Split (1)

train · 58.4k rows

threads listlengths 1 2.99k
[ { "msg_contents": "Running pg_basebackup with a throttling of say 10M runs it into the risk of\nthe I/O on the server actually being slower than pg_basebackup (I have\npreproduced similar issues on fake-slow disks with lower rate limits).\n\nWhat happens in this case in basebackup.c is that the value for \"slee...
[ { "msg_contents": "While poking around the pg_basebackup code, I noticed there are a lot of\nerror messa\n\n-- \n Magnus Hagander\n Me: http://www.hagander.net/\n Work: http://www.redpill-linpro.com/\n\nWhile poking around the pg_basebackup code, I noticed there are a lot of error messa-- Magnus Hagander Me: h...
[ { "msg_contents": "I would like to propose extending composite types with constraints.\nCurrently there is a preliminar patch for not null constraints.\n\n===\nChanges :\n -Use the parser from create table to get the constraints\n'OptTableElementList' instead of 'OptTableFuncElementList'.\n - Add a new transf...
[ { "msg_contents": "Hi,\n\nWhen the number of sync standbys is set to 0 in s_s_names, the assersion\nfailure happens as follows. This means that current multiple syncrep code\nassumes that the num of sync standbys must be greater than 0. But we forgot\nto forbid users from setting that num to 0. This is an overs...
[ { "msg_contents": "On Thu, Dec 15, 2016 at 10:40 AM, Dmitry Ivanov <d.ivanov@postgrespro.ru> wrote:\n> Hi everyone,\n>\n> Looks like \"sql_inheritance\" GUC is affecting partitioned tables:\n>\n> explain (costs off) select * from test;\n> QUERY PLAN ------------------------------\n> Append\n> ...
[ { "msg_contents": "Simplify LWLock tranche machinery by removing array_base/array_stride.\n\narray_base and array_stride were added so that we could identify the\noffset of an LWLock within a tranche, but this facility is only very\nmarginally used apart from the main tranche. So, give every lock in\nthe main ...
[ { "msg_contents": "Hi\n\nI am trying to fix slow query on PostgreSQL 9.5.4.\n\nThe data are almost in RAM\n\nI have a problem with too low cost slow Bitmap index scan on date column,\nthat returns 300K rows.\n\nSlow part\n-> Bitmap Heap Scan on \"Zasilka\" (cost=5097.39..5670.64 rows=1 width=12)\n(actual time...
[ { "msg_contents": "Hi all,\n\nWhile going through the temporary replication slot commits, I have\nnoticed two typos as attached.\nThanks,\n-- \nMichael\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-...
[ { "msg_contents": "It seems that there is a bug in CREATE OR REPLACE VIEW's handling of\nWITH CHECK OPTION (noticed while thinking about the recent change to\npg_dump's handling of circular dependencies in views -- d8c05af). If\nyou use CREATE OR REPLACE VIEW on a view that isn't auto-updatable and\nturn it int...
[ { "msg_contents": "I've been thinking about how to fix the problem Andreas Seltenreich\nreported at\nhttps://postgr.es/m/87eg1y2s3x.fsf@credativ.de\n\nThe core of that problem is that the phrase-search patch attempts to\nrestructure tsquery trees so that there are no operators underneath a\nPHRASE operator, exc...
[ { "msg_contents": "Hello,\n\nI have been trying to build Postgres and migrate my data to the newest\nversion. Postgres builds just fine, but I also need the uuid-ossp module,\nwhich used to build fine for me and now does not...\n\nI am currently \"git pull\"ed to commit b645a05fc6112a4857ceac574d4aa2\n4174a704...
[ { "msg_contents": "Hi,\n\nI don't think PG works on any linux without dlopen(). And DLD (what's\nused in the dlopen replacement) hasn't been maintained in a while.\n\nSee https://www.gnu.org/software/dld/\n\nShouldn't we just remove that code?\n\nGreetings,\n\nAndres Freund\n\n\n-- \nSent via pgsql-hackers mail...
[ { "msg_contents": "Hi,\n\nas the logical replication patch is progressing closer to integration it\nseems to be good time to revisit what to do about preexisting data.\n\nThis patch implements data synchronization for the logical replication.\nIt works both for initial setup of subscription as well as for table...
[ { "msg_contents": "Hello, recently one of my customer stumbled over an immoderate\ncatcache bloat.\n\nThis is a known issue living on the Todo page in the PostgreSQL\nwiki.\n\nhttps://wiki.postgresql.org/wiki/Todo#Cache_Usage\n> Fix memory leak caused by negative catcache entries\n\nhttps://www.postgresql.org/m...
[ { "msg_contents": "Hi,\n\nIsn't it better to forbid the conbination of the options \"-D -\", \"-F t\" and\n\"-X stream\" in pg_basebackup? This is obviously invalid setting and the docs\nwarns this as follows. But currently users can specify such setting and\npg_basebackup can exit unexpectedly with an error.\n...
[ { "msg_contents": "\nHello,\n\nWhile investigating a performance issue, I tried to get informations from \npg_stat_statements, however I ran into another issue: it seems that when \nusing combined queries pg_stat_statements query normalization does not \nwork properly... 2 queries that should have been mapped t...
[ { "msg_contents": "Add pg_sequence system catalog\n\nMove sequence metadata (start, increment, etc.) into a proper system\ncatalog instead of storing it in the sequence heap object. This\nseparates the metadata from the sequence data. Sequence metadata is now\noperated on transactionally by DDL commands, wher...
[ { "msg_contents": "\nRecently a client was confused because there was a substantial \ndifference between the reported table_len of a table and the sum of the \ncorresponding tuple_len, dead_tuple_len and free_space. The docs are \nfairly silent on this point, and I agree that in the absence of \nexplanation it ...
[ { "msg_contents": "I just had a client issue with table bloat that I traced back to a\nstale xmin value in a replication slot. xmin value from hot standby\nfeedback is stored in replication slot and used for vacuum xmin\ncalculation. If hot standby feedback is turned off while walreceiver\nis active then the xm...
[ { "msg_contents": "Hi Postgres Devs,\n\nI had a suggestion regarding the output pg_ctl gives when you use it to\nstart the postgres server. At first I was going to write a patch, but then\nI decided to just ask you guys first to see what you think.\n\nI had an issue earlier where I was trying to upgrade my pos...
[ { "msg_contents": "Why do functions that accept composite types delay type resolution until \nexecution? I have a naive patch that speeds up plpy.execute() by 8% by \ncaching interred python strings for the dictionary key names (which are \nrepeated over and over). The next step is to just pre-allocate those \n...
[ { "msg_contents": "I've been looking at the performance of SPI calls within plpython. \nThere's a roughly 1.5x difference from equivalent python code just in \npulling data out of the SPI tuplestore. Some of that is due to an \ninefficiency in how plpython is creating result dictionaries, but fixing \nthat is u...
[ { "msg_contents": "In commit b30d3ea824c5ccba43d3e942704f20686e7dbab8, when Andres added\nthe simplehash stuff, he also added SH_TYPE, SH_ITERATOR, and\nSH_STATUS to typedefs.list. When I subsequently updated typedefs.list\nfrom the buildfarm in acddbe221b084956a0efd6e4b6c6586e8fd994d7, it of\ncourse removed t...
[ { "msg_contents": "Hi hackers,\n\nI would be good if it would be possible to quickly set NOT NULL for an\nexisting column in a table\nthat have no rows where the column IS NULL and where there is a full\nindex on the column\nallowing the logics to quickly understand there are no NULL values,\nand just have to t...
[ { "msg_contents": "Hi,\nI am starting this as a separate thread for this since the declarative\npartitioning thread has many issues reported and it's better to keep\nthis discussion separate from the issues reported on that thread.\n\nWhile expanding inheritance, any inheritance hierarchy is flattened\nout incl...
[ { "msg_contents": "Hi All,\n\nI have introduced a new function 'pgstathashindex()' inside pgstatuple\nextension to view the statistics related to hash index table. I could\nhave used 'pgstattuple()' function to view hash index stats instead of\nadding this new function but there are certain limitations when usi...
[ { "msg_contents": "Hello to everybody,\nI had a problem with a Postgresql 9.3.5 on 32 bit linux, old 2.6.26 kernel:\nProgram: pg_dump\nProblem: if you have tables with big blob fields and you try to dump them with --inserts you could get errors like\n pg_dump: [archiver (db)] query failed: lost synchronizati...
[ { "msg_contents": "We have a report in pgsql-general of a dblink query failing with\n\tERROR: unknown error\nThis is, to say the least, unhelpful. And it violates our error\nmessage style guidelines.\n\nWhere that is coming from is a situation where we've failed to extract\nany primary message from a remote er...
[ { "msg_contents": "Hi all,\n\n2PC files are created using RecreateTwoPhaseFile() in two places currently:\n- at replay on a XLOG_XACT_PREPARE record.\n- At checkpoint with CheckPointTwoPhase().\n\nNow RecreateTwoPhaseFile() is careful to call pg_fsync() to be sure\nthat the 2PC files find their way into disk. B...
[ { "msg_contents": "Hello,\n\nWhile investigating some other issue, we found a trivial bug of ecpg. The attached is a fix for that.\n\nIf you specify an input file which ends with \".\" (e.g. run \"ecpg file.\"), ecpg writes one byte past the end of the allocated memory.\n\nIn addition, the following statement ...
[ { "msg_contents": "There seems to be some base bug in _hash_doinsert().\n+ * XXX this is useless code if we are only storing hash keys.\n\n+ */\n\n+ if (itemsz > HashMaxItemSize((Page) metap))\n\n+ ereport(ERROR,\n\n+ (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),\n\n+ e...
[ { "msg_contents": "Hello all,\n\nIn a recent post[1] by Robert, wait events for different LWLOCKS have been\nanalyzed. The results clearly indicate a significant lock contention\noverhead on WAL Write locks. To get an idea of this overhead, we did the\nfollowing two tests.\n\n1. Hacked the code to comment out W...
[ { "msg_contents": "I looked into the problem reported in bug #14472,\nhttps://www.postgresql.org/message-id/20161221214744.25622.71454@wrigleys.postgresql.org\nAlthough the submitter thought it was related to bug #14414, it isn't\nparticularly. The failure scenario is that the input value to a\nCoerceToDomain ...
[ { "msg_contents": "Skip checkpoints, archiving on idle systems.\n\nSome background activity (like checkpoints, archive timeout, standby\nsnapshots) is not supposed to happen on an idle system. Unfortunately\nso far it was not easy to determine when a system is idle, which\ndefeated some of the attempts to avoid...
[ { "msg_contents": "Currently an Append plan node does not execute its subplans in\nparallel. There is no distribution of workers across its subplans. The\nsecond subplan starts running only after the first subplan finishes,\nalthough the individual subplans may be running parallel scans.\n\nSecondly, we create ...
[ { "msg_contents": "Hi,\n\nfor testing - like yesterday's 6ef2eba3f - it's annoying that\ncheckpoint_timeout has 30s minimum. I've now locally patched that to be\n1s a significant number of times.\n\nWhile it's not a particularly good idea to set it to 1s on a production\nsystem, I don't see why we need to preve...
[ { "msg_contents": "Hi,\n\nWhile running sqlsmith against PG v10 , found a crash . Not sure \nwhether it is reported earlier or not . Please refer the standalone \ntestcase for the same -\n\n[centos@tusharcentos7 bin]$ ./psql postgres -p 9000\npsql (10devel)\nType \"help\" for help.\n\npostgres=# select\npostg...
[ { "msg_contents": "Recently, we find PG fails to generate an effective plan for following SQL:\r\n select * from (select * from table1 execpt select * from table2) as foo where foo.a > 0;\r\n Because PG does not pushdown qual to the none of the subquery. And I check the source code, find some comments in ...
[ { "msg_contents": "Pursuant to the report here:\nhttps://www.postgresql.org/message-id/7d0809ee-6f25-c9d6-8e74-5b2967830d49@enterprisedb.com\nI tried to test all the built-in functions that take \"text\" (rather\nthan \"name\") arguments representing cataloged objects. I was able\nto provoke the same assertion...
[ { "msg_contents": "I am seeing this compiler warning in the 9.4 branch:\n\n 9.4: basebackup.c:1284:6: warning: variable 'wait_result' set but not used [-Wunused-but-set-variable]\n\nThis is on Debian Jessie with gcc version 4.9.2. It is from this commit:\n\n\tcommit f6508827afe76b2c3735a9ce073620e708d60c79\n\...
[ { "msg_contents": "Hi hackers,\n\nWe already have xact_start, query_start and backend_start\nto get the timestamptz for when different things happened.\n\nI would like to propose adding a fourth such column, \"waiting_start\",\nwhich would tell how long time a backend has been waiting.\n\nThe column would be NU...
[ { "msg_contents": "Is there any chance that the PostgreSQL developers would accept a new\ncluster wide configuration option to control how the system handles\nsymbol case folding? \n\nCurrently PostgreSQL folds all un-quoted symbols to lower case. \n\nWe would like to add a global configuration option with a na...
[ { "msg_contents": "On December 24, 2016 9:52 PM Craig Ringer\n[mailto:craig.ringer@2ndquadrant.com] wrote:\n> Personally I can see such an option being ok as an initdb-time setting\nor at CREATE DATABASE time. Case folding can know the current db from\nglobal context.\n>\n> It'd have to be fast though. Very fa...
[ { "msg_contents": "On Saturday, December 24, 2016 10:49 PM Tom Lane\n[mailto:tgl@sss.pgh.pa.us] wrote:\n> No. This has been looked into repeatedly in the past, and we simply\ndon't want to deal with it. \n\nFair enough. We will not pursue the issue then. That is why I asked.\n\nIan Lewis (www.mstarlabs.com)\n...
[ { "msg_contents": "Tom Lane [mailto:tgl@sss.pgh.pa.us] wrote:\n> Quite aside from the impact on the server (which would be extensive),\nit would break every nontrivial application, and force them all to try\nto deal with each possible folding behavior.\nI have read through the various threads related to this i...
[ { "msg_contents": "'the the' -> 'the'\n\nand\n\n'ie' -> 'i.e.'\n\nAlthough (concening the latter change) the present counts are 'ie' 428, \nand 'i.e.' 428.\nso it might be debatable (but let's not)\n\nthanks,\n\nErik Rijkers\n\n\n\n\n-- \nSent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)\nTo ma...
[ { "msg_contents": "I suspect the following is a bug:\n\ncreate table foo (a int) with oids;\nCREATE TABLE\ncreate table bar (a int);\nCREATE TABLE\nalter table bar inherit foo;\nERROR: table \"bar\" without OIDs cannot inherit from table \"foo\" with OIDs\n\nalter table bar set with oids;\nALTER TABLE\nalter t...
[ { "msg_contents": "In postmaster.c signal handler pmdie() calls ereport() and\nerrmsg_internal(), which could call palloc() then malloc() if\nnecessary. Because it is possible that pmdie() gets called while\nmalloc() gets called in postmaster, I think it is possible that a\ndeadlock situation could occur throug...
[ { "msg_contents": "Hi all,\n\nSince 56c7d8d4, pg_basebackup supports tar format when streaming WAL\nrecords. This has been done by introducing a new transparent routine\nlayer to control the method used to fetch WAL walmethods.c: plain or\ntar.\n\npg_receivexlog does not make use of that yet, but I think that i...
[ { "msg_contents": "Hi all,\n\nAs mentioned a couple of weeks back, I am fine to work as commit fest\nmanager for 2017-01.\n\nHere is the commit fest status at the moment I am writing this email:\nNeeds review: 68.\nWaiting on Author: 17.\nReady for Committer: 18.\nCommitted: 27.\nRejected: 2.\nReturned with Fee...
[ { "msg_contents": "Hi\n\nI reread ideas described on page https://github.com/trustly/plpgsql2\n\nSome points are well and can be benefit for PlpgSQL.\n\nFirst I describe my initial position. I am strongly against introduction\n\"new\" language - plpgsql2 or new plpgsql, or any else. The trust of\ndevelopers to ...
[ { "msg_contents": "Hi,\nWe report planning and execution time when EXPLAIN ANALYZE is issued.\nWe do not have facility to report planning time as part EXPLAIN\noutput. In order to get the planning time, one has to issue EXPLAIN\nANALYZE which involves executing the plan, which is unnecessary.\n\nWe report plann...
[ { "msg_contents": "Folks,\n\nOne of our hidden treasures is the hook system, documented only in\nrandom presentations, if you can find them, and in the source code, if\nyou know to look.\n\nI'd like to document the hooks that we consider public APIs.\n\nTo do this, I need to figure out whether there are hooks t...
[ { "msg_contents": "Hi,\n\ntesting master as of fe591f8bf6 produced a crash reading\npg_stat_activity (backtrace below). Digging around with with gdb\nrevealed that pgstat_get_wait_event() returned an invalid pointer for a\nclassId PG_WAIT_LWLOCK.\n\nI think the culprit is dsa.c passing a pointer to memory that...
[ { "msg_contents": "Hi,\n\nSync rep with multiple standbys allows queries run on standbys to see\ntransactions that haven't been flushed on the configured number of\nstandbys. That means that it's susceptible to lost updates or a kind\nof \"dirty read\" in certain cluster reconfiguration scenarios. To\nclose t...
[ { "msg_contents": "Here is a patch to add some information about the systemd RemoveIPC\nissue to the documentation, sort of in the spirit of the OOM discussion\nnearby.\n\n-- \nPeter Eisentraut http://www.2ndQuadrant.com/\nPostgreSQL Development, 24x7 Support, Remote DBA, Training & Services\n\n\n\...
[ { "msg_contents": "> \r\n> I'm looking forward to seeing your patch.\r\n>\r\n\r\nI created a patch.\r\n\r\nI marked [WIP] to the title because some documentation is lacked and format needs some fixing.\r\n\r\nI'm going to post this patch to the January CF. \r\nBut it's my first time to send a patch so please ex...
[ { "msg_contents": "Currently, queries that have references to SubPlans or\nAlternativeSubPlans are considered parallel-restricted. I think we\ncan lift this restriction in many cases especially when SubPlans are\nparallel-safe. To make this work, we need to propagate the\nparallel-safety information from path...
[ { "msg_contents": "By seeing the subject line, one might wonder why we need to consider\nparallelizing the queries containing initplans differently from\nqueries containing subplans considering that I have posted a mail to\nachieve later a few hours back. The reason is that both are treated\ndifferently with r...
[ { "msg_contents": "Hi all\n\nThere's a minor race between commit_ts SLRU truncation and concurrent\ncommit_ts lookups, where a lookup can check the lower valid bound xid\nwithout knowing it's already been truncated away. This would result in\na SLRU lookup error.\n\nIt's pretty low-harm since it's hard to trigg...
[ { "msg_contents": "Here is a small cleanup patch to make more use of the RoleSpec\nstruct/node throughout the parser to avoid casts and make the code more\nself-documenting.\n\n-- \nPeter Eisentraut http://www.2ndQuadrant.com/\nPostgreSQL Development, 24x7 Support, Remote DBA, Training & Services\n...
[ { "msg_contents": "By chance I happened to notice that the recent partition patch pushed\nus over the number of available node tags between\n\n\tT_A_Expr = 900,\n\nand\n\n\tT_TriggerData = 950,\t\t/* in commands/trigger.h */\n\nSpecifically we now have some of the replication grammar node type\ncodes conflictin...
[ { "msg_contents": "Pursuant to my comments at\nhttps://www.postgresql.org/message-id/20161223192245.hx4rbrxbrwtgwj6i@alvherre.pgsql\nand because of a stupid bug I found in my indirect indexes patch, I\nrewrote (read: removed) HeapSatisfiesHOTAndKey. The replacement\nfunction HeapDetermineModifiedColumns return...
[ { "msg_contents": "\n[Oops, resent, wrong from address, please accept my apologies]\n\nHello Pavel,\n\n> There are two concepts - both can be implemented, and used (can be used\n> together).\n\nThat is one point I would like to ascertain clearly and explicitely, so having \nvarious designs side by side, eg in t...
[ { "msg_contents": "Hi there\r\nAny suggestion how to get around this issue I am having with vacuum command I’m running on 8.1 version of prostgre SQL.\r\nThe VACUUM FULL command seems to get stuck on vacuuming \"pg_catalog.pg_largeobject\" (last message for Verbose)\r\n Now attempting below - but not hopeful t...
[ { "msg_contents": "Hi there\r\nCan this be forwarded to someone who can assist with this query?\r\nThank you, Patrick Hayes\r\nFrom: Hayes, Patrick\r\nSent: 29 December 2016 12:26\r\nTo: 'pgsql-hackers@postgresql.org' <pgsql-hackers@postgresql.org>\r\nSubject: FW: VACUUM FULL Error\r\n\r\nHi there\r\nAny sugges...
[ { "msg_contents": "Is there any reason that src/backend/fr.po is in ISO-8859 whereas all\nthe other translations are in UTF-8?\n\nHere's a quick-fix patch for consistency.\n-- \nVik Fearing +33 6 46 75 15 36\nhttp://2ndQuadrant.fr PostgreSQL : Expertise, Formation et...
[ { "msg_contents": "Hi All,\n\nFujitsu was interested in developing a columnar storage extension with\nminimal\nchanges the server backend.\n\nThe columnar store is implemented as an extension using index access\nmethods.\nThis can be easily enhanced with pluggable storage methods once they are\navailable.\n\nA ...
[ { "msg_contents": "In pursuit of knowledge about clock_gettime(), I installed FreeBSD 10.3\non a PowerBook I had laying about. I was disappointed to find out that\nHEAD fails regression tests on that platform:\n\n*** /usr/home/tgl/pgsql/src/test/regress/expected/lock.out Thu Dec 29 19:37:50 2016\n--- /usr...
[ { "msg_contents": "I've been trying to troubleshoot the cause of this PostGIS recheck bug we\nhave reported by two people so far. The last test was a nice simple\nrepeatable one that triggered the issue:\n\nhttps://trac.osgeo.org/postgis/ticket/3418\n\n\nfrom what I have seen this only affects cases where we a...
[ { "msg_contents": "While trying to find a case where spgist wins over btree for text, I\ncame across the following behavior which I would consider a bug:\n\nCREATE TABLE texts (value text);\nINSERT INTO texts SELECT repeat('a', (2^20)::integer);\nCREATE INDEX ON texts USING spgist (value);\nSET enable_seqscan =...
[ { "msg_contents": "Hi all,\n\nThe attached patch is reworked from a previous one [1] to better deal with\nget_object_address and pg_get_object_address.\n\nRegards,\n\n[1] https://www.postgresql.org/message-id/20150317171836.GC10492@momjian.us\n\n-- \nFabrízio de Royes Mello\nConsultoria/Coaching PostgreSQL\n>> ...
[ { "msg_contents": "During the review of Group update Clog patch [1], Dilip noticed an\nissue with the patch where it can leak the semaphore count in one of\nthe corner case. I have checked and found that similar issue exists\nfor Group clear xid (ProcArrayGroupClearXid) as well. I think the\nreason why this p...
[ { "msg_contents": "During some recent patch reviews, there was some back and forth about\nwhere to put prototypes for fmgr-callable functions. We don't actually\nneed these prototypes except to satisfy the compiler, so we end up\nsprinkling them around random header files.\n\nI figured we could generate these ...
[ { "msg_contents": "A further step toward getting rid of the DSSSL tool chain requirement,\nhere is a patch to change the generation of the text INSTALL file to use\nXLST and Pandoc.\n\nThe change to Pandoc is not essential to this change, but it creates\nmuch better looking output and simplifies the locale/enco...
[ { "msg_contents": "Here is a patch to add 'use strict' to all Perl programs (that I could\nfind), or move it to the right place where it was already there. I\nthink that is a pretty standard thing to do nowadays.\n\nI tried testing the changes in pgcheckdefines, but it just spits out\nnonsense before and after...
[ { "msg_contents": "In order to reduce the number of useless casts and make the useful casts\nmore interesting, here is a patch that automatically casts the result of\ncopyNode() back to the input type, if the compiler supports something\nlike typeof(), which most current compilers appear to. That gets us\nsome...
[ { "msg_contents": "Here is resubmission of patch to implement logical decoding of two-phase transactions (instead of treating them\nas usual transaction when commit) [1] I’ve slightly polished things and used test_decoding output plugin as client.\n\nGeneral idea quite simple here:\n\n* Write gid along with com...
[ { "msg_contents": "Hi\n\nnow the code in tabcomplete become large part of psql. Is there some\npossibility to write regress tests?\n\nI found only this link\n\nhttp://stackoverflow.com/questions/22795767/how-to-test-python-readline-completion\n\nRegards\n\nPavel\n\nHinow the code in tabcomplete become large par...
[ { "msg_contents": "Cycling back to this topic again, but this time at the beginning of a CF.\n\nHere's an actual patch to change:\n\n\nwal_level=replica\nmax_wal_senders=10\nmax_replication_slots=20\n\nBased on feedback from last year (\nhttps://www.postgresql.org/message-id/CABUevEwfV7zDutescm2PHGvsJdYA0RWHFMT...
[ { "msg_contents": "There is a common coding pattern that goes like this:\n\n RestrictInfo rinfo = (RestrictInfo ) lfirst(lc);\n Assert(IsA(rinfo, RestrictInfo));\n\n(Arguably, the Assert should come before the cast, but I guess it's done\nthis way out of convenience.)\n\n(Not to mention the other common...
[ { "msg_contents": "pgbench's -l option is coded using inappropriate familiarity with the\ncontents of struct instr_time. I recall complaining about that when\nthe code went in, but to little avail. However, it needs to be fixed\nif we're to switch over to using clock_gettime() as discussed in\nthe gettimeofda...
[ { "msg_contents": "Hi hackers,\n\nHere is a small patch to add a test exercising SERIALIZABLE READ ONLY\nDEFERRABLE. It shows a well known example of a serialisation anomaly\ncaused by a read-only transaction under REPEATABLE READ (snapshot\nisolation), then shows the different ways that SERIALIZABLE and\nSERI...
[ { "msg_contents": "The following documentation comment has been logged on the website:\n\nPage: https://www.postgresql.org/docs/9.6/static/pgcrypto.html\nDescription:\n\nHello,\r\n\r\nThe C source code of gen_random_uuid reads:\r\n\r\n /\r\n Generate random bits. pg_backend_random() will do here, we do...
[ { "msg_contents": "In the attached patch (snippet below), I'm seeing something strange with \nargs->in.r.atts[]. Prior to entering the PG_TRY block, I can inspect \nthings in lldb just fine:\n\n(lldb) call args->in.r.atts[0].func\n(PLyDatumToObFunc) $49 = 0x000000010fc4dc70 \n(plpython3.so`PLyString_FromDatum a...
[ { "msg_contents": "The recent thread about compiler warnings got me thinking about how it's \nessentially impossible to notice warnings with default make output. \nPerhaps everyone just uses make -s by default, though that's a bit \nannoying since you get no output unless something does warn (and then \nyou don...
[ { "msg_contents": "I originally sent to psql-general some months ago, but it appears it was never\ndelivered (perhaps I wasn't properly subscribed?).\n\nFailed to alter table eric_umts_rnc_utrancell_metrics: ERROR: attribute 361 has wrong type\nDETAIL: Table has type integer, but query expects smallint.\n\nWe...
[ { "msg_contents": "gsstark@gmail.com [mailto:gsstark@gmail.com] On Behalf Of Greg Stark\nwrote:\n\n> But the problem with configurable quoting rules is a bit different.\n> Imagine your application later decides to depend on PostGIS. So you\nload the PostGIS extension and perhaps also some useful functions you\n...
[ { "msg_contents": "From: Robert Haas [mailto:robertmhaas@gmail.com] wrote:\n\n> I'm not sure there's any way to split the baby here: tool authors will\nobviously prefer that PostgreSQL's behavior in this area be invariable,\nwhile people trying to develop portable database applications will\nprefer configurabil...
[ { "msg_contents": "Hi hackers,\n\nHere is a new version of my \"causal reads\" patch (see the earlier\nthread from the 9.6 development cycle[1]), which provides a way to\navoid stale reads when load balancing with streaming replication.\n\nTo try it out:\n\nSet up a primary and some standbys, and put \"causal_r...
[ { "msg_contents": "Hi,\nI'm developing a logical decoding plugin to extract changes from the\ndatabase. This is working pretty well so far, but I have a question on the\narchitectural aspect of the application. I want to filter changes based on\nthe tables that are subscribed for CDC in my application and avoid...
[ { "msg_contents": "Update copyright for 2017\n\nBackpatch-through: certain files through 9.2\n\nBranch\n------\nREL9_2_STABLE\n\nDetails\n-------\nhttp://git.postgresql.org/pg/commitdiff/19371e148207c33d15fded06a178d58d0781141d\n\nModified Files\n--------------\nCOPYRIGHT ...
[ { "msg_contents": "Robert Haas [mailto:robertmhaas@gmail.com] wrote:\n> The issue is, rather, that every extension written for\n> PostgreSQL, whether in or out of core, needs to handle this issue and\n> every general-purpose client tool (pgAdmin, etc.) needs to be aware of\n> it. ....\n\nI can see the accuracy ...
[ { "msg_contents": "Tom Lane [mailto:tgl@sss.pgh.pa.us] wrote:\n>> 2. If the folding mode is chosen through a GUC variable, which\n>> is certainly what people would expect, then it turns out that\n>> it breaks client libraries/applications anyway, because an\n>> installation-wide setting could impose itself on...
[ { "msg_contents": "Hi and happy new year.\n\nThe lazy vacuum calls lazy_cleanup_index to update statistics of\nindexes on a table such as relpages, reltuples at the end of the\nlazy_scan_heap. In all type of indexes the lazy_cleanup_index scans\nall index pages. It happens even if table has not been updated at ...
[ { "msg_contents": "The following bug has been logged on the website:\n\nBug reference: 14486\nLogged by: Per Modin\nEmail address: web+postgresql@modin.io\nPostgreSQL version: 9.6.1\nOperating system: Linux 303a92173594 4.8.15-moby #1 SMP Sat Dec 17 0\nDescription: \n\nFound this bug...
[ { "msg_contents": "My next thought is ALTER SYSTEM support for pg_hba.conf, especially\nsince that would make it easier to do a formal test of Haribabu's\npg_hba view patch by adding each of the options one by one and then\njuggling them.\n\nALTER SYSTEM\nENABLE [LOCAL \| REMOTE] ACCESS\nFOR\n[DATABASE <name> \|...
[ { "msg_contents": "Robert Haas [mailto:robertmhaas@gmail.com] wrote:\n>> Where you get into trouble there is that you might run CREATE\nEXTENSION \n>> from that session\n\nYes. I can see this problem. And, while I can imagine resolving it with\ncontext belonging to the extension, separate from the current sessi...
[ { "msg_contents": "In:\nhttps://www.postgresql.org/docs/devel/static/runtime-config-file-locations.html\n\n---------------------------------------------------\nident_file (string)\n\n Specifies the configuration file for Section 20.2, “User Name Maps” user name mapping (customarily called pg_ident.conf). Thi...
[ { "msg_contents": ">\n>\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...

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.