threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "In going back through old emails to see what issues might have been\nraised but not yet addressed for the SSI patch, I found the subject\nissue described in a review by Jeff Davis here:\n \nhttp://archives.postgresql.org/pgsql-hackers/2010-10/msg01159.php\n \nI think this is already handled b... |
[
{
"msg_contents": "There's an issue where we don't seem to have consensus yet, so I\nfigured I'd bounce it off the list.\n \nIf the SSI patch were to be accepted as is, REPEATABLE READ would\ncontinue to provide the exact same snapshot isolation behavior which\nboth it and SERIALIZABLE do through 9.0, and SERIA... |
[
{
"msg_contents": "Patching:\n\npatching file doc/src/sgml/func.sgml\nHunk #6 succeeded at 10567 (offset 1 line).\nHunk #7 succeeded at 10621 (offset 1 line).\nHunk #8 succeeded at 10721 (offset 1 line).\nHunk #9 succeeded at 10775 (offset 1 line).\npatching file src/backend/nodes/makefuncs.c\npatching file src... |
[
{
"msg_contents": "hstore's \"hstore ?& text[]\" operator is defined as \"contains all\", ie,\nit will return true if all the key names found in the text array are\npresent in the hstore.\n\nISTM that a sane definition of this operator would provide that if the\narray is empty, it returns true: every set contai... |
[
{
"msg_contents": "This begins the patch series for the design I recently proposed[1] for avoiding\nsome table rewrites in ALTER TABLE ... ALTER COLUMN ... TYPE. I'm posting these\npatches today:\n\n0 - new test cases\n1 - recheck UNIQUE constraints on ALTER TYPE\n2 - skip cases where we can already prove ther... |
[
{
"msg_contents": "When ALTER TABLE rewrites a table, it reindexes, but the reindex does not\nrevalidate UNIQUE/EXCLUDE constraints. This behaves badly in cases like this,\nneglecting to throw an error on the new UNIQUE violation:\n\nCREATE TABLE t (c numeric UNIQUE);\nINSERT INTO t VALUES (1.1),(1.2);\nALTER ... |
[
{
"msg_contents": "This patch removes ALTER TYPE rewrites in cases we can already prove valid. I\nadd a function GetCoerceExemptions() that walks an Expr according to rules\ndiscussed in the design thread, simplified slightly pending additions in the\nnext patch. See the comment at that function for a refresh... |
[
{
"msg_contents": "Here I add the notion of an \"exemptor function\", a property of a cast that\ndetermines when calls to the cast would be superfluous. Such calls can be\nremoved, reduced to RelabelType expressions, or annotated (via a new field in\nFuncExpr) with the applicable exemptions. I modify various ... |
[
{
"msg_contents": "Add exemptor functions to avoid rewrites for conversions involving the temporal\ndata types. I needed a find-last-set function for the interval_scale exemptor\nfunction, so I imported one from FreeBSD. To improve timestamp<->timestamptz\nwhen the timezone \"is\" UTC/GMT, I compare the curre... |
[
{
"msg_contents": "Add exemptor functions for bit and varbit. These are probably the simplest\nexamples of the full range of optimizations. I would have used them as the test\ncase in the initial exemptor function patch if it were a more mainstream use\ncase.",
"msg_date": "Sun, 9 Jan 2011 17:06:15 -0500"... |
[
{
"msg_contents": "Add an exemptor function for numeric. We store the scale in every datum, making\nnumeric(7,2)->numeric(8,3) unoptimizable. Precision changes work, though.",
"msg_date": "Sun, 9 Jan 2011 17:07:09 -0500",
"msg_from": "Noah Misch <noah@leadboat.com>",
"msg_from_op": true,
"msg_... |
[
{
"msg_contents": "Whilst fooling around with GIN over the past few days, I noticed the\nfollowing rather surprising behavior:\n\nregression=# create table t1 (f1 int[]);\nCREATE TABLE\nregression=# insert into t1 values (array[42]);\nINSERT 0 1\nregression=# create index ti1 on t1 using gin (f1);\nCREATE INDEX... |
[
{
"msg_contents": "Due to popular request (Hey, David's popular, right?), I'm posting a\npatch for Serializable Snapshot Isolation (SSI), although I don't\nyet have everything in it that I was planning on submitting before\nthe CF. I will probably be submitting another version before the\ndeadline with the fol... |
[
{
"msg_contents": "Attached is a rebased roll-up of the 3 and 3a patches from last month.\n \n-Kevin",
"msg_date": "Mon, 10 Jan 2011 10:27:50 -0600",
"msg_from": "\"Kevin Grittner\" <Kevin.Grittner@wicourts.gov>",
"msg_from_op": true,
"msg_subject": "READ ONLY fixes"
},
{
"msg_contents":... |
[
{
"msg_contents": "To All,\r\nI am attempting to setup a server to use SSPI for mapping operating system users/groups to various postgres roles. In process I found that everything is driven off of the username in the mapping with no group but the mapping file supports regular expressions to do some mapping.\r\n... |
[
{
"msg_contents": "The attached, applied patch removes toast relid from the relation array\nas it is no longer needed. Also other remaming was done.\n\n-- \n Bruce Momjian <bruce@momjian.us> http://momjian.us\n EnterpriseDB http://enterprisedb.com\n\n + It's impossible fo... |
[
{
"msg_contents": "Hi. I try to compile postgres with mingw32. When configure runs, it\ntells that found perl 5.6 which is too old. I install perl 5.10 from\nactivestate but configure cant find it. How to set up path to newer\nperl?\n\n-- \nSent from my mobile device\n\n------------\npasman\n",
"msg_date": ... |
[
{
"msg_contents": "Attached is an updated version of Heikki's patch to use a parser for\nthe walsender commands, instead of parsing things manually. It also\ndoes some minor refactoring in walsender.c to break out\nIdentifySystem() and StartReplication() to their own functions to make\nit more readable.\n\nWhil... |
[
{
"msg_contents": "Set process title to indicate base backup is running\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=2e36343f82377fbb50834bba6557f8f243fecb34\n\nModified Files\n--------------\nsrc/backend/replication/basebackup.c | 10 ++++++++... |
[
{
"msg_contents": "Here's a patch that adds a few PL/Python functions for quoting strings.\nIt's an incremental patch on top of the plpython-refactor patch sent in\nhttp://archives.postgresql.org/message-id/4D135170.3080705@wulczer.org.\n\nGit branch for this patch:\nhttps://github.com/wulczer/postgres/tree/fun... |
[
{
"msg_contents": "When a function is created, the system validates the syntax and\ncomplains if any function the created function attempts to call is\nmissing.\nI think this is really good, since it traps typos and warns you if you\nhave forgotten to install any functions your function depends on.\n\nIt would ... |
[
{
"msg_contents": "I can't seem to read a UDT properly from a stored function with the\npostgres JDBC driver. This is some sample code:\n\n====================================\nCREATE TYPE u_country AS ENUM ('Brazil', 'England', 'Germany')\n\nCREATE TYPE u_street_type AS (\n street VARCHAR(100),\n no VARCHAR(... |
[
{
"msg_contents": "Josh Berkus wrote:\n \n>> Really, the biggest risk of such a GUC is the confusion factor\n>> when supporting people.\n \n> How is this different from our other backwards-compatibility GUCs?\n \nI thought Tom might be concerned about such a GUC destabilizing\nthings in other ways. I just wan... |
[
{
"msg_contents": "I wrote:\n \n> (3) Recent changes to index page split/combine logic are still\n> being tested.\n \nAnd we have a new patch now that this has been finished. Two new\nlines to clean up linkages which were missed, and there was one\nover-zealous Assert elsewhere which was tickled by these change... |
[
{
"msg_contents": "I wrote:\n \n> Last time we did stress tests, it uncovered some race conditions.\n> Those were fixed at the time, and hopefully we haven't introduced\n> any new ones; but it's a Very Good Thing that Dan is able to run\n> some more DBT-2 tests, even if that test isn't ideal for\n> highlighting... |
[
{
"msg_contents": "Tried asking this in pgsql-general but I got no response, so I thought\nI'd give hackers a shot:\n\npostgres=# select (((1.7976931348623157081e+308)::double\nprecision)::text)::double precision;\nERROR: \"1.79769313486232e+308\" is out of range for type double precision\n\nI'm working on a p... |
[
{
"msg_contents": "Ok, I have made some progress. This is still a proof-of-concept patch,\nbut the important pieces are working together.\n\nSynopsis:\n\n CREATE TYPE numrange AS RANGE (SUBTYPE=numeric, \n SUBTYPE_CMP=numeric_cmp);\n\n SELECT range_eq('[1,2.2)'::numrange,'[1,2.2]');\n SELECT range_lbound(... |
[
{
"msg_contents": "A reminder about PGCon 2011; the deadline is 19 January 2011.\n\nPGCon 2011 will be held 19-20 May 2011, in Ottawa at the University of\nOttawa. It will be preceded by two days of tutorials on 17-18 May 2011.\n\nWe are now accepting proposals for talks. Proposals can be quite \nsimple. We d... |
[
{
"msg_contents": "Has anyone written a in-depth description on how to traverse the pg_depend tree?\nThe 'a' and 'i' deptype really makes it hard to figure out the\ndependency order, a topological sort does not work.\n\nMy latest attempt involved trying to group by all objects connected to\neach other via depty... |
[
{
"msg_contents": "Now that we have a basic over-the-wire base backup capability in \nwalsender, it would be nice to allow taking multiple base backups at the \nsame time. It might not seem very useful at first, but it makes it \neasier to set up standbys for small databases. At the moment, if you \nwant to set... |
[
{
"msg_contents": "Hello,\n\nHere's the patch that improves handling of arrays as pl/perl function input\narguments, converting postgres arrays of arbitrary dimensions into perl array\nreferences. It includes regression tests and a documentation changes, and it\nbuilds and runs successfully on my mac os x and l... |
[
{
"msg_contents": "Is it a good idea that we run make check with MULTIBYTE = SQL_ASCII by\ndefault? We run it with the user's locale by default, so shouldn't we\nuse the encoding that belongs to the locale by default? Otherwise we\nare testing a fairly unrepresentative environment. If you really want\nto tes... |
[
{
"msg_contents": "We have 46 patches in this CommitFest so far and I know that there are\nquite a few patches that have been posted but not added to the\nCommitFest application yet (please fix this, if you are a patch author\nwho has failed to do this) and that there will be lots more patches\nposted over the ... |
[
{
"msg_contents": "Attached are WIP version of patches for FDW API. Basically not\nchanged from last version but rebased to current HEAD.\n\nTo make review easier, I split core functionality into 3 patches. \nPlease apply these patches in the following order.\n\n1) fdw_handler - this patch adds HANDLER option ... |
[
{
"msg_contents": "Greetings,\n\nMinor enhancement, but a valuable one imv. Hopefully there aren't any\nissues with it. :)\n\n\tThanks!\n\n\t\tStephen\n\ncommit 3cb707aa9f228e629e7127625a76a223751a778b\nAuthor: Stephen Frost <sfrost@snowman.net>\nDate: Wed Jan 12 09:17:31 2011 -0500\n\n Add support for lo... |
[
{
"msg_contents": "On Wed, Sep 15, 2010 at 11:14 PM, Heikki Linnakangas\n<heikki.linnakangas@enterprisedb.com> wrote:\n> On 15/09/10 16:55, Tom Lane wrote:\n>>\n>> So I'm wondering if we couldn't eliminate the five-second sleep\n>> requirement here too. It's problematic anyhow, since somebody looking\n>> for e... |
[
{
"msg_contents": "I found a crash case (assertion failure) when runing psql -f\nutf8_encoded_script.sql against client_encoding = shift_jis in\npostgresql.conf. Though encoding mismatch is obviously user's fault, a\ncrash doesn't explain anything to him.\n\nThe thing is, prepare_buffer() in psqlscan.l assumes ... |
[
{
"msg_contents": "The attached patch is a collection of libpq documentation cleanups\nrecommended in a list of changes emailed to me by Leslie S Satenstein.\n\nLeslie found a number of places our libpq documentation that were unclear\nor awkward, and this diff generated by me attempts to address them.\n\nI hav... |
[
{
"msg_contents": "I've been looking into Gordon Shannon's crash report here:\nhttp://archives.postgresql.org/pgsql-general/2010-12/msg01030.php\nAfter some groveling around in the core dump (thanks to Gordon for\nmaking that available), I figured out the cause of the problem.\n\nThe missing piece of informatio... |
[
{
"msg_contents": "Hello,\n\n Today I was reading a blog post from a fellow coworker\nhttp://www.depesz.com/index.php/2010/12/11/waiting-for-9-1-knngist/ and\nstarted to mess around with the trigram contrib package for postgres and\nplaying with some different word dictionaries for English and German. I wa... |
[
{
"msg_contents": "Hi,\n\nIn the case where there are multiple standbys, when a failover\nhappens, we usually calculate most advanced standby by using\npg_last_xlog_receive_location or pg_last_xlog_replay_location,\nand promote it to new master. The problem is that neither\nfunction might return the right last ... |
[
{
"msg_contents": "Hi,\n\nWhen the master shuts down or crashes, there seems to be\nthe case where walreceiver exits without flushing WAL which\nhas already been written. This might lead startup process to\nreplay un-flushed WAL and break a Write-Ahead-Logging rule.\n\nwalreceiver.c\n>\t\t/* Wait a while for da... |
[
{
"msg_contents": "The example from Tom Lane below results in a database which is not\npossible to correctly dump using pg_dump.\n\nThe view v1 strangely becomes a table in the dump output?!\n\nIt's probably a quite useless database to dump in the first place, but\nthat is no excuse to generate an invalid dump,... |
[
{
"msg_contents": "Are multiple identical entires in pg_depend possible?\nIf so, how do they occur, and what is the purpose of representing\nexactly the same dependency two times in pg_depend?\n\nI expected the following query not to return any rows, but it did:\n\nglue=# select count(*), * from pg_depend group... |
[
{
"msg_contents": "While testing Fujii-san's patch to flush any already-written WAL on \nerror in walreceiver, I added a debugging elog to XLogWalRcvFlush() to \nprint out how far it has written and flushed.\n\nI saw an unexpected pattern while the standby catches up with the master:\n\nLOG: streaming replicat... |
[
{
"msg_contents": "Hello, Pgsql-hackers.\n\nI'm getting such warnings:\n\npg_dump.c: In function 'dumpSequence':\npg_dump.c:11449:2: warning: unknown conversion type character 'l' in format\npg_dump.c:11449:2: warning: too many arguments for format\npg_dump.c:11450:2: warning: unknown conversion type character ... |
[
{
"msg_contents": "Folks,\n\nI've noticed over the years that we give people dire warnings never to\nsend a KILL signal to the postmaster, but I'm unsure as to what are\npotential consequences of this, as in just exactly how this can result\nin problems. Is there some reference I can look to for explanations\n... |
[
{
"msg_contents": "Hi,\n\nPlease find an updated (merged) version 24 of the extension patch set,\nsimplified to follow on Tom's comment about dependency tracking when\ndoing ALTER EXTENSION SET SCHEMA. We now follow only one level of\ndependencies, and that just works (thanks Tom).\n\nOn the same idea, I also ... |
[
{
"msg_contents": "Attached is a new version of the SSI patch with these changes:\n \n- Performance is markedly better on table scans. Fixed by this\ncommit:\n \nhttp://git.postgresql.org/gitweb?p=users/kgrittn/postgres.git;a=commitdiff;h=64ca508a0e2fa9c21dc76a5d6a5f549c27f511fa\n \n- A race condition which wa... |
[
{
"msg_contents": "Hello,\n\nAs previously commented, here's a proposal with patch to turn foreign\nkey checks into something less intrusive.\n\nThe basic idea, as proposed by Simon Riggs, was discussed in a previous\npgsql-hackers thread here:\nhttp://archives.postgresql.org/message-id/AANLkTimo9XVcEzfiBR-ut3K... |
[
{
"msg_contents": "On Thu, Jan 6, 2011 at 11:37 PM, Greg Smith <greg@2ndquadrant.com> wrote:\n> If it defaulted to 3% of shared_buffers, min 64K & max 16MB for the auto\n> setting, it would for the most part become an autotuned parameter. That\n> would make it 0.75 to 1MB at the standard anemic Linux default k... |
[
{
"msg_contents": "When the earlier patches in this series allow ALTER TABLE ... ALTER COLUMN\n... TYPE to skip the table rewrite, it still rebuilds indexes and rechecks\nconstraints depending on that column. With this patch, we'll skip an index\nrebuild when the change does not cross operator families and the... |
[
{
"msg_contents": "https://github.com/gluefinance/pov/blob/master/doc/example_database.sql\n\nPlease feel free to put any of this on the PostgreSQL wiki like\nsuggested by David Fetter.\n\nThis is an example of some functionality provided and used by pov\n(PostgreSQL Object version control system).\nMost of, if... |
[
{
"msg_contents": "I whipped up the attached patch tonight. It's pretty quick and dirty,\nso it's possible I've missed something, but the intent is to suppress\nwriting of hint bits by buffers allocating backends, and by\ncheckpoints, and write them only from the background writer cleaning\nscan. It therefore... |
[
{
"msg_contents": "Functions to control recovery, to aid PITR and Hot Standby.\npg_is_xlog_replay_paused()\npg_xlog_replay_pause()\npg_xlog_replay_resume()\n\nrecovery.conf parameter: pause_at_recovery_target (bool)\n\n-- \n Simon Riggs http://www.2ndQuadrant.com/books/\n PostgreSQL Development, 24x7 ... |
[
{
"msg_contents": "Exit from base backups when shutdown is requested\n\nWhen the exit waits until the whole backup completes, it may take\na very long time.\n\nIn passing, add back an error check in the main loop so we detect\nclients that disconnect much earlier if the backup is large.\n\nBranch\n------\nmaste... |
[
{
"msg_contents": "Hi all!\n\n\nIn preperation of some further improvments to our infrastructure, the \nsysadmin team needs to move coridian.postgresql.org (aka \ncommitfest.postgresql.org) and mintaka.postgresql.org \n(media.postgresql.org) to a different host within the same datacenter.\n\nWe are planning to ... |
[
{
"msg_contents": "Here's the patch to add ENCODING option to COPY command.\nThe fundamental issue was explained months ago:\n\nhttp://archives.postgresql.org/message-id/AANLkTikCt6bHXZjO_oX+JS7+G=jAQ7gVZPu0Owjcsbfb@mail.gmail.com\n\nIn short, client_encoding is not appropriate for copy operation so we\nshould ... |
[
{
"msg_contents": "Hi,\n\nHere is a patch that implements \"named restore points\".\n\nIt allows DBAs to specify an exact point to which they can recover\nbut that point will have a name, so they have a better control of when\nthey want to stop recovery (ie: DBA's won't depend of remember\nspecific times, date... |
[
{
"msg_contents": "Dear pghackers,\n\nDuring the last two weeks, I have enjoyed a total of 113 hours of\ndevelopment of pov - PostgreSQL Object Version control system.\n\nVersion 1.0 is now finally released!\n\nPOV will not touch any of your \"data\" (tables, sequences, indexes) -\nit will only keep track of yo... |
[
{
"msg_contents": "I've been going over this patch with a fine-tooth comb for the last two\nweeks, fixed some small bugs, added comments, made initdb a little\nfriendlier, but no substantial changes.\n\nI'm going to start working on SQL-level CREATE/DROP/ALTER COLLATION\nsupport and associated things now.",
... |
[
{
"msg_contents": "Hi,\n\nI propose the patch which reduces the amount of data loss\non the standby.\n\nIf you create the trigger file after replication is terminated,\nthe standby tries to replay all the WAL available in pg_xlog\nand the archive. This behavior is OK, and it's helpful to reduce\nthe amount of d... |
[
{
"msg_contents": "Dan found a few low-frequency assertions when he let the patch cook\nin DBT-2 for extended periods. For the most part they were\nover-zealous assertions, although he made a couple minor adjustments\nrelated to their causes in tracking the process local predicate lock\ninformation used to mak... |
[
{
"msg_contents": "On Sat, Jan 15, 2011 at 9:25 AM, Greg Smith <greg@2ndquadrant.com> wrote:\n> I'm going to share it before someone announces a deadline has passed or\n> something. (whistling)\n\nSpeaking of that, it is now Saturday, January 15, 2011 and CommitFest\n2011-01 is now in progress. If you have an... |
[
{
"msg_contents": "diff --git a/src/backend/replication/.gitignore\nb/src/backend/replication/.gitignore\nnew file mode 100644\nindex 0000000..a0332b2\n--- /dev/null\n+++ b/src/backend/replication/.gitignore\n@@ -0,0 +1,3 @@\n+/repl_gram.c\n+/repl_gram.h\n+/repl_scanner.c\n\n\n",
"msg_date": "Sat, 15 Jan 20... |
[
{
"msg_contents": "\nI've been waiting for the latest FDW patches as patiently as I can, and \nI've been reviewing them this morning, in particular the file_fdw patch \nand how it interacts with the newly exposed COPY API. Overall it seems \nto be a whole lot cleaner, and the wholesale duplication of the copy \... |
[
{
"msg_contents": "This patch creates pg_basebackup in bin/, being a client program for\nthe streaming base backup feature.\n\nI think it's more or less done now. I've again split it out of\npg_streamrecv, because it had very little shared code with that\n(basically just the PQconnectdb() wrapper).\n\nOne thing... |
[
{
"msg_contents": "I have modified test_fsync to use modular C so there is less duplicate\ncode and it can be enhanced easier. Applied patch attached, though the\ndiff is larger than the C file so you might want to just look at the C\nfile in git.\n\n-- \n Bruce Momjian <bruce@momjian.us> http://momji... |
[
{
"msg_contents": "Here's a cutdown version of the idea about including WAL in the base\nbackup. What I initially wanted was to introduce a way to guarantee\nthat the required WAL (with some sort of cutoff of course) would be\navailable for the backup, but I ran out of time for that. We can\nalways add that lat... |
[
{
"msg_contents": "Here's the latest patch for sync rep.\n\n>From here, I will be developing the patch further on public git\nrepository towards commit. My expectation is that commit is at least 2\nweeks away, though there are no major unresolved problems. I expect\nessential follow on patches to continue for a... |
[
{
"msg_contents": "Last I remember, we were going to add this as an option. But I don't\nsee a patch in the queue. Am I missing it? Was I supposed to write it?\n\n-- \n -- Josh Berkus\n PostgreSQL Experts Inc.\n ... |
[
{
"msg_contents": "Hi,\n\nwhat follows is a review of the FDW API patch from\nhttp://archives.postgresql.org/message-id/20110114212358.82C7.6989961C@metrosystems.co.jp\n\nAll three patches apply cleanly and compile without warnings. Regression\ntests pass.\n\nLet me go patch by patch, starting with the first on... |
[
{
"msg_contents": "In 9.0, we specifically require using \"replication\" as database name\nto start a replication session. In 9.1 we will have the REPLICATION\nattribute to a role - should we change it so that \"all\" in database\nincludes replication connections? It certainly goes in the \"principle\nof least ... |
[
{
"msg_contents": "Currently, replication connections *always* logs something like:\nLOG: replication connection authorized: user=mha host=[local]\n\nThere's no way to turn that off.\n\nI can't find the reasoning behind this - why is this one not\ncontrolled by log_connections like normal ones? There's a comme... |
[
{
"msg_contents": "pg_stat_replication shows all replication information to all users, no\nrequirement to be a superuser or anything. That leaks a bunch of\ninformation that regular pg_stat_activity doesn't - such as clients IP\naddresses. And also of course all the replication info itself, which\nmay or may no... |
[
{
"msg_contents": "Since we now show the application name in pg_stat_replication, I think\nit would make sense to have the walreceiver set\nfallback_application_name on the connection string, like so:\n\ndiff --git a/src/backend/replication/libpqwalreceiver/libpqwalreceiver.c\nb/src/backend/replication/libpqwal... |
[
{
"msg_contents": "Robert Haas wrote:\n \n> a quick-and-dirty attempt to limit the amount of I/O caused by hint\n> bits. I'm still very interested in knowing what people think about\n> that.\n \nI found the elimination of the response-time spike promising. I\ndon't think I've seen enough data yet to feel comf... |
[
{
"msg_contents": "This is a review for the patch sent as\nhttps://commitfest.postgresql.org/action/patch_view?id=456\n\n== Submission ==\nThe patch applied cleanly atop of plpython refactor patches. The\nformat is git diff (though refactor patches is format-patch). I did\npatch -p1.\nIt includes adequate amoun... |
[
{
"msg_contents": "Robert Haas wrote:\n \n> I think you may be confused about what the patch does - currently,\n> pages with hint bit changes are considered dirty, period.\n> Therefore, they are written whenever any other dirty page would be\n> written: by the background writer cleaning scan, at checkpoints,\n... |
[
{
"msg_contents": "Hi,\n\nThe v24 extension patch is already suffering of bitrot after recent\ncleaning of src/include/commands/defrem.h as seen here:\n\n http://git.postgresql.org/gitweb?p=postgresql.git;a=commit;h=6ca452ba7fca14dad16425a56ffa1c8a93496b5f\n\nPFA version 25 of the main patch, and v3 of the UPG... |
[
{
"msg_contents": "The attached, applied patch adds getopt() support to test_fsync.\n\n-- \n Bruce Momjian <bruce@momjian.us> http://momjian.us\n EnterpriseDB http://enterprisedb.com\n\n + It's impossible for everything to be true. +",
"msg_date": "Mon, 17 Jan 2011 09:... |
[
{
"msg_contents": "I used the patch from CommitFest application and applied the following \ncommit to fix a known issue:\n\nhttp://git.postgresql.org/gitweb?p=postgresql-extension.git;a=commitdiff;h=d4991d35283ae0ceeb7f9e4203cf6a9dfb5d128d\n\nIs the patch in context diff format?\nYes.\n\nDoes the patch apply cl... |
[
{
"msg_contents": "Is there a value to this test_fsync test?\n\n\tCompare open_sync with different sizes:\n\t(This is designed to compare the cost of one large\n\tsync'ed write and two smaller sync'ed writes.)\n\t open_sync 16k write 242.563 ops/sec\n\t 2 open_sync 8k writes ... |
[
{
"msg_contents": "Is there value in moving test_fsync to /contrib?\n\n-- \n Bruce Momjian <bruce@momjian.us> http://momjian.us\n EnterpriseDB http://enterprisedb.com\n\n + It's impossible for everything to be true. +\n",
"msg_date": "Mon, 17 Jan 2011 11:02:57 -0500 (E... |
[
{
"msg_contents": "Anssi K��ri�inen wrote:\n \n> I tried to break the version 11 of the patch (some of the work was\n> against earlier versions). In total I have used a full work day\n> just trying to break things, but haven't been able to find anything\n> after version 8. I can verify that the partial in... |
[
{
"msg_contents": "Jeff Janes wrote:\n \n> A review:\n \nThanks! Very thorough!\n \n> None of the issues I raise above are severe. Does that mean I\n> should change the status to \"ready for committer\"?\n \nI see that notion was endorsed by Robert, so I'll leave it alone for\nnow. If a committer asks me to ... |
[
{
"msg_contents": "Itagaki Takahiro wrote:\n> Shigeru HANADA wrote:\n \n>> Attached patch would avoid this leak by adding per-copy context to\n>> CopyState. This would be overkill, and ResetCopyFrom() might be\n>> reasonable though.\n>\n> Good catch. I merged your fix into the attached patch.\n \nReview for C... |
[
{
"msg_contents": "> Dan Ports wrote:\n \n> Yes, that comment was supposed to be attached to\n> possibleUnsafeConflicts.\n \n> Actually, I think that \"other\" hash no longer exists\n \n> The comment above SERIALIZABLEXACT also needs to be updated since\n> it refers to said hash table. And if I'm not mistaken ... |
[
{
"msg_contents": "> Heikki Linnakangas wrote:\n \n> Setting the high bit in OldSetXidAdd() seems a bit weird. How about\n> just using UINT64_MAX instead of 0 to mean no conflicts? Or 1, and\n> start the sequence counter from 2.\n \nSure. I think I like reserving 1 and starting at 2 better. Will do.\n \n> Re... |
[
{
"msg_contents": "I've gotten permission to move pg_filedump from its former home at\nsources.redhat.com to pgfoundry. You can find the historical release\ntarballs as well as current sources at\n\thttp://pgfoundry.org/projects/pgfiledump/\n\nOne advantage of doing this is it will be a lot easier to let other... |
[
{
"msg_contents": "> Dan Ports wrote:\n> On Mon, Jan 17, 2011 at 07:20:20PM -0600, Kevin Grittner wrote:\n>> OK. I may need to bounce some questions off the list to resolve\n>> some of them. The biggest, in my mind, is whether\n>> MySerializableXact needs to be declared volatile. I don't have my\n>> head aroun... |
[
{
"msg_contents": "I'm compiling postgresql 9.0.2 using msys + mingw + gcc 4.5.2 (latest \nofficial release from mingw). This is on Windows 7 64-bit.\n\nUnfortunately the built dlls, at least libpq.dll, crash if they need to \nbe relocated. This happens to me when loading libpq.dll into a project \nthat has a... |
[
{
"msg_contents": "Shouldn't the comment read \"If first time through\"?\n\n\t/*\n\t * If not first time through, get workspace to remember main XIDs in. We\n\t * malloc it permanently to avoid repeated palloc/pfree overhead.\n\t */\n\tif (xids == NULL)\n\t{\n...\n\t\txids = (TransactionId *) malloc(maxxids * s... |
[
{
"msg_contents": "\nThe following bug has been logged online:\n\nBug reference: 5842\nLogged by: Daniel Popowich\nEmail address: danielpopowich@gmail.com\nPostgreSQL version: 8.4.6\nOperating system: x86_64-pc-linux-gnu (Ubuntu 10.04.1)\nDescription: Memory leak in PL/Python when ta... |
[
{
"msg_contents": "I just noticed that if you execute the same DO command over and over\nwithin a session, it gets slower and slower. And if you keep it up\nyou'll notice the backend's RAM consumption bloating too. The cause\nappears to be that we leak the cached plans created for any SQL\nstatements or expre... |
[
{
"msg_contents": "Hey folks,\n\nPgEast is being held in NYC this year from 03/22-03-25. Get your papers\nin, the deadline is soon!\n\nhttp://www.postgresqlconference.org/\n\nJoshua D. Drake\n\n-- \nPostgreSQL.org Major Contributor\nCommand Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579\nConsulting, ... |
[
{
"msg_contents": "Log replication connections only when log_connections is on\n\nPreviously we'd always log replication connections, with no\nway to turn them off.\n\nBranch\n------\nmaster\n\nDetails\n-------\nhttp://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=6e1726d082b5a9c305ed44a69ff50bc3407... |
[
{
"msg_contents": "I have modified test_fsync to use test labels that match wal_sync_method\nvalues, and and added more tests for open_sync with different sizes. \nThis should make the program easier for novices to understand. Here is\na test run for Ubuntu 11.04:\n\n\t$ ./test_fsync\n\t2000 operations per tes... |
[
{
"msg_contents": "I just got annoyed by the fact that contrib/intarray has support for\nqueries on GIN indexes on integer[] columns, but they only work if you\nuse the intarray-provided opclass, not the core-provided GIN opclass for\ninteger[] columns. In general, of course, two different GIN opclasses\naren'... |
[
{
"msg_contents": "Hello\n\nThe EXECUTE statement doesn't support a parametrization via\nSPI_execute_with_args call and PQexecParams too. It can be a security\nissue. If somebody use a prepared statement as protection to sql\ninjection, then all security goes out, because he has to call EXECUTE\nwithout paramet... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.