threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "This patch increases test coverage for pltcl, from 70% to 83%. Aside \nfrom that, the work on this uncovered 2 new bugs (the trigger return one \nI just submitted, as well as a bug in the SRF/composite patch).\n-- \nJim Nasby, Data Architect, Blue Treble Consulting, Austin TX\nExperts in Anal... |
[
{
"msg_contents": "Hi,\n\nWhile debugging something on 9.6, I've noticed that auto_explain handles \nparallel queries in a slightly strange way - both the leader and all the \nworkers log their chunk of the query (i.e. the leader logs explain for \nthe whole query, while workers only log the parts they've been ... |
[
{
"msg_contents": "Hi hackers,\n\nIn PostgreSQL 9.6, hash joins can be parallelised under certain\nconditions, but a copy of the hash table is built in every\nparticipating backend. That means that memory and CPU time are\nwasted. In many cases, that's OK: if the hash table contents are\nsmall and cheap to co... |
[
{
"msg_contents": "Hi All,\n\n\nI didn't find any relevant thread about the discussion of Bulk insert\nsupport in ECPG using arrays. Oracle supports the same and details\nare available in [1].\n\n\nI see some performance benefits in supporting the same in ECPG also.\nDoes any one worked on this area before? Are... |
[
{
"msg_contents": "Here is a patch series that implements several changes in the internal\ngrammar and node representation of function signatures. They are not\nnecessarily meant to be applied separately, but they explain the\nprogression of the changes nicely, so I left them like that for review.\n\nThe end g... |
[
{
"msg_contents": "Hi,\n\nI ran into a typo in a comment in contrib/postgres_fdw/deparse.c. \nPlease find attached a patch.\n\nBest regards,\nEtsuro Fujita\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/p... |
[
{
"msg_contents": "Add make rules to download raw Unicode mapping files\n\nThis serves as implicit documentation and is handy if someone wants to\ntweak things. The rules are not part of a normal build, like this\nentire directory.\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/pg/com... |
[
{
"msg_contents": "Hi,\nConsider this situation:\n1. I have a background worker process.\n2. The process creates a latch, initializes it using InitLatch & resets it.\n3. It then creates a thread and passes the latch created in step 2 to it.\n To pass it, the process uses the last argument of pthread_create.\... |
[
{
"msg_contents": "Hi All,\n\nThe 2016-11 commitfest is officially started. Please add any further\ndevelopment patches into 2017-01 commitfest.\n\nThe current status summary is:\n\nNeeds review: 98\nWaiting on author: 9\nReady for Commiter: 18\nCommited: 19\nMoved to next CF: 0\nRejected: 0\nReturned with feed... |
[
{
"msg_contents": "Hi,\nI send this email after reading the xml2 module « Deprecation notice » in\nhttps://www.postgresql.org/docs/9.6/static/xml2.html.\n\nI still use xml2 module for one thing: xslt_process(). We store some\ndata in xml fields, and we sometime need to correct / upgrade the format.\nBeing able ... |
[
{
"msg_contents": "In 9.6, \"make installcheck\" in contrib/postgres_fdw takes a shade\nunder 3 seconds on my machine. In HEAD, it's taking 10 seconds.\nI am not happy, especially not since there's no parallelization\nof the contrib regression tests. That's a direct consumption of\nmy time and all other devel... |
[
{
"msg_contents": "Hi,\n\nwhile eye-balling some explain plans for parallel queries, I got a bit \nconfused by the row count estimates. I wonder whether I'm alone.\n\nConsider for example a simple seq scan query, which in non-parallel \nexplain looks like this:\n\n QUERY PLAN\n----... |
[
{
"msg_contents": "Hi all\n\nA very common operation that users perform is reloading tables.\nSometimes as part of an ETL process. Sometimes as part of a dump and\nreload. Sometimes loading data from external DBs, etc.\n\nRight now users have to jump through a bunch of hoops to do this efficiently:\n\nBEGIN;\n\... |
[
{
"msg_contents": "Row level security feature implementation in Postgres is through policy and the row security qualifier is attached as a subquery to the main query before query planning. The RLS is implemented through ALTER TABLE STATEMENT. \nBut my doubt is why this feature is not enabled in case of Foreign ... |
[
{
"msg_contents": "Hi,\n\nIt looks like for about 3 years, PL/Java has been calling\nInitializeTimeouts before calling RegisterTimeout. Looking over\nthe callers of InitializeTimeouts in core, though, it appears\nthat an extension like PL/Java should be able to assume it has\nalready been called, and in fact wo... |
[
{
"msg_contents": "Hi Guillaume,\n\nWith your v2 patch, -B options working as expected but --no-blobs\noptions is still unrecognized, this happens is because of you have\nforgot to add entry for 'no-blobs' in long_options[] array.\n\nApart from this concern patch looks good to me. Thanks\n\nRegards,\nAmul\n\n\n... |
[
{
"msg_contents": "Hi,\n\nThere's two things I found while working on faster expression\nevaluation, slot deforming and batched execution. As those two issues\noften seemed quite dominant cost-wise it seemed worthwhile to evaluate\nthem independently.\n\n1) We atm do one ExecProject() to compute each aggregate'... |
[
{
"msg_contents": "libpq: Allow connection strings and URIs to specify multiple hosts.\n\nIt's also possible to specify a separate port for each host.\n\nPreviously, we'd loop over every address returned by looking up the\nhost name; now, we'll try every address for every host name.\n\nPatch by me. Victor Wagn... |
[
{
"msg_contents": "Craig's post yesterday about exposing syntax for disabling indexes reminded\nme of another feature I think we're lacking in areas where we have to do\ntable management.\n\nThe issue is to create a *something* that has the exact permissions of\nanother *something*. Usually it's creating a tabl... |
[
{
"msg_contents": "Hi all,\n\nApologies in advance if this isn't the right place to be posting this.\n\nI've started work on a plugin in C (https://github.com/tanglebones/pg_tuid)\nfor generating generally monotonically ascending UUIDs (aka TUIDs), and\nafter googling around I couldn't find any guidence on a fe... |
[
{
"msg_contents": "Hello Hackers,\n\nI have a question regarding the contents being written to the backup_label\nfile and the .backup file in the pg_wal location generated when the online\nbackup is done.\n\nbackup_label file contents are as follows, which do not contain backup stop\nposition (timestamp and STO... |
[
{
"msg_contents": "pg_xlogdump: Add NLS\n\nReviewed-by: Michael Paquier <michael.paquier@gmail.com>\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/pg/commitdiff/69d590fffbdcfb50a31a8c78ce87e602002a869f\n\nModified Files\n--------------\nsrc/bin/pg_xlogdump/nls.mk | 6 ++++\nsrc/... |
[
{
"msg_contents": "pg_test_timing: Add NLS\n\nAlso straighten out use of time unit abbreviations a bit.\n\nReviewed-by: Michael Paquier <michael.paquier@gmail.com>\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/pg/commitdiff/59fa9d2d9da46097dd4da5c5f1f07e22a288fccf\n\nModified Files\n-... |
[
{
"msg_contents": "Hi\n\nAs noted in a nearby review of some similar code[1], commit 7a2fe9bd\nmade merge append very slightly more efficient, but nobody told\ncost_merge_append about the change. I doubt it makes much difference\nto the final cost in practice but I figured it might be worth\ncorrecting the com... |
[
{
"msg_contents": "Hi,\n\nHere's a patch to remove a copy-and-pasto from xact.h.\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:\nhttp://www.postgresql.org/mailpref/pgsql-hackers",
"msg_da... |
[
{
"msg_contents": "Hi all,\n\nare there plans to introduce temporal tables?\n\nbest,\n\nStefan\n\n\n\n-------- Weitergeleitete Nachricht --------\nBetreff: \tRe: [CORE] temporal tables (SQL2011)\nDatum: \tFri, 4 Nov 2016 10:27:40 -0400\nVon: \tPeter Eisentraut <peter_e@gmx.net>\nAn: \tStefan Scheid <ssc@4brainc... |
[
{
"msg_contents": "The intent of SPI_push/SPI_pop seems to be to draw a boundary line between\nnested layers of SPI callers. Which is fine, but the SPI_connect and\nSPI_finish calls of the inner layer would suffice for that. AFAICS,\nthe only thing that SPI_push/SPI_pop buy for us is the ability to detect\na ... |
[
{
"msg_contents": "Hi guys,\n Although, usually, we do not change the system catalog or modify the\ncatalog schema, or adding a new system catalog, but in these system catalog\nhead files, such as pg_xxx.h, i think we should use more meaningful\nvariables. As we known, in pg_xxx.h files, we insert some initia... |
[
{
"msg_contents": "Hi hackers\n\nHere is an experimental WIP patch to allow SERIALIZABLE READ ONLY\nDEFERRABLE transactions on standby servers without serialisation\nanomalies, based loosely on an old email from Kevin Grittner[1]. I'm\nnot sure how far this is from what he had in mind or whether I've\nmisunder... |
[
{
"msg_contents": "Hi all,\n\nAlthough we restrict the WAL segment size to 64 MB as upper limit, the\nfollowing piece of code in guc.c (line 715) seems confusing to me.\n\n#if XLOG_SEG_SIZE < (1024*1024) || XLOG_BLCKSZ > (1024*1024*1024)\n#error XLOG_SEG_SIZE must be between 1MB and 1GB\n#endif\n\nEither the co... |
[
{
"msg_contents": "Hello,\n\nI was surprised to find that\nsrc/backend/utils/mb/Unicode/Makefile got changed (3a47c70) to\nhave download feature of character conversion authority files. It\nseems fine according to the previous discussion as the commnet of\nthat is saying.\n\nSo, I did 'make maintainer-clean; ma... |
[
{
"msg_contents": "Hi all,\n\nI just bumped into the following:\n--- a/src/backend/commands/event_trigger.c\n+++ b/src/backend/commands/event_trigger.c\n@@ -742,7 +742,7 @@ EventTriggerCommonSetup(Node *parsetree,\n\n /*\n * Filter list of event triggers by command tag, and copy them into our\n- * mem... |
[
{
"msg_contents": "When compiling with --enable-profiling using:\n\ngcc version 5.4.0 20160609 (Ubuntu 5.4.0-6ubuntu1~16.04.2)\n\nI get the warning:\n\nipc.c: In function 'proc_exit':\nipc.c:137:3: warning: ignoring return value of 'chdir', declared with\nattribute warn_unused_result [-Wunused-result]\n chdi... |
[
{
"msg_contents": "Per some discussions with a number of different people at pgconfeu, here is\na patch that changes the default mode of pg_basebackup to be streaming the\nwal, as this is what most users would want -- and those that don't want it\nhave to make other changes as well. Doing the \"most safe\" thin... |
[
{
"msg_contents": "Hi hackers,\n\nCurrently we don't generate parallel plans in SERIALIZABLE. What\nproblems need to be solved to be able to do that? I'm probably\nsteamrolling over a ton of subtleties and assumptions here, but it\noccurred to me that a first step might be something like this:\n\n1. Hand the... |
[
{
"msg_contents": "Hi hackers,\n\nThe SLRU managed by predicate.c can wrap around and overwrite data if\nyou have more than 1 billion active XIDs. That's because when SSI was\nimplemented, slru.c was limited to four digit segment names, which\nimplied a page limit that wasn't enough for pg_serial to have space... |
[
{
"msg_contents": "Hi hackers.\n\nWhile working on jsonbstatistics, I found the following bug:\nan empty jsonb array is considered to be lesser than any scalar,\nbut it is expected that objects > arrays > scalars.\n\n# select '[]'::jsonb < 'null'::jsonb;\n ?column?\n----------\n t\n(1 row)\n\nAttached patch c... |
[
{
"msg_contents": "Hi all,\n\nThe documentation does not explain at all what means \"sync\" or \"async\"\non pg_stat_replication. The paragraph \"Planning for high availability\"\nmentions \"catchup\" and \"streaming\", but it does not say that users can\nrefer to it directly in pg_stat_replication.\n\nThoughts... |
[
{
"msg_contents": "Hi all,\n\nIn brin_doupdate(line 290), REGBUF_STANDARD is used to register\nrevmap buffer reference in WAL record. But, revmap buffer page doesn't\nhave a standard page layout and it doesn't update pd_upper and\npd_lower as well.\n\nEither we should register revmapbuf as following:\nXLogRegis... |
[
{
"msg_contents": "While testing bitmap performance, I have observed that some of the\nTPCH queries taking huge time in BitmapIndexScan node, when there are\nlossy pages.\n\nI suspected 75ae538bc3168bf44475240d4e0487ee2f3bb376 commit, because\nprior to that it used to take very less time. So I tested by reverti... |
[
{
"msg_contents": "Hi, hackers\n\nI wonder if such behavior can be considered as a bug:\n\nknizhnik@knizhnik:~/dtm-data$ psql postgres\npsql (10devel)\nType \"help\" for help.\n\npostgres=# create tablespace fs location '/home/knizhnik/dtm-data/fs';\nCREATE TABLESPACE\npostgres=# set default_tablespace=fs;\nSET... |
[
{
"msg_contents": "The system will let you set the \"user_catalog_table\" reloption to \"true\"\non a materialized view. Is this sensible, or is it a bug caused by the\nfact that reloptions.c fails to distinguish matviews from heaps at all?\n\nIf it is sensible, then I broke it in e3e66d8a9 ...\n\n\t\t\tregard... |
[
{
"msg_contents": "pgbench: Allow the transaction log file prefix to be changed.\n\nMasahiko Sawada, reviewed by Fabien Coelho and Beena Emerson, with\nsome a bit of wordsmithing and cosmetic adjustment by me.\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/pg/commitdiff/41124a91e61fc6d... |
[
{
"msg_contents": "I am looking for ways to ensure referential integrity on large objects. Something like having a column myoid in a table that holds an oid of a large object, and which throws an error when the referenced large object should be unlinked. Like \"myoid references pg_largeobject(loid)\", which doe... |
[
{
"msg_contents": "> On Tue, Nov 8, 2016 at 6:54 AM, Jeevan Chalke\r\n> <jeevan.chalke@enterprisedb.com> wrote:\r\n> > 1. ps_numTuples is declared as long, however offset and count members in\r\n> > LimitState struct and bound member in SortState struct is int64. However\r\n> > long on 32 bit machine may be 32... |
[
{
"msg_contents": "On Wed, Nov 9, 2016 at 11:32 PM, Kuntal Ghosh\n<kuntalghosh.2007@gmail.com> wrote:\n> On Fri, Nov 4, 2016 at 1:32 PM, Michael Paquier\n> <michael.paquier@gmail.com> wrote:\n>> Thank you for the new patch. This will be hopefully the last round of\n>> reviews, we are getting close to something ... |
[
{
"msg_contents": "Some work has been going on recently to be able to update our\ndocumentation build tool chain. After discussion on pgsql-docs, the\npeople involved agree that it is time to move forward.\n\nWe are now proposing that we change the way the HTML documentation is\nbuilt from jade/openjade+docboo... |
[
{
"msg_contents": "configuration parameters *_flush_after were added in 9.6. They are not in\npostgresql.conf.sample, but are also not marked GUC_NOT_IN_SAMPLE. Is this\nintentional and/or desirable?\n\nCheers,\n\nJeff\n\nconfiguration parameters *_flush_after were added in 9.6. They are not in postgresql.co... |
[
{
"msg_contents": "I noticed that buildfarm member piculet fell over this afternoon:\nhttp://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=piculet&dt=2016-11-10%2020%3A10%3A02\nwith this interesting failure during startup of the \"collate\" test:\npsql: FATAL: cache lookup failed for relation 1255\n\n1255 is... |
[
{
"msg_contents": "I am a newbie to databases and Postgres and I am trying to analyze the shared\nmemory being calculated and allocated by Postgres in the method\n\"CreateSharedMemoryAndSemaphores\" for different major versions for different\npostgres.conf file\n\nMy idea was to create a utility in Postgres and... |
[
{
"msg_contents": "\nHi all:\n\n\n As we know, the synchronous commit process is blocked while \nreceives from acknowledgement from standby in\n\nPostgreSQL.This is good for data consistence in master and standby, and \napplication can get important data from standby.But\n\nwhen the standby crash or network... |
[
{
"msg_contents": "Per prior discussion, this small Makefile change allows anything that\nuses $(prove_check) or $(prove_installcheck) to have the list of tests\noverridden by passing a PROVE_TESTS as a list of space-separated test\nfile paths.\n\nParticularly handy for src/test/recovery when you want to run\nt... |
[
{
"msg_contents": "Hi,\n\nI'm running some tests on a cluster with 4kB blocks, and it seems \nthere's a bug in xlogreader.c, causing FATAL errors for example when \nrunning pg_xlogdump:\n\npg_xlogdump: FATAL: error in WAL record at 48/63970258: \nBKPIMAGE_HAS_HOLE not set, but hole offset 0 length 4096 at 48/6... |
[
{
"msg_contents": "While pursuing the unprotected-use-of-CatalogSnapshot problem I noted\nearlier, I came across a couple of related issues:\n\n* ProcArrayInstallImportedXmin and ProcArrayInstallRestoredXmin will\nforcibly overwrite MyPgXact->xmin with some random xmin or other.\nWhile they take pains to ensure... |
[
{
"msg_contents": "Scenario:\n\nLarge table, *mostly* insert+select only, either a sequence or a timestamp\ncolumn set to current time. This is an ideal usecase for a BRIN index.\nAwesome.\n\nBut then consider the same table. Except rows are typically updated once or\ntwice when they are new, and *then* go read... |
[
{
"msg_contents": "Hi,\n\nI don't quite know what the hook in $subject was originally designed\nfor, but right now it has the problem that it's not invoked for EXPLAIN\nEXECUTE. That's because ExplainExecuteQuery directly calls\nExplainOnePlan instead of ExplainOneQuery_hook.\n\nUnfortunately that's not entire... |
[
{
"msg_contents": "Every test I write with the TAP framework for recovery seems to need\nto wait for one node to catch up to another or examine replication\nslot state. So: attached is a patch to add helper methods for these\ntasks.\n\nAlso add a new pg_lsn.pm helper class to deal with PostgreSQL's\nawkward rep... |
[
{
"msg_contents": "Anastasia, et al,\n\nThis is a review of including_columns_9.7_v5.patch.\n\nI looked through the commit fest list and this patch was interesting and I wanted to try it.\n\nI have used include columns on Microsoft SQL Server; DB2 for Linux, Unix, Windows; and DB2 for z/OS.\n\nAfter reading the... |
[
{
"msg_contents": "Hi,\nMaking developing ECPG application more efficiently and improve portability,\nI'd like to suggest DECLARE STATEMENT support in ECPG.\n\nThis DECLARE STATEMENT is one of the statement that lets users declare an identifier pointing a connection.\nThis identifier will be used in other embed... |
[
{
"msg_contents": "Hi, \n\nI would like to add the following support for a trigger. \nThis implementation enables to create a trigger efficiently \nin single command. \n\nIt had been discussed before. The link is as shown below.\nhttps://www.postgresql.org/message-id/CAA-aLv4m%3Df9cc1zcUzM49pE8%2B2NpytUDraTgfBm... |
[
{
"msg_contents": "Is there a reason why we don't log which relation triggered the snapshot\ntoo old error when it happens?\n\nSince we do have Relation available in TestForOldSnapshot_impl(), shouldn't\nwe be able to include it?\n\nHaving that in the log message would be very helpful, but is there a reason\nwh... |
[
{
"msg_contents": "\nI know we've discussed this before, but I have just had the unpleasant \nexperience of trying to get around the difficulty of inserting into a \nforeign table with a serial field, surely a common enough scenario that \nwe should try to deal with it better. The solution of using a local \nse... |
[
{
"msg_contents": "Tobias Bussmann has discovered an oddity with prepared statements.\r\n\r\nParallel scan is used with prepared statements, but only if they have\r\nbeen created with protocol V3 \"Parse\".\r\nIf a prepared statement has been prepared with the SQL statement PREPARE,\r\nit will never use a paral... |
[
{
"msg_contents": "Attached is an implantation of jsonb_delete that instead of taking a single\nkey to remove accepts an array of keys (it still does just keys, so it's\nusing the - operator, it's not like the path delete function that also\ntakes an array, but uses a different operator).\n\nIn some simple test... |
[
{
"msg_contents": "Account for catalog snapshot in PGXACT->xmin updates.\n\nThe CatalogSnapshot was not plugged into SnapshotResetXmin()'s accounting\nfor whether MyPgXact->xmin could be cleared or advanced. In normal\ntransactions this was masked by the fact that the transaction snapshot\nwould be older, but ... |
[
{
"msg_contents": "Hello,\n\nI have a question as subject line.\n\nhttps://www.postgresql.org/docs/devel/static/catalog-pg-proc.html\nAccording to the documentation, the purpose of pg_proc.probin is\nintroduced as follows:\n| Additional information about how to invoke the function. Again, the\ninterpretation is... |
[
{
"msg_contents": "I found it a real pain to set up Perl 5.8.8 to re-check recent TAP\nchanges against it. It turns out there's a much easier way than\nmessing with VMs or manually doing a source install, so I documented\nit to save others the future pain.\n\nIt'd be nice to commit this at least to 9.6 and v10.... |
[
{
"msg_contents": "Hi\n\nI have a repeated problem with regress tests\n\nmaster, Fedora 25,\n\nrunning on port 50848 with PID 5548\n============== creating database \"regression\" ==============\nCREATE DATABASE\nALTER DATABASE\n============== running regression test queries ==============\ntest ... |
[
{
"msg_contents": "Build HTML documentation using XSLT stylesheets by default\n\nThe old DSSSL build is still available for a while using the make target\n\"oldhtml\".\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/pg/commitdiff/e36ddab11735052841b4eff96642187ec9a8a7bc\n\nModified File... |
[
{
"msg_contents": "I looked into the problem reported at\nhttps://www.postgresql.org/message-id/b3690957-fd8c-6def-d3ec-e589887dd0f1%40codata.eu\n\nIt's easy to reproduce. Given this simple database:\n\ncreate table tt (f1 int primary key, f2 text);\ncreate view vv as select * from tt group by f1;\n\npg_dump w... |
[
{
"msg_contents": "Here is a tip for building the documentation faster during development.\nWith the new toolchain, you can build only a part of the documentation,\nlike this:\n\nmake html XSLTPROCFLAGS='--stringparam rootid pageinspect'\n\nwhere \"pageinspect\" is some XML id (see the top of pageinspect.sgml i... |
[
{
"msg_contents": "Following UNION of two queries with constant literals runs successfully.\n\nCASE 1:\npostgres=# SELECT 'abc' UNION SELECT 'bcd' ;\n ?column?\n----------\n abc\n bcd\n(2 rows)\n\nwhereas when these literals are part of a view, the UNION fails.\n\nCASE 2:\npostgres=# create view v as select 'ab... |
[
{
"msg_contents": "Hello hackers,\n\nI attached the test extension. It just creates event trigger on \nddl_command_end. And pg_catalog.pg_event_trigger_ddl_commands() is \nexecuted from extension's C-function handler.\n\nThe pg_event_trigger_ddl_commands() returns the error:\n\nERROR: unrecognized object class... |
[
{
"msg_contents": "I love seeing references to email threads in commit messages. It would \nmake them a lot friendlier though if a full http URL were included \ninstead of just a Message-ID, i.e. instead of <foo1234@bar.com> put \n<https://www.postgresql.org/message-id/foo1234@bar.com>. I know this is \na bit m... |
[
{
"msg_contents": "Hi all\n\nWhile adding support for logical decoding on standby I noticed that\nthe walsender doesn't respect\nSIGUSR1 with PROCSIG_RECOVERY_CONFLICT_DATABASE set.\n\nIt blindly assumes that it means there's new WAL:\n\nWalSndSignals(void)\n{\n...\n pqsignal(SIGUSR1, WalSndXLogSendHandl... |
[
{
"msg_contents": "Hello.\n\nWe had too-early WAL recycling during a test we had on a sync\nreplication set. This is not a bug and a bit extreme case but is\ncontrary to expectation on synchronous replication.\n\n> FATAL: could not receive data from WAL stream: ERROR: requested WAL segment 0000000100000000000... |
[
{
"msg_contents": "Hi\n\nIn one application I see slow queries. There is often used views like\n\nCREATE VIEW v AS SELECT min(a) M1, min(b) M2, max(c) M3, x, y, z\n FROM t1 GROUP BY x, y, z;\n\nand queries like\n\nSELECT * FROM v\n WHERE M2 = const1\n AND M3 > const2\n\nIsn't possible in this case pus... |
[
{
"msg_contents": "Hi,\nIn create_merge_append_path() we have following code\n1331\n1332 /* Now we can compute total costs of the MergeAppend */\n1333 cost_merge_append(&pathnode->path, root,\n1334 pathkeys, list_length(subpaths),\n1335 input_startup_cost, inp... |
[
{
"msg_contents": "Hi Experts,\n\nAs in the example below, i think the plan which hash table is created on\ntesttbl2 (the fewer tuples) should be choosen.\nBecause creating of hash table should faster in testtbl2. But it did not.\n\nI have tried to change the ordering of table by tuning parameter even if\nusing... |
[
{
"msg_contents": "Hi,\n\nsqlsmith just made a GUC that tricks the serialization code into\ndereferencing a nullpointer. Here's a recipe:\n\n--8<---------------cut here---------------start------------->8---\nset min_parallel_relation_size to 0;\nset max_parallel_workers_per_gather to 2;\nset force_parallel_mod... |
[
{
"msg_contents": "All,\n\nWhile reviewing and considering the patch to add the --no-blobs option\nto pg_dump, which I entirely agree with, I came across our claim that\npassing -b to pg_dump would cause blobs to be output, regardless of\nother options set.\n\nTo wit, our docs specifically say:\n\n----\nInclude... |
[
{
"msg_contents": "You get an odd result if you write \"something.*\" in an UPDATE's\nsource expressions:\n\nregression=# create table t1 (f1 int, f2 int, f3 int);\nCREATE TABLE\nregression=# update t1 set f1 = t1.*;\nERROR: UPDATE target count mismatch --- internal error\n\nThe reason for that is that we use ... |
[
{
"msg_contents": "Hi,\n\nthe following query appears to reliably crash parallel workers on master\nas of 0832f2d.\n\n--8<---------------cut here---------------start------------->8---\nset max_parallel_workers_per_gather to 2;\nset force_parallel_mode to 1;\n\nselect subq.context from pg_settings,\n lateral ... |
[
{
"msg_contents": "Hi,\n\nWhen tab-completing after ALTER DEFAULT PRIVILEGES ... GRANT|REVOKE,\ncurrently psql injects completion from the GRANT|REVOKE order, rather\nthan the one expected.\n\nA patch is attached. It adds the right completion to GRANT|REVOKE after\nALTER DEFAULT PRIVILEGES and after FOR ROLE|US... |
[
{
"msg_contents": "Hi,\n\nthe query below triggers a parallel worker assertion for me when run on\nthe regression database of master as of 0832f2d. The plan sports a\ncouple of InitPlan nodes below Gather.\n\nregards,\nAndreas\n\n Gather (cost=1.64..84.29 rows=128 width=4)\n Workers Planned: 1\n Single Co... |
[
{
"msg_contents": "I ran into this today:\n\nselect current_database() as current_db \\gset\nCREATE EXTENSION postgres_fdw;\nCREATE EXTENSION\nCREATE EXTENSION dblink;\nCREATE EXTENSION\nCREATE ROLE bob LOGIN PASSWORD 'bob';\nCREATE ROLE\nCREATE SERVER bob_srv FOREIGN DATA WRAPPER postgres_fdw OPTIONS ( host\n'... |
[
{
"msg_contents": "Hi all\n\nI've prepared a working initial, somewhat raw implementation for\nlogical decoding on physical standbys. Since it's a series of 20\nsmallish patches at the moment I won't attach it. You can find the\ncurrent version at time of writing here:\n\n https://github.com/postgres/postgre... |
[
{
"msg_contents": "Hello,\r\n\r\nThe attached patch is a revised version of pass-down LIMIT to FDW/CSP.\r\n\r\nBelow is the updates from the last version.\r\n\r\n'ps_numTuples' of PlanState was declared as uint64, instead of long\r\nto avoid problems on 32bits machine when a large LIMIT clause is\r\nsupplied.\r... |
[
{
"msg_contents": "Hi,\nI have this question. Looked for a help on http://dba.stackexchange.com/\nNo success.\nMaybe you can answer?..\nThank you in advance\n\n\n\"FOREIGN_TABLE\" created with postgres_fdw. LOCAL_TABLE is just a local\ntable...\n\nSymptoms:\n\n 1. I run in psql query SELECT * from FOREIGN_TAB... |
[
{
"msg_contents": "While fooling around trying to document the behavior of *-expansion,\nI got annoyed about the fact that it doesn't work in the context of\n\tUPDATE tab SET (c1,c2,c3) = (x,y,z) ...\nThe righthand side of this is a row-expression according to the SQL\nstandard, and \"foo.*\" would be expanded ... |
[
{
"msg_contents": "autovacuum: Drop orphan temp tables more quickly but with more caution.\n\nPreviously, we only dropped an orphan temp table when it became old\nenough to threaten wraparound; instead, doing it immediately. The\nonly value of waiting is that someone might be able to examine the\ncontents of t... |
[
{
"msg_contents": "Hi all,\n\nI noticed that we have no option to set default privileges for newly\ncreated schemas, other than calling GRANT explicitly. At work I use ALTER\nDEFAULT PRIVILEGE (ADP) command extensively, as the developers are\npermitted to manage DDL on the databases, and all work fine except fo... |
[
{
"msg_contents": "Doc: improve documentation about composite-value usage.\n\nCreate a section specifically for the syntactic rules around whole-row\nvariable usage, such as expansion of \"foo.*\". This was previously\ndocumented only haphazardly, with some critical info buried in\nunexpected places like xfunc... |
[
{
"msg_contents": "PostgreSQL tables and indexes not infrequently become bloated, and\nthis bloat is difficult to reverse once it has occurred. When a row\nis updated, a new row version is created and the space occupied by the\nold version can be eventually be recycled. Since there is always a\nshort delay an... |
[
{
"msg_contents": "\nI love MATERIALIZED VIEWs.\n\nBut we needed a method for recording deltas from REFRESHes, and that's\nnot supported. So I coded up my own version of materialized views, in\nPlPgSQL, that does provide a history feature.\n\nBesides a history feature, this includes the ability to record chang... |
[
{
"msg_contents": "Hi all\n\nToday I ran into an issue where commit timestamp lookups were failing with\n\n ERROR: cannot retrieve commit timestamp for transaction 2\n\nwhich is of course FrozenTransactionId.\n\nTransactionIdGetCommitTsData(...) ERRORs on !TransactionIdIsNormal(),\nwhich I think is wrong... |
[
{
"msg_contents": "I've found a typo in a comment in function do_analyze_rel() in\nfile analyze.c [line 348, master branch].\n\nAttached a patch.\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_contents": "In short:\n\nThe available timecounters on Google Compute Instances seem to be random.\n\nThe setting in the official FreeBSD image is wrong (not available on any \nof my test instances). FreeBSD will pick up a timecounter at random.\n\nWhen either the TSC or the TSC-low counter is used, expl... |
[
{
"msg_contents": "Hi All,\n\nI would like to take over pg_background patch and repost for\ndiscussion and review.\n\nInitially Robert Haas has share this for parallelism demonstration[1]\nand abandoned later with\nsummary of open issue[2] with this pg_background patch need to be\nfixed, most of them seems to b... |
[
{
"msg_contents": "I'm working on an extension that is meant to enable metaprogramming \nwithin Postgres, namely around the creation of objects. For example, a \n\"lookup table\" class might create a table, grant specific permissions on \nthat table, and create several functions to support getting data from \nt... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.