranch\n-... |
[
{
"msg_contents": "A few things caught my eye while hacking on the latest round of sorting \npatches. Starting a new thread because these are orthogonal to the \nthings discussed on the other threads:\n\n1. SortTuple.tupindex is not used when the sort fits in memory. If we \nalso got rid of the isnull1 flag, we... |
[
{
"msg_contents": "I often see mention of project policies for various things (the one\ntriggering this email is in commit 967a7b0).\n\nWhere can I find documentation for these policies?\n-- \nVik Fearing +33 6 46 75 15 36\nhttp://2ndQuadrant.fr PostgreSQL : Expertis... |
[
{
"msg_contents": "Hello Andrew (from back in 2004),\n\nHave things improved at all regarding Postgres and the configurability of reduced public privileges for databases?\n\nWe have a goal to hide unneeded database objects from our database-challenged users - as they might see in a simple database viewer applic... |
[
{
"msg_contents": "For external sorts, cost_sort() assumes the following:\n\n * The disk traffic is assumed to be 3/4ths sequential and 1/4th random\n * accesses (XXX can't we refine that guess?)\n\n...\n /* Assume 3/4ths of accesses are sequential, 1/4th are not */\n startup_cost += npageaccesses... |
[
{
"msg_contents": "Hi,\nattached is a patch which adds support to localized month names in\nto_date() and to_timestamp() functions.\n\nThe patch is fairly simple but I want to discuss the approach and\nimplementation:\n\nUsing the TM modifier as in to_char() was already discussed some years\nago: 10710.12021708... |
[
{
"msg_contents": "Hi,\n\nAs I understand it, a merge join will currently read all tuples from both\nsubqueries (besides early termination). I believe it should be possible to\ntake advantages of the indexes on one or both of the tables being read from\nto skip a large number of tuples that would currently be r... |
[
{
"msg_contents": "I happened to read the pg_usleep() code recently. I'm wondering\nif we could implement it using the posix function nanosleep(),\ninstead of by select().\n\nnanosleep() is designed with higher time resolution, besides it provide\nremaining time if is interrupted by signal so that pg_usleep() c... |
[
{
"msg_contents": "The IDENTIFICATION filename in src/backend/storage/ipc/dsm_impl.c is\nincorrectly labelling the file dsm.c. Patch fixing the typo attached.\n\ncheers ./daniel\n\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://www.p... |
[
{
"msg_contents": "Hi all,\n\nThe current status summary is:\n\nNeeds review: 81\nWaiting on author: 35\nReady for Commiter: 12\nCommited: 78\nMoved to next CF: 1\nRejected: 6\nReturned with feedback: 6\nTOTAL: 219\n\nThe current progress is ~39%. The things moving fast but 27 patches still\nwith no signed revi... |
[
{
"msg_contents": "2016-09-12 8:14 GMT+02:00 Craig Ringer <craig@2ndquadrant.com>:\n\n> On 12 September 2016 at 14:02, Pavel Stehule <pavel.stehule@gmail.com>\n> wrote:\n> > Hi\n> >\n> > There is some opened questions - the standard (and some other databases)\n> > requires entering XPath expression as string li... |
[
{
"msg_contents": "Hi hackers,\n\nHere's a patch to add psql tab completion for the recently-added ALTER\nTYPE … RENAME VALUE feature (thanks to Tom for fixing it up and\ncommitting it).\n\nIt's modelled on the ALTER TYPE … RENAME ATTRIBUTE completion, but\ntweaked to return string literals instead of identifie... |
[
{
"msg_contents": "Friends,\n\nthere are several places in the code where variables defined as\n(char *) or as (const char *) are passed to the NameGetDatum()\nmacro. I believe it would be better form to use CStringGetDatum()\nin these locations. I am aware that these two macros are internally\nthe same.\n\ns... |
[
{
"msg_contents": "-- \n Craig Ringer http://www.2ndQuadrant.com/\n PostgreSQL Development, 24x7 Support, Training & Services\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://www.postgresql.org/mailpref/pgsql-hackers\n"... |
[
{
"msg_contents": "Hi all\n\nWhile updating an extension for 9.6 I noticed that while the\n$(prove_check) definition is exposed for use by PGXS in\nMakefile.global, extensions can't actually use the TAP tests because\nwe don't install the required Perl modules like PostgresNode.pm.\n\nI don't see any reason not... |
[
{
"msg_contents": "I just noticed something surprising:\r\n\r\n-- create a larger local table\r\nCREATE TABLE llarge (id integer NOT NULL, val integer NOT NULL);\r\nINSERT INTO llarge SELECT i, i%100 FROM generate_series(1, 10000) i;\r\nALTER TABLE llarge ADD PRIMARY KEY (id);\r\n\r\n-- create a small local tab... |
[
{
"msg_contents": "While working on write-ahead-logging of hash indexes, I noticed that\nthis function allocates buckets in batches and the mechanism it uses\nis that it initialize the last page of batch with zeros and expect\nthat the filesystem will ensure the intervening pages read as zeroes\ntoo.\n\nI think... |
[
{
"msg_contents": "Hi hackers!\n\nI've encountered need to extend functionality of PostgresNode class\nfrom the TAP test framework. What I want can easily be done via perl\nobject inheritation. \n\nBut documentation in the PostgresNode.pm recommends to use get_new_node\nfunction rather than call PostgresNode co... |
[
{
"msg_contents": "Simple question: Is there any reason for event trigger to not support\nCREATE/ALTER ROLE/USER?\n\nBest regards,\n--\nTatsuo Ishii\nSRA OSS, Inc. Japan\nEnglish: http://www.sraoss.co.jp/index_en.php\nJapanese:http://www.sraoss.co.jp\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@... |
[
{
"msg_contents": "Spotted a typo in the header comment of execIndexing.c. Attached fixes it.\n\ns/exclusive constraints/exclusion constraints/g\n\nThanks,\nAmit\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://www.postgresql.org/mail... |
[
{
"msg_contents": "Hello,\n\nIt looks to me pg_buffercache tries to allocate more than 1GB using\npalloc(), when shared_buffers is more than 256GB.\n\n# show shared_buffers ;\n shared_buffers\n----------------\n 280GB\n(1 row)\n\n# SELECT buffers, d.datname, coalesce(c.relname, '???')\n FROM (SELECT count(*)... |
[
{
"msg_contents": "I have Continued with testing declarative partitioning with the latest\npatch. Got some more observation, given below\n\n-- Observation 1 : Getting overlap error with START with EXCLUSIVE in range\npartition.\n\ncreate table test_range_bound ( a int) partition by range(a);\n--creating a parti... |
[
{
"msg_contents": "Hi All,\nWhile working on partition-wise join, I had to examine Relids objects\nmany times. Printing the Bitmapset::words[] in binary format and then\ninferring the relids takes time and is error prone. Instead I wrote a\nfunction bms_to_char() which allocates a StringInfo, calls\noutBitmapse... |
[
{
"msg_contents": "Hi, I'm kind of new to Postgres and I'm trying to create a custom output\nplugin for logical replication (Postgres is 9.5.4 version and I'm building\nthe project from a Windows 8/64 bit machine - same machine where the db is\ninstalled).\nI started from the code of the sample test_decoding, a... |
[
{
"msg_contents": "Hi hackers!\n\nCurrently GiST treats NaN penalty as zero penalty, in terms of\ngeneralized tree this means \"perfect fit\". I think that this situation\nshould be considered \"worst fit\" instead.\nHere is a patch to highlight place in the code.\nI could not construct test to generate bad tre... |
[
{
"msg_contents": "Support OpenSSL 1.1.0.\n\nChanges needed to build at all:\n\n- Check for SSL_new in configure, now that SSL_library_init is a macro.\n- Do not access struct members directly. This includes some new code in\n pgcrypto, to use the resource owner mechanism to ensure that we don't\n leak OpenSS... |
[
{
"msg_contents": "When building with --with-blocksize=16, the select_parallel test fails\nwith this difference:\n\n explain (costs off)\n select sum(parallel_restricted(unique1)) from tenk1\n group by(parallel_restricted(unique1));\n- QUERY PLAN\n-----------------------------... |
[
{
"msg_contents": "Hi,\n\nthe current implementation of PostmasterIsAlive() uses a pipe to\nmonitor the existence of the postmaster process.\nOne end of the pipe is held open in the postmaster, while the other end is\ninherited to all the auxiliary and background processes when they fork.\nThis leads to multipl... |
[
{
"msg_contents": "Hi, hackers!\n\nDo we have any standard set of functions to operate with files?\n\nI mean link(), copy(), rename(), unlink(), ftruncate() and so on.\nMostly I worry about the first two since they have different\nimplementation on Windows.\nI found a couple of functions in src/backend/storage/... |
[
{
"msg_contents": "Hello!\nDoes parallel secscan works in plpgsql?\n\n-- \nAlex Ignatov\nPostgres Professional: http://www.postgrespro.com\nThe Russian Postgres Company\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://www.postgresql.or... |
[
{
"msg_contents": "On Mon, Sep 5, 2016 at 1:44 PM, Pavel Stehule <pavel.stehule@gmail.com> wrote:\n...\n\n> I wrote XMLTABLE function, and I am thinking about JSON_TABLE function. But\n> there is one blocker - missing JsonPath support in our JSON implementation.\n>\n> So one idea - implement JsonPath support an... |
[
{
"msg_contents": "Hi Hackers,\n\n\nI checked if there is update transaction on master table involved in\npartition.\nPostgresql takes RowExclusiveLock on all partition tables.\n\nconstraint exclusion is set to on.\n\nMy question is why it locks on all partition tables instead only one\npartition tables wher... |
[
{
"msg_contents": "Currently README of hash module contain algorithms written in below form.\n\nThe insertion algorithm is rather similar:\n\npin meta page and take buffer content lock in shared mode\nloop:\ncompute bucket number for target hash key\nrelease meta page buffer content lock\nif (correct bucket pag... |
[
{
"msg_contents": "Hi,\n\nFollowing the discussion on forbidding an AUTOCOMMIT off->on\nswitch mid-transaction [1], attached is a patch that let the hooks\nreturn a boolean indicating whether a change is allowed.\n\nUsing the hooks, bogus assignments to built-in variables can\nbe dealt with more strictly.\n\nFo... |
[
{
"msg_contents": "On Mon, May 23, 2016 at 6:31 PM, Alvaro Herrera\n<alvherre@2ndquadrant.com> wrote:\n> Tom Lane wrote:\n>> My feeling about it is that we need to provide a partitioning feature\n>> that doesn't rely on the current notion of inheritance at all. We've\n>> heard from multiple users who want to u... |
[
{
"msg_contents": "Current PostgreSQL Documentation on recovery.conf has this about\nrecovery_min_apply_delay[1]:\n\n---<---\n\nThis parameter is intended for use with streaming replication deployments;\nhowever, if the parameter is specified it will be honored in all cases.\nSynchronous replication is not affe... |
[
{
"msg_contents": "Current PostgreSQL Documentation on recovery.conf has this about\nrecovery_min_apply_delay[1]:\n\n---<---\n\nThis parameter is intended for use with streaming replication deployments;\nhowever, if the parameter is specified it will be honored in all cases.\nSynchronous replication is not affe... |
[
{
"msg_contents": "Hello,\n\npg_recvlogical is not included in the Windows client installation,\nwhich is performed by running \"install <instdir> client\". The\nattached patch based on HEAD fixes this. I confirmed nothing else is\nmissing in the client installation.\n\n\nRegards\nMauMau (= Takayuki Tsunakawa... |
[
{
"msg_contents": "I've recently seen this when using 9.6:\n\n#0 0x00007f147892f0c7 in raise () from /lib64/libc.so.6\n#1 0x00007f1478930478 in abort () from /lib64/libc.so.6\n#2 0x00000000009683a1 in ExceptionalCondition (conditionName=0x9f2ef8 \"!(((PageHeader) (page))->pd_special >= (__builtin_offsetof (P... |
[
{
"msg_contents": "Hello,\n\n> From: pgsql-hackers-owner@postgresql.org\n> [mailto:pgsql-hackers-owner@postgresql.org] On Behalf Of Magnus Hagander\n> On Wed, Aug 24, 2016 at 4:35 AM, Tsunakawa, Takayuki\n> <tsunakawa.takay@jp.fujitsu.com> wrote:\n> \tAs a similar topic, I wonder whether the following still hol... |
[
{
"msg_contents": "Hi,\nThere's code in add_paths_to_joinrel() which computes the set of\ntarget relations that should overlap parameterization of any proposed\njoin path.\n\n 120 /*\n 121 * Decide whether it's sensible to generate parameterized\npaths for this\n 122 * joinrel, and if so, which re... |
[
{
"msg_contents": "Hello,\n\nPlease let me ask you about possible causes of a certain problem, slow shutdown of postmaster when a backend crashes, and whether to fix PostgreSQL.\n\nOur customer is using 64-bit PostgreSQL 9.2.8 on RHEL 6.4. Yes, the PostgreSQL version is rather old but there's no relevant bug f... |
[
{
"msg_contents": "Hi,\n\nI happened to notice this comment in src/include/storage/itemptr.h\n\n * Note: because there is an item pointer in each tuple header and index\n * tuple header on disk, it's very important not to waste space with\n * structure padding bytes. The struct is designed to be six bytes long... |
[
{
"msg_contents": "Hi,\n\nI work in an environment, where servers are administered by people\nwith different user names and identical uid (0). The attached patch\nfixes a bug exposed in such environments: where the logic of\nretrieving a personal configuration file relies solely on\nget_home_path(), the differe... |
[
{
"msg_contents": "Friends,\n\nThere are places in the code that cast away const for no apparent\nreason, fixed like such:\n\n\n\ndiff --git a/src/backend/executor/nodeIndexscan.c b/src/backend/executor/nodeIndexscan.c\nindex 3143bd9..fe2cfc2 100644\n--- a/src/backend/executor/nodeIndexscan.c\n+++ b/src/backend... |
[
{
"msg_contents": "Friends, per the recent thread \"gratuitous casting away const\", the\nassignment on line 1247 of localtime.c has const lvalue and rvalue,\nyet casts through (char *) rather than (const char *). Fix attached.\n\nMark Dilger\n\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@pos... |
[
{
"msg_contents": "Hi,\n\nI got a server crash with partition_wise_join, steps to reproduce given\nbelow.\n\npostgres=# set enable_partition_wise_join=true;\nSET\npostgres=# CREATE TABLE tbl (a int,c text) PARTITION BY LIST(a);\nCREATE TABLE\npostgres=# CREATE TABLE tbl_p1 PARTITION OF tbl FOR VALUES IN (1, 2);... |
[
{
"msg_contents": "On Thus, July 7,2016 at 08:39 PM, Fabien COELHO <coelho@cri.ensmp.fr> wrote:\n> Also there is still the bug under -t which displays a 0 latency.\nYour patch clearly fixed the issue.\n\n> The attached patch still fixes that and make it consistent the other way\n> around, i.e. by using \"=\" fo... |
[
{
"msg_contents": "Hi all,\n\nWhile going through pgstat.h I bumped into that:\n- * as 4-bytes where first byte repersents the wait event class (type of\n+ * as 4-bytes where first byte represents the wait event class (type of\nRegards,\n-- \nMichael\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hacker... |
[
{
"msg_contents": "There is apparently some misbehavior if max_wal_size is less than 5 *\nwal_segment_size.\n\nFor example, if you build with --with-wal-segsize=64, then the recovery\ntest fails unless you set max_wal_size to at least 320MB in\nPostgresNode.pm. The issue is that pg_basebackup fails with:\n\npg... |
[
{
"msg_contents": "pg_ctl: Detect current standby state from pg_control\n\npg_ctl used to determine whether a server was in standby mode by looking\nfor a recovery.conf file. With this change, it instead looks into\npg_control, which is potentially more accurate. There are also\noccasional discussions about r... |
[
{
"msg_contents": "I looked into the misbehavior reported in bug #14328,\nhttps://www.postgresql.org/message-id/20160919160136.1348.55251@wrigleys.postgresql.org\n\nWhat is happening is that during the EvalPlanQual recheck to see whether\nthe updated row still satisfies the SELECT's quals, we run ExecInitCteSca... |
[
{
"msg_contents": "Ran into a typo in libpq-int.h while reading/hacking:\n\n- \tchar \t *gsslib; \t\t\t/* What GSS librart to use (\"gssapi\" or\n+ \tchar \t *gsslib; \t\t\t/* What GSS library to use (\"gssapi” or\n\nPatch attached.\n\ncheers ./daniel\n\n\n\n\n-- \nSent via pgsql-hackers... |
[
{
"msg_contents": "Hello psql-hackers!\n\nWe thought it would be advantageous to be able to specify a 'custom' \npgpassfile within the connection string along the lines of the existing \nparameters sslkey and sslcert.\n\nWhich is exactly what this very compact patch does.\nThe patch is minimally invasive - when... |
[
{
"msg_contents": "\nI have just encountered an apparent bug in pg_upgrade (or possibly pg_dump).\n\n\nTo recreate, install the cranges extension - which can be obtained via \n\"git clone https://bitbucket.org/adunstan/pg-closed-ranges.git\" - for \nboth 9.4 and 9.5. Create a fresh 9.4 instance, create a databa... |
[
{
"msg_contents": "Hello.\n\nVery simple small patch - see attachment.\n\n-- \nBest regards,\nAleksander Alekseev\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://www.postgresql.org/mailpref/pgsql-hackers",
"msg_date": "Fri, 23 Sep... |
[
{
"msg_contents": "Sorry to be asking another phrase search syntax question, and so close\nto final release, but ...\n\nWhy does the phrase distance operator assume <1> means adjacent words,\nand not <0>. (FYI, <-> is the same as <1>.)\n\nFor example:\n\n select to_tsvector('park a a house') @@ to_tsquery('pa... |
[
{
"msg_contents": "Currently, hash indexes always store the hash code in the index, but\nnot the actual Datum. It's recently been noted that this can make a\nhash index smaller than the corresponding btree index would be if the\ncolumn is wide. However, if the index is being built on a fixed-width\ncolumn wit... |
[
{
"msg_contents": "Apologies for the late notice on this, but the release team has concluded\nthat 9.6 is about as ready as it's going to get. We are planning to go\nahead with 9.6.0 release next week --- that is, wrap Monday 26th for\npublic announcement on Thursday 29th.\n\nThanks to the Release Management T... |
[
{
"msg_contents": "Hi List,\n\nNote beforehand: this question is a result of a stack-exchange that can be\nseen here:\nhttp://stackoverflow.com/questions/39624241/inconsistent-behaviour-of-set-returning-functions-in-sub-query-with-random\n\nI'm often using the WHERE clause random() > 0.5 to pick a random subset... |
[
{
"msg_contents": "Hi,\n\nPlaying around with my hashtable implementation [1] and using it for\nbitmapscans/tidbitmap.c I noticed something curious. When using small\nwork_mem (4MB in my case) for large-ish tables (~5GB), the performance\ntanks. That's primarily caused by some issues in the hashtable code\n(sin... |
[
{
"msg_contents": "[ Adjusting subject line to reflect the actual topic of discussion better. ]\n\nOn Fri, Sep 23, 2016 at 9:29 AM, Robert Haas <robertmhaas@gmail.com> wrote:\n> On Fri, Sep 23, 2016 at 8:45 AM, Amit Khandekar <amitdkhan.pg@gmail.com> wrote:\n>> For e.g., in the above plan which you specified, s... |
[
{
"msg_contents": "Hi hackers\n\nStartupXLOG is 1549 lines of code. Its unwieldy size came up in a\ndiscussion in an IRC channel where some PG hackers lurk and I decided\nto see how it might be chopped up into subroutines with a clear\npurpose and reasonable size, as an exercise.\n\nI came up with the followin... |
[
{
"msg_contents": "Hi\n\nAfter LOCK TABLE ... IN ACCESS|ROW|SHARE we run out of completions.\nHere's a patch to improve that, for November.\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 subscription:\nhtt... |
[
{
"msg_contents": "\nHello,\n\nThe checkpoint time limit has just been raised to one day after a \ndiscussion started by Andres Freund:\n\nhttps://www.postgresql.org/message-id/20160202001320.GP8743%40awork2.anarazel.de\n\nI would have gone further up, say one week or even one month, but I think \nthat this new... |
[
{
"msg_contents": "Apple has decided to rename Mac OS X to \"macOS\", and apparently is now\nretroactively referring to old releases that way too. I propose that\nwe should do likewise, ie run around and clean up our various references\nto \"Mac OS X\", \"OS X\", or \"OSX\" to uniformly say \"macOS\". I think... |
[
{
"msg_contents": "I am trying to do a macOS build of postgresql (9.6 stable branch from\ngithub) with the uuid-ossp contrib by typing \"make world\" but it fails due\nto an openjade error (I did install openjade using homebrew using this\nsetup https://github.com/petere/homebrew-sgml).\n\nIs there a way to bui... |
[
{
"msg_contents": "Hi all,\n\nThe current status summary is:\n\nNeeds review: 65\nWaiting on author: 34\nReady for Commiter: 9\nCommited: 88\nMoved to next CF: 1\nRejected: 11\nReturned with feedback: 11\nTOTAL: 219\n\nThe current progress is ~51%. We're in the last week of this CF and we\nstill with about ~30%... |
[
{
"msg_contents": "Hello,\n\nThe attached patch implements huge_pages on Windows. I'll add this to the CommitFest.\n\nThe performance improvement was about 2% with the following select-only pgbench. The scaling factor is 200, where the database size is roughly 3GB. I ran the benchmark on my Windows 10 PC wit... |
[
{
"msg_contents": "On Tue, Jul 5, 2016 at 06:39 AM, Haribabu Kommi\nkommi(dot)haribabu(at)gmail(dot)com wrote:\n\nStill i feel the GRANT statements should be present, as the create\ndatabase statement\nis generated only with -C option. So attached patch produces the GRANT\nstatements based\non the -x option.\n\... |
[
{
"msg_contents": "When using pg_xlogdump in --follow mode, there is what I believe is a small\nrace condition when files are switched.\n\nIf pg_xlogdump detects then end of one file (by looking at the record) it\nwill immediately try to open the following file. If that file has not yet\nbeen created, it will f... |
[
{
"msg_contents": "test:\r\ncreate type h3 as (id int,name char(10));\r\n\r\nCREATE or replace FUNCTION proc17() \r\nRETURNS SETOF h3 AS $$ \r\nDECLARE\r\nv_rec h3;\r\nBEGIN \r\n create temp table abc(id int,name varchar) on commit drop;\r\ninsert into abc select 1,'lw';\r\ninsert into abc select 2,'lw2';\... |
[
{
"msg_contents": "Hackers,\n\nAt work we use several major versions of PostgreSQL, and developers\nuse non-local clusters for developing and debugging.\nWe do dump/restore schemas/data via custom/dir formats and we have to\nkeep several client versions for 9.2, 9.4 and 9.5 versions on local\nworkstations becau... |
[
{
"msg_contents": "Hi,\n I am working in optimizer module of postgreSQL 9.4.1. My project is\nconcerned with the plan returned by the optimizer.\n\nI am trying to return a subplan for a query instead of full plan. For this\nI need to return a path form *RelOptInfo *standard_join_search() *at*\nallpaths.c *oth... |
[
{
"msg_contents": "Dear PostgresSQL Hackers,\n I am working in optimizer module of postgreSQL 9.4.1. I am trying to\nreturn a subplan for a query instead of full plan. For this I need to\nreturn an intermediate plan (or path) from the DP lattice (i.e. from\n*RelOptInfo\n*standard_join_search() *at* allpaths.... |
[
{
"msg_contents": "When running ./configure on a system without Flex/Bison it’s easy to miss the\nwarning that flies past and then run into compilation error instead. When it\nhappened to a colleague yesterday a brief discussion came to the conclusion\nthat it would be neat it the flex and bison checks took th... |
[
{
"msg_contents": "On Tue, Sep 27, 2016 at 3:08 PM, Robert Haas <robertmhaas@gmail.com> wrote:\n> I don't know what any of that means. You said we need something like\n> an LWLock, but I think we don't. The workers just write the results\n> of their own final merges into shm_mqs. The leader can read from any... |
[
{
"msg_contents": "Hello.\n\nDuring processing of logical messages in ReorderBufferSerializeChange()\npointer to ondisk structure isn’t updated after possible reorder buffer realloc by\nReorderBufferSerializeReserve().\n\nActual reason spotted by Konstantin Knizhnik.\n\n\n\n\n\n-- \nStas Kelvich\nPostgres Profe... |
[
{
"msg_contents": "Hi,\n\ndebugging a citus valgrind bleat I noticed that hash_create() accesses\nthe result of palloc(0) as an hash element:\nHTAB *\nhash_create(const char *tabname, long nelem, HASHCTL *info, int flags)\n{\n...\n\tif ((flags & HASH_SHARED_MEM) ||\n\t\tnelem < hctl->nelem_alloc)\n\t{\n\t\tif (... |
[
{
"msg_contents": "The function\n\n static void xlog_outdesc(StringInfo buf, XLogReaderState *record);\n\nin src/backend/access/transam/xlog.c is called by XLogInsertRecord,\nand after returning a string describing an XLogRecord, it clears the\nstate data in its XLogReaderState argument. That mixes the read-o... |
[
{
"msg_contents": "Friends,\n\ncomparators usually take arguments like\n\n (const void *a, const void *b)\n\nand do something read-only to a and b. In our\nsources, we typically cast these to something else,\nlike (char *), and do something read-only. This \ngenerates a lot of warnings if using -Wcast-qual.\... |
[
{
"msg_contents": "Friends,\n\nalong the lines of other similar emails from me of late,\nI tried to avoid casting away const when using the FileName\ntypedef. There are several calls where a (const char *) has to\nbe cast to (char *) due to FileName being typedef'd as\nnon-const. But changing the typedef to c... |
[
{
"msg_contents": "rangetypes_spgist.c contains a call to detoast a pointer which was\njust returned from a detoast operation. This seems to do no harm,\nbut is to my mind a sloppy thinko. Fix attached.\n\ntsgistidx.c contains a detoast of a pointer where detoast of a datum\nwas likely intended. Fix attached... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.