threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "On Thu, Jan 5, 2017 at 8:20 AM, Etsuro Fujita\n<fujita.etsuro@lab.ntt.co.jp> wrote:\n> On 2016/12/27 16:41, Etsuro Fujita wrote:\n>>\n>> On 2016/12/22 1:04, Ashutosh Bapat wrote:\n>>>\n>>> 2. We should try to look for other not-so-cheap paths if the cheapest\n>>> one is\n>>> paramterized. You... |
[
{
"msg_contents": ">\n>> Hmm. If I understand the patch correctly, it does not return any path\n>> when merge join is allowed and there are merge clauses but no hash\n>> clauses. In this case we will not create a foreign join path, loosing\n>> some optimization. If we remove GetExistingLocalJoinPath, which\n>> ... |
[
{
"msg_contents": "\n<Argh, wrong sender again, sorry [again] for the repost.\n I must tweak my mail client configuration...>\n\n>>> Good. So we seem to agree that GUCS are transactional?\n> \n> I'm surprised, I never knew this.\n\nI must admit that it was also a (good) surprise for me.\n\nThe documentation sa... |
[
{
"msg_contents": "Fix possible crash reading pg_stat_activity.\n\nWith the old code, a backend that read pg_stat_activity without ever\nhaving executed a parallel query might see a backend in the midst of\nexecuting one waiting on a DSA LWLock, resulting in a crash. The\nsolution is for backends to register t... |
[
{
"msg_contents": "Attached please find a patch for PostgreSQL 9.4 which changes the maximum\namount of data that the wal sender will send at any point in time from the\nhard-coded value of 128KiB to a user-controllable value up to 16MiB. It has\nbeen primarily tested under 9.4 but there has been some testing w... |
[
{
"msg_contents": "I noticed a strange behavior when T_Float value is serialized, then deserialized\non the worker process for cpu parallel execution.\n\nSimplified description of what I did is:\n fval = makeFloat(psprintf(\"%.0f\", plan_nrows));\n custom_scan->custom_private = list_make1(fval);\n\nThis strin... |
[
{
"msg_contents": "Herewith a patch for doing grouping sets via hashing or mixed hashing\nand sorting.\n\nThe principal objective is to pick whatever combination of grouping sets\nhas an estimated size that fits in work_mem, and minimizes the number of\nsorting passes we need to do over the data, and hash those... |
[
{
"msg_contents": "Hi all,\n\nWhen reading through \"Support for pg_receivexlog --format=plain|tar\"[1], I\ncame across a notion from Magnus Hagander that has crossed my mind a few\ntimes as well in the past years. As the feature proposed here is not\ndirectly related to that thread, I thought it best to start ... |
[
{
"msg_contents": "While reviewing Etsuro-san's patch to force replanning after FDW option\nchanges, I noticed that there is a great lack of consistency about where\nInvokeObjectPostAlterHook calls have been placed relative to other actions\nsuch as forced relcache invals. I wonder exactly what expectations a\... |
[
{
"msg_contents": " Hi,\n\nWhen testing the patch at https://commitfest.postgresql.org/12/768/\n(\"sequence data type\" by Peter E.), I notice that there's a preexisting\noddity in the fact that sequences created with a negative increment\nin current releases initialize the minval to -(2^63)+1 instead of -2^63,... |
[
{
"msg_contents": "It looks to to me like the recent hash index changes have left\n_hash_addovflpage slightly broken. I think that if that function\nreaches the point where it calls _hash_getbuf() to fetch the next page\nin the bucket chain, we also need to clear retain_pin. Otherwise,\nwe'll erroneously thin... |
[
{
"msg_contents": "It seems that commit 01ec2563 has a subtle bug, which stems from the\nfact that logtape.c no longer follows the rule described above\nltsGetFreeBlock():\n\n/*\n * Select a currently unused block for writing to.\n *\n * NB: should only be called when writer is ready to write immediately,\n * t... |
[
{
"msg_contents": "So in the grouping sets patch post, I said:\n\n >> There is one current weakness which I don't see a good solution for:\n >> the planner code still has to pick a single value for group_pathkeys\n >> before planning the input path. This means that we sometimes can't\n >> choose a minimal set ... |
[
{
"msg_contents": "Hey Postgres team!\n\nI've been gleefully using the inheritance feature of Postgres for the last\n6 months or so, and it's really great! I especially like how easily you\ncan ALTER TABLE foo INHERIT bar, and get helpful error messages about what\ncolumns need to be there before the inherit c... |
[
{
"msg_contents": "While running a couple of parallel make (install)check-worlds I was \nsurprised to see my system load at 5x what I'd think it'd be. Looking at \nCPU % consumption it turns out this was because mds (the OS X indexing \nservice) and a few other things were going bonkers. That's because they \nw... |
[
{
"msg_contents": "Hi all,\n\nIs anyone working on porting PostgreSQL to Rust?\n\nCorrode looks a bit limited for the task, but maybe it can be a start.\nIt doesn't support goto or switch, but maybe the gotos patterns are not too\ncomplicated.\n\nMy motivation is primarily I don't want to learn all the over-com... |
[
{
"msg_contents": "Collegues!\n\nRecently I've found out that PL/Python have very nice feature - explicit\nsubtransaction object, which allows to execute block of code in the\ncontext of subtransaction.\n\nI've quite surprised that other PL languages, shipped with PostgreSQL do\nnot have such useful constructio... |
[
{
"msg_contents": "I'm seeing the following warnings on a recent checkout. make check still \npasses fine though...\n\n> ccache clang -Qunused-arguments -fcolor-diagnostics -Wall -Wmissing-prototypes -Wpointer-arith -Wdeclaration-after-statement -Wendif-labels -Wmissing-format-attribute -Wformat-security -fno-s... |
[
{
"msg_contents": "Attached is a draft patch that lets partial aggregation happen at base\nrelation level. If the relations contain relatively small number of groups,\nthe number of input rows of the aggregation at the query level can be reduced\nthis way. Also, if append relation and postgres_fdw planning is ... |
[
{
"msg_contents": "Greetings,\n\nFor reasons which seem likely to be entirely unintentional, pg_restore\nwill accept a '-1' for -j:\n\npg_restore -j -1\n\nThis seems to result in the parallel state being NULL and so things\ndon't outright break, but it hardly seems likely to be what the user was\nasking for- my... |
[
{
"msg_contents": "I happened to notice that we have dead code to parse floating point\nnumbers in the boostrap scanner/parser. This seems unused since forever\n-- or, more precisely, I couldn't find any floats in 8.2's postgres.bki\n(the oldest I have around). I would just rip it out, per the attached.\n\n--... |
[
{
"msg_contents": "Hi all!\n\nIn 2016 PostgreSQL project didn't pass to GSoC program. In my\nunderstanding the reasons for that are following.\n\n1. We did last-minute submission of our application to GSoC.\n2. In 2016 GSoC application form for mentoring organizations has been\nchanged. In particular, it requ... |
[
{
"msg_contents": "Two options are on the table to attack the problem of updates causing\nwrite amplification: WARM and indirect indexes. They are completely\ndifferent approaches but have overlapping effects on what scenarios are\nimproved. Here's a recap of both features, with the intent that we make\na wel... |
[
{
"msg_contents": "It has come to my attention that when a user has a CONNECTION LIMIT\nset, and they make use of parallel query, that their queries can fail\ndue to the connection limit being exceeded.\n\nSimple test case:\n\npostgres=# CREATE USER user1 LOGIN CONNECTION LIMIT 2;\nCREATE ROLE\npostgres=# \\c p... |
[
{
"msg_contents": "(This came up on IRC, but I'm not sure to what extent it should be\nconsidered a \"bug\")\n\nIf you do PQprepare(conn, \"myfetch\", \"FETCH ALL FROM mycursor\", ...);\nthen the results are unpredictable in two ways:\n\nFirstly, nothing causes the plancache entry to be revalidated just\nbecau... |
[
{
"msg_contents": "Hello everybody,\n\nCurrently, query string is not passed to the workers and only master has\nit. In the events, when multiple queries are running on a server and for\none query some worker crashes then it becomes quite burdensome to find the\nquery with the crashed worker, since on the worke... |
[
{
"msg_contents": "Hi,\n\nAttached fixes comment typos in dsa.c.\n\nRegards,\n\n--\nMasahiko Sawada\nNIPPON TELEGRAPH AND TELEPHONE CORPORATION\nNTT Open Source Software Center\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://www.postg... |
[
{
"msg_contents": "The comment above ParallelQueryMain() still refers to ParallelMain() as\nits caller which is no longer the case. Attached fixes that.\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.... |
[
{
"msg_contents": "Hi,\n\nIn transformDeleteStmt:\n\n qry->hasWindowFuncs = pstate->p_hasWindowFuncs;\n qry->hasAggs = pstate->p_hasAggs;\n if (pstate->p_hasAggs)\n parseCheckAggregates(pstate, qry);\n\nThese code set agg and window function status for delete query,\nbut there is no similar code... |
[
{
"msg_contents": "Hi\n\nI am starting new thread for this patch (related to merging some ideas from\nplpgsql2 project thread).\n\nHere is simple patch with two new extra_warnings, extra_errors checks\n\n1. strict_multi_assignment - checks the number of target variables and\nsource values.\n2. too_many_rows - c... |
[
{
"msg_contents": "There are approximately 11 copies of atooid() and 3 of oid_cmp() or\nequivalent, and pending patches are proposing to add more. I propose\nthese two patches to collect them in central places.\n\n-- \nPeter Eisentraut http://www.2ndQuadrant.com/\nPostgreSQL Development, 24x7 Supp... |
[
{
"msg_contents": "-hackers,\n\nI know we have talked about this before but today it was impressed upon \nme rather firmly. I presented a Webinar: Postgres for Oracle People. The \nattendees were 90% pl/pgsql developers. 330 people registered for an \nevent that was only allowed to host 100 people. The webinar ... |
[
{
"msg_contents": "Hackers:\n\nYou will have noticed that I haven't been very active for the past year.\n My new work on Linux containers and Kubernetes has been even more\nabsorbing than I anticipated, and I just haven't had a lot of time for\nPostgreSQL work.\n\nFor that reason, as of today, I am stepping dow... |
[
{
"msg_contents": "Hello,\n\nWhen working with a production (P) and a DR (D) environment it is often a\nrequirement to be able to protect against data loss when promoting within a\nsite, and also when losing A and promoting a master at D.\n\nThe current synchronous_standby_names do not allow this.\n\nIn a simpl... |
[
{
"msg_contents": "Hi,\n\nWhile working on pushing down more joins/updates to the remote, I \nnoticed that in contrib/postgres_fdw/postgres_fdw.h the declaration of \nget_jointype_name is misplaced in the section of shippable.c. Since \nthat function is defined in contrib/postgres_fdw/deparse.c, we should ... |
[
{
"msg_contents": "Hi,\nThe member root in foreign_glob_cxt isn't used anywhere by\npostgres_fdw code. Without that member the code compiles and\nregression passes. The member was added by d0d75c40. I looked at that\ncommit briefly but did not find any code using it there. So, possibly\nit's unused since it was... |
[
{
"msg_contents": "In 9.5, the default pg_ctl stop mode was changed from \"smart\" to \"fast\".\n\nIn pg_upgrade, there is this code:\n\nvoid\nstop_postmaster(bool fast)\n{\n ...\n exec_prog(SERVER_STOP_LOG_FILE, NULL, !fast,\n \"\\\"%s/pg_ctl\\\" -w -D \\\"%s\\\" -o \\\"%s\\\" %s stop\",\n ... |
[
{
"msg_contents": "Fix field order in struct catcache.\n\nSomebody failed to grasp the point of having the #ifdef CATCACHE_STATS\nfields at the end of the struct. Put that back the way it should be,\nand add a comment making it more explicit why it should be that way.\n\nBranch\n------\nmaster\n\nDetails\n----... |
[
{
"msg_contents": "There is a typo in comments of function _hash_first(); Adding a fix for same.\n\n-- \nThanks and Regards\nMithun C Y\nEnterpriseDB: http://www.enterprisedb.com\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscription:\nhttp://www.pos... |
[
{
"msg_contents": "In a server where autovacuum is disabled and its databases reach\nautovacuum_freeze_max_age limit, an autovacuum is forced to prevent\nxid wraparound issues. At this stage, when the server is loaded with a\nlot of DML operations, an exceedingly high number of autovacuum\nworkers keep on getti... |
[
{
"msg_contents": "Hi\n\nI would to do import without RI check - so I disable RI triggers.\n\nI would to invalidate constraint on Foreign Keys and then I would to use\nALTER TABLE VALIDATE CONSTRAINT ...\n\nI didn't find how to invalidate constraint without direct update\npg_constraint.\n\nIs there some clean w... |
[
{
"msg_contents": "I implemented IF NOT EXISTS option for CREATE SERVER and CREATE USER \nMAPPING statements\nfor one of our customers.\nI think other users can also find it useful for scripting and automated \ntasks.\nThe patches themselves are small and simple. Documentation is updated as \nwell.\n\n-- \nAnas... |
[
{
"msg_contents": "Fix cardinality estimates for parallel joins.\n\nFor a partial path, the cardinality estimate needs to reflect the\nnumber of rows we think each worker will see, rather than the total\nnumber of rows; otherwise, costing will go wrong. The previous coding\ngot this completely wrong for parall... |
[
{
"msg_contents": "Fix a bug in how we generate partition constraints.\n\nMove the code for doing parent attnos to child attnos mapping for Vars\nin partition constraint expressions to a separate function\nmap_partition_varattnos() and call it from the appropriate places.\nDoing it in get_qual_from_partbound(),... |
[
{
"msg_contents": "Hi\n\nIn last discussion related to PLpgSQL was mentioned weak of stack trace\nPLpgSQL. The function parameters are not printed.\n\nCREATE OR REPLACE FUNCTION public.foo(a double precision, b double\nprecision)\n RETURNS double precision\n LANGUAGE plpgsql\nAS $function$\nbegin\n return 100/... |
[
{
"msg_contents": "Not sure how many people still use [1], as referenced by our git \nwiki[2], but it appears git worktrees are a viable replacement for that \ntechnique. In short, if you're already in your checkout:\n\ngit worktree add ../9.6 REL9_6_STABLE\n\nwould give you a checkout of 9.6 in the ../9.6 dire... |
[
{
"msg_contents": "Hi,\n\nwhile working on [1] (works, harmless regression check failures aside,\nneeds cleanup), I noticed $subject. Without fixing EXPR_KIND_VALUES to\nset p_hasTargetSRFs the query has Query->hasTargetSRF wrongly set.\nWhich in turn breaks your planner code, as it's not being triggered\nanym... |
[
{
"msg_contents": "Hi,\n\nAttached patch fixes comment typos in condition_variable.c\n\nRegards,\n\n--\nMasahiko Sawada\nNIPPON TELEGRAPH AND TELEPHONE CORPORATION\nNTT Open Source Software Center\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo make changes to your subscriptio... |
[
{
"msg_contents": "Test:\n------\ncreate table seq_tab(t int);\ninsert into seq_tab select generate_series(1, 10000000);\nselect count(distinct t) from seq_tab;\n\n#0 0x000000000094a9ad in pfree (pointer=0x0) at mcxt.c:1007\n#1 0x0000000000953be3 in mergeonerun (state=0x1611450) at tuplesort.c:2803\n#2 0x000... |
[
{
"msg_contents": "I am new here and I really want to contribute, I have read same resource\nthat help understanding database system and postgresql. I would like to\nstart implementing sql syntax corresponding by clause because I believe\nimplementing sql syntax gives an opportunity to familiarize many part of... |
[
{
"msg_contents": "Fix an assertion failure related to an exclusive backup.\n\nPreviously multiple sessions could execute pg_start_backup() and\npg_stop_backup() to start and stop an exclusive backup at the same time.\nThis could trigger the assertion failure of\n\"FailedAssertion(\"!(XLogCtl->Insert.exclusiveB... |
[
{
"msg_contents": "Hi there,\n\nI just bought ISO/IEC 9075-2:2016\nhttp://www.iso.org/iso/home/store/catalogue_tc/catalogue_detail.htm?csnumber=63556\nto satisfy my interest on json support in SQL. I am not ready to discuss\nhere implementation details, but there is one problem with the status of\nthis documen... |
[
{
"msg_contents": "Hello,\n\nThere are two days left in the PGCon 2017 CFP, which closes on 19 January.\nPlease get your submissions in soon.\n\nPGCon 2017 will be on 23-26 May 2017 at University of Ottawa.\n\n* 23-24 (Tue-Wed) tutorials\n* 24 (Wed) The Unconference\n* 25-26 (Thu-Fri) talks - the main part of t... |
[
{
"msg_contents": "Generate fmgr prototypes automatically\n\nGen_fmgrtab.pl creates a new file fmgrprotos.h, which contains\nprototypes for all functions registered in pg_proc.h. This avoids\nhaving to manually maintain these prototypes across a random variety of\nheader files. It also automatically enforces ... |
[
{
"msg_contents": "Hi hackers,\n\nI'd like to present a series of patches that implements k-Nearest \nNeighbors (kNN)\nsearch forbtree, which can be usedto speed up ORDER BY distance queries \nlike this:\nSELECT * FROM eventsORDER BY date <-> '2000-01-01'::date ASC LIMIT 100;\nNow only GiST supports kNN, but kN... |
[
{
"msg_contents": "Add function to import operating system collations\n\nMove this logic out of initdb into a user-callable function. This\nsimplifies the code and makes it possible to update the standard\ncollations later on if additional operating system collations appear.\n\nReviewed-by: Andres Freund <andr... |
[
{
"msg_contents": "I looked into bug #14504,\nhttps://www.postgresql.org/message-id/20170118144828.1432.52823@wrigleys.postgresql.org\n\nThe problem is that timestamp_zone_transform() has the cute idea that\nit can simplify timezone('UTC', timestamptzvalue) into a RelabelType\nthat claims the timestamptzvalue i... |
[
{
"msg_contents": "On Tue, Jan 10, 2017 at 4:17 AM, Amit Langote\n<Langote_Amit_f8@lab.ntt.co.jp> wrote:\n> On 2017/01/10 14:44, Keith Fiske wrote:\n>> Is there any reason for the exclusion of parent tables from the pg_tables\n>> system catalog view? They do not show up in information_schema.tables as\n>> well.... |
[
{
"msg_contents": "Attached patch fixes typo in regression test src/test/regress/sql/create_index.sql\nthat was introduced in commit a4523c5\n(\"Improve planning of btree index scans using ScalarArrayOpExpr quals.\").\n\nIn this commit the following lines were added to create_index.sql:\n\nSET enable_indexonlys... |
[
{
"msg_contents": "On Wed, Jan 11, 2017 at 10:53 PM, Amit Langote\n<Langote_Amit_f8@lab.ntt.co.jp> wrote:\n> On 2017/01/06 20:23, Amit Langote wrote:\n>> On 2017/01/05 3:26, Robert Haas wrote:\n>>> It's unclear to me why we need to do 0002. It doesn't seem like it\n>>> should be necessary, it doesn't seem like... |
[
{
"msg_contents": "https://www.postgresql.org/message-id/CAJEAwVE4UAmm8fr%2BNW8XTnKV6M--ACoNhL3ES8yoKL2sKhbaiw%40mail.gmail.com\n\nLet me re-summarize what's been done here to make sure I understand:\n\nEach key in GIN has its own posting tree, which is itself a btree,\nholding all of the tuples that contain th... |
[
{
"msg_contents": "Hi,\n\nHere is a patch that adds temporary column's description to\npg_replication_slots document.\n\nRegards,\n--\nAyumi Ishii\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-hacke... |
[
{
"msg_contents": "On Tue, Jan 3, 2017 at 3:43 AM, Thomas Munro\n<thomas.munro@enterprisedb.com> wrote:\n> Here is a new version of my \"causal reads\" patch (see the earlier\n> thread from the 9.6 development cycle[1]), which provides a way to\n> avoid stale reads when load balancing with streaming replication... |
[
{
"msg_contents": "Hi,\n\nThis may come from my lack of experience with Postgres, but I'm trying to\nextract the byte portion of a Datum that is of type VarBit - bit/bit(n). I\nsee that the Datum pointer contains the value content of the bytes (after a\nfew bytes for the header) but I would need to point to the... |
[
{
"msg_contents": "ProcessStartupPacket() in src/backend/postmaster/postmaster.c contains\nthe following comment:\n\n /*\n * Due to backward compatibility\nconcerns the replication\n * parameter is a hybrid beast whic... |
[
{
"msg_contents": "The travis job https://travis-ci.org/pgjdbc/pgjdbc/jobs/192517342 is seeing\na segfault when we are testing against HEAD with REPLICATION turned on.\n\nLogs can be found here\nhttps://drive.google.com/drive/folders/0B-Heg5ZYCWbreEE4Uk5LdnJ5eWM?usp=sharing\n\nRegards,\n\nDave Cramer\n\nThe tra... |
[
{
"msg_contents": "According to usages, tm_mon (month) field has origin 1, not 0. This is\ndifference from C-library version of struct tm, which has origin 0 for\nreal.\nFor example:\n\n```C\nif (DateOrder == DATEORDER_DMY)\n sprintf(str, \"%02d/%02d\", tm->tm_mday, tm->tm_mon);\nelse\n sprintf(str, \"%02... |
[
{
"msg_contents": "A customer is on 9.6.1, and complains of a segfault observed at least\n3 times. In all cases, the logs look like this:\n\nJan 11 16:11:07 ip-10-0-118-82 kernel: [41913.530453] traps:\npostgres[6561] general protection ip:55fcf08b0491 sp:7ffc17dfa650\nerror:0 in postgres[55fcf0557000+638000]\n... |
[
{
"msg_contents": "Hi Andrew, All,\n\nThere's a buildfarm failure (just on sittella so far) that I can't quite\ninterpret from here. Timing wise it looks like it might be correlated\nwith ea15e18677fc2eff3135023e27f69ed8821554ed - \"Remove obsoleted code\nrelating to targetlist SRF evaluation.\" but I'm not qui... |
[
{
"msg_contents": "Greetings,\n\nThere's some inconsistency when it comes to if we actually use\nSysCacheGetAttr() when pulling an attribute for a tuple we got via\nSearchSysCache(), or if we use heap_getattr().\n\nMaybe I'm missing something, but that seems less than ideal. I've\ngenerally been under the beli... |
[
{
"msg_contents": "Hi,\n\nI've found a mistake in a comment of StrategyNotifyBgWriter\nin freelist.c. bgwriterLatch was replaced by bgwprocno in\nthe following commit, but this is remained in the comment.\n\ncommit d72731a70450b5e7084991b9caa15cb58a2820df\nAuthor: Andres Freund <andres@anarazel.de>\nDate: Thu... |
[
{
"msg_contents": "skink has been unhappy since commit d26fa4f went in, but I think\nthat just exposed a pre-existing bug. Running valgrind here\nduplicates the failure:\n\n==00:00:02:01.653 16626== Conditional jump or move depends on uninitialised value(s)\n==00:00:02:01.653 16626== at 0x4BDF6B: btint4cmp ... |
[
{
"msg_contents": "Logical replication\n\n- Add PUBLICATION catalogs and DDL\n- Add SUBSCRIPTION catalog and DDL\n- Define logical replication protocol and output plugin\n- Add logical replication workers\n\nFrom: Petr Jelinek <petr@2ndquadrant.com>\nReviewed-by: Steve Singer <steve@ssinger.info>\nReviewed-by: ... |
[
{
"msg_contents": "Hi,\n \nI'm getting this error building docs (from commit \ne4c27f5befbfc80a1bf96fc93256dce08b148238):\n \n$ make docs \n make -C doc all\n make[1]: Entering directory '/home/andreak/dev/postgresql/doc'\n make -C src all\n make[2]: Entering directory '/home/andreak/dev/postgresql/doc/src'\n ... |
[
{
"msg_contents": "logical-replication.sgml.diff changes\n\nErik Rijkers\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, 20 Jan 2017 17:00:34 +0100",
"msg_from": "Eri... |
[
{
"msg_contents": "I see a consistent failure while running \"make -C\nsrc/test/modules/commit_ts/ check\" on a server compiled with\n--enable-tap-tests. This is on the master branch (193a7d791)\n\nt/004_restart.pl ....\n1..16\nok 1 - getting ts of InvalidTransactionId reports error\nok 2 - expected error from ... |
[
{
"msg_contents": "Hi,\n\nIn src/backend/replication/logical/worker.c:\n\n pq_sendbyte(reply_message, 'r');\n pq_sendint64(reply_message, recvpos); /* write */\n pq_sendint64(reply_message, flushpos); /* flush */\n pq_sendint64(reply_message, writepos); /* apply */\n\nIs 'writepos' r... |
[
{
"msg_contents": "Is it time to enable checksums by default, and give initdb a switch to turn\nit off instead?\n\nI keep running into situations where people haven't enabled it, because (a)\nthey didn't know about it, or (b) their packaging system ran initdb for\nthem so they didn't even know they could. And o... |
[
{
"msg_contents": "Hi,\n\nWith the pdf increasingly in readability-decline (more and more text \nparts fall off the (right) side of the page and quite a few tables \ncontain unreable bits) I would like to have a single page html. (cf the \nbash scipting guide at http://tldp.org/guides.html even if that is \nsm... |
[
{
"msg_contents": "On Sun, Jan 22, 2017 at 12:18 AM, Petr Jelinek\n<petr.jelinek@2ndquadrant.com> wrote:\n> On 21/01/17 11:39, Magnus Hagander wrote:\n>> Is it time to enable checksums by default, and give initdb a switch to\n>> turn it off instead?\n>\n> I'd like to see benchmark first, both in terms of CPU an... |
[
{
"msg_contents": "So, that post I made about checksums certainly spurred a lot of discussion\n:) One summary is that life would be a lot easier if we could turn\nchecksums on (and off) without re-initdbing. I'm breaking out this\nquestion into this thread to talk about it separately.\n\n\nI've been toying wit... |
[
{
"msg_contents": "Hi,\n\nThere's an issue which I haven't seen documented as expected\nbehaviour, where replicating data to a table which has a foreign key\nresults in a replication failure. This produces the following log\nentries:\n\nLOG: starting logical replication worker for subscription \"contacts_sub\... |
[
{
"msg_contents": ">\n>\n> Fabien is pressed for time, so I've been speaking with him out-of-thread\n> about how I should go about implementing it.\n>\n> The v1 patch will be \\if <expr>, \\elseif <expr>, \\else, \\endif, where\n> <expr> will be naively evaluated via ParseVariableBool().\n>\n> \\ifs and \\endif... |
[
{
"msg_contents": "Hi all,\n\nAs now wal_level = replica has become the default for Postgres 10,\ncould we consider as well making replication connections enabled by\ndefault in pg_hba.conf? This requires just uncommenting a couple of\nlines in pg_hba.conf.sample.\nThanks,\n-- \nMichael\n\n\n\n-- \nSent via pgs... |
[
{
"msg_contents": "Hi all,\n\nWhen spawning a new instance, I found the following thing, which is\nsurprising at first sight:\npostgres: bgworker: logical replication launcher\n\nThere is perhaps no problem to keep that enabled by default until the\nrelease 10 wraps to give it some buildfarm coverage similarly ... |
[
{
"msg_contents": "While working on a bug in postgres_fdw, I found this:\n\n NestLoop\n -> Seq Scan on SmallTable1 A\n NestLoop\n -> Seq Scan on SmallTable2 B\n NestLoop\n -> Index Scan using XYIndex on LargeTable C\n Index... |
[
{
"msg_contents": "Hi,\n\nI've developed a decoding plugin and tested in a Windows environment. In\nWindows, what I really have to do is build a library project from Visual\nStudio and copy the dll into the lib folder of Postgres.\n\nI am now in the process of testing the same code under Linux. From what I've\n... |
[
{
"msg_contents": "So as mentioned on IRC, I have a patch that I am working to rebase on HEAD with the following design. It is very similar to what you have proposed, so maybe we can use my development notes as a jumping-off point for discussion/refinement.\n\n* Incremental Checksums\n\nPostgreSQL users should... |
[
{
"msg_contents": "I was giving some thought to how psql handles undefined variables.\n\nI would like an option where either psql can provide an alternate value\nwhen an undefined variable is referenced, or a way to detect that a\nspecific variable is undefined and replace it with a defined variable.\n\nMy firs... |
[
{
"msg_contents": "Another suggestion to come out of the \\quit_if discussion was that someone\nmight want to send a return code other than 0 back. Something like this:\n\n\\quit 4\n\nor even:\n\n\n\\set exit_code 127\n\\quit :exit_code\n\nThis isn't a personal need of mine, but I figured it was an idea worth\n... |
[
{
"msg_contents": "Hi,\n\nthis patch is based on discussions related to plpgsql2 project.\n\nCurrently we cannot to control plan cache from plpgsql directly. We can use\ndynamic SQL if we can enforce oneshot plan - but it means little bit less\nreadable code (if we enforce dynamic SQL from performance reasons).... |
[
{
"msg_contents": "Dear hackers,\n\nCurrently, I am developing a backend index access method for my research\nproject.\n\nI built corresponding routines such as MyIndexbuild, MyIndexInsert, and so\non and put them in \"src/backend/access/hippo\" (hippo is my index's name). I\nalso added new entries in correspon... |
[
{
"msg_contents": "src/test/README wasn't updated for the new src/test/subscription entry.\n\nPatch attached.\n\n-- \n Craig Ringer http://www.2ndQuadrant.com/\n PostgreSQL Development, 24x7 Support, Training & Services\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.... |
[
{
"msg_contents": "Attached is an alternative implementation of MATERIALIZED VIEWs.\n\nThe idea is to explore possible enahncements to the PostgreSQL\nMATERIALIZED VIEW features.\n\nFeatures:\n\n - All SQL-coded.\n\n - Keeps history of deltas computed at each refresh.\n\n - Allows DMLs of the materialized view,... |
[
{
"msg_contents": "psql(1) does not output notifications asynchronously, as it does not\ncheck for them when idle. This makes it difficult to script handling of\nNOTIFYs.\n\nAttached is pqasyncnotifier.c, a simple command that allows one to\nhandle NOTIFYs asynchronously.\n\nCheers,\n\nNico\n-- \n\n\n-- \nSent... |
[
{
"msg_contents": "According to the documentation for COPY IN mode, \"If the COPY command\nwas issued via an extended-query message, the backend will now discard\nfrontend messages until a Sync message is received, then it will issue\nReadyForQuery and return to normal processing.\" I added a similar\nnote to ... |
[
{
"msg_contents": "Some contrib functions fail to fail sooner when relations of unsupported\nrelkinds are passed, resulting in error message like one below:\n\ncreate table foo (a int);\ncreate view foov as select * from foo;\nselect pg_visibility('foov', 0);\nERROR: could not open file \"base/13123/16488\": N... |
[
{
"msg_contents": "Hi,\n\nWe are doing an assessment for for migrating our Perl applications to Windows 2016 server.\nI am trying to install PostgreSQL 8.2 version on my Windows server 2016. But it is giving me following error:\nMalformed permissions property: 'langid'\n\nWe could not find any relavant informat... |
[
{
"msg_contents": "So I was thinking about various annoying admin/security issues\nrecently, so I came up with this: a new type of user called a\n“superowner”. It’s somewhere between a superuser and a normal user.\nSuperowner would own all objects defined by users, so it would do\nuseful things in contexts wh... |
[
{
"msg_contents": "Hi Hackers,\n\nWe are running Postgres at AIX and encoountered two strqange problems: \nactive zombies process and deadlock in XLOG writer.\nFirst problem I will explain in separate mail, now I am mostly \nconcerning about deadlock.\nIt is irregularly reproduced with standard pgbench launched... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.