FAQ

PostgreSQL > pgsql-hackers > May 1997

Search Discussions

158 discussions - 476 posts

  • I thought all of you should see this. </chris Those who cannot remember the past are doomed to buy Microsoft products. - ---------- Forwarded message ---------- Date: Thu, 08 May 1997 09:30:25 -0500 ...
    Christopher HicksChristopher Hicks
    May 8, 1997 at 5:16 pm
    May 29, 1997 at 2:19 am
  • Just curious ... Vadim P.S. Yet another 3 days holyday... ------------------------------
    Vadim B. MikheevVadim B. Mikheev
    May 8, 1997 at 4:45 pm
    May 27, 1997 at 3:28 am
  • OK, now that we have some time, here are the open issues I remember seeing. I am sure there are others: - --------------------------------------------------------------------------- Have we updated ...
    Bruce MomjianBruce Momjian
    May 19, 1997 at 4:53 pm
    May 27, 1997 at 1:31 am
  • I was bored: http://www.hcs.harvard.edu/~lynders/postgresql.html Lemme know what you think. sean Sean Lyndersay Time is natures way of making sure lynders@hcs.harvard.edu things don't happen all at ...
    Sean LyndersaySean Lyndersay
    May 29, 1997 at 7:59 am
    May 31, 1997 at 12:54 pm
  • I pop'd into a friends work the other night to help her with an Informix related problem, and found something that I don't *believe* we do now, but think it would be a nice extension to the Sequence ...
    The Hermit HackerThe Hermit Hacker
    May 9, 1997 at 3:37 am
    May 15, 1997 at 3:35 pm
  • Do we need to contact Openlink to get new drivers for 6.1 prepared, now that the libpq communiction to the backend has changed? - -- Bruce Momjian maillist@candle.pha.pa.us ...
    Bruce MomjianBruce Momjian
    May 23, 1997 at 2:21 pm
    May 25, 1997 at 12:15 pm
  • I am glad we postponed the release (thanks Marc), and I am glad that bug reports AND FIXES are contining to be reported and applied. One item I can't figure out how to fix is to give pg_dumpall the ...
    Bruce MomjianBruce Momjian
    May 20, 1997 at 2:52 pm
    May 20, 1997 at 11:51 pm
  • It's time I migrate to a newer version than 1.06. What version is likely to be the most stable? 6.0? Or one of the 6.1 snapshots? Another member of the League for Programming Freedom (LPF) ...
    Brian BartholomewBrian Bartholomew
    May 12, 1997 at 10:09 pm
    May 14, 1997 at 12:45 pm
  • The only regression diff's I am seeing now are: < |Mon May 10 23:59:12 1943 PST - --- Seems my system knows 1943 was using Pacific War Time: # Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S Rule US ...
    Bruce MomjianBruce Momjian
    May 29, 1997 at 6:36 pm
    May 30, 1997 at 2:32 pm
  • How hard would it be to add a sql knob that disabled or shorter- circuited various features of the back end? One setting would disable all code related to transactions. Another might disable some of ...
    Brian BartholomewBrian Bartholomew
    May 15, 1997 at 7:52 pm
    May 28, 1997 at 6:04 am
  • OK, to clarify: you are running on a machine which does _not_ set HAVE_INT_TIMEZONE but _does_ set USE_POSIX_TIME ? Those are cases which I can't test, but your case results should help me track it ...
    Thomas G. LockhartThomas G. Lockhart
    May 26, 1997 at 4:51 pm
    May 27, 1997 at 6:10 pm
  • Hi, there is a memory leak in the sort code of postgres (970428). If I do a query without sort the total memory remains the same, if I add a sort clause it grows of about 12K per query, as you can ...
    Massimo Dal ZottoMassimo Dal Zotto
    May 2, 1997 at 3:35 pm
    May 21, 1997 at 6:28 pm
  • Hi, Somebody forwarded this to me, and as the author of CVSup I thought it might be worthwhile to reply: Before you dismiss CVSup because of the language it's written in, I hope you'll take a look at ...
    John PolstraJohn Polstra
    May 21, 1997 at 3:39 am
    May 29, 1997 at 1:01 pm
  • When I found GEQO in Postgres I decided to do some tests. Suprisingly, exhaustive optimalization is always faster than GEQO. I wonder how sophisticated the query has to be to allow GEQO outperform ...
    Michał MosiewiczMichał Mosiewicz
    May 15, 1997 at 1:21 am
    May 16, 1997 at 2:11 am
  • BSDI, which just shipped their release two months ago, shipped flex 2.5.3, so people are still using it. Yes people can upgrade, but I prefer they get an explicit warning before it compiles, and ...
    Bruce MomjianBruce Momjian
    May 14, 1997 at 2:20 pm
    May 15, 1997 at 12:50 am
  • Are transactions supposed to be aborted if there is a syntax error in a query? I am not sure whether it was the same in 6.0, but any 6.1 source does this. You can start a transaction, and if you have ...
    System AdministratorSystem Administrator
    May 2, 1997 at 9:50 pm
    May 10, 1997 at 1:17 am
  • When I create a table with super-user account, everything works well. But when I create it without the super-user account, the table appears 4 times: siaac- CREATE TABLE foo (int foo1, int foo2, int ...
    Robson Paniago de MirandaRobson Paniago de Miranda
    May 28, 1997 at 12:07 pm
    May 31, 1997 at 12:11 am
  • I think these need to be fixed before the release of 6.1. Otherwise this release is basicly a non-starter for any serious use, IMO. I say this because purify detects loads of *serious* problems in ...
    Robert WithrowRobert Withrow
    May 30, 1997 at 2:46 am
    May 30, 1997 at 2:22 pm
  • Hi, I'm getting the following "strange" behaviour with the new date/time code. Platform is SPARC-Solaris 2.5 (again!!) I'm guessing it may be something to do with alignment? rundh= show datestyle; ...
    Keith ParksKeith Parks
    May 25, 1997 at 11:00 am
    May 27, 1997 at 12:24 am
  • from what I noticed there is still no checking for database ownerships when users connect/delete databases...I can create my own database, then su to another user who has a privilege to ...
    System AdministratorSystem Administrator
    May 20, 1997 at 11:11 pm
    May 21, 1997 at 3:15 pm
  • this is inconsistent with other comments on this list. As far as I understand it, a select in a transaction does not imply locking (in postgres). I've seen several solutions involving doing an update ...
    Sean LyndersaySean Lyndersay
    May 30, 1997 at 3:12 pm
    May 31, 1997 at 11:03 am
  • Hi! Next attempt to get better execution plans from optimizer... My test involved one BIG (700K tuples), SMALL (266 tuples) and VERY_SMALL (18 tuples) tables. Query returns 49K tuples. Current ...
    Vadim B. MikheevVadim B. Mikheev
    May 16, 1997 at 5:47 pm
    May 31, 1997 at 2:53 am
  • I have patched PQsetdb and PQfnumber() to properly lowercase the database and field names supplied. I had forgotten that we lowercase identifiers supplied by PQexec(), but identifiers can be supplied ...
    Bruce MomjianBruce Momjian
    May 20, 1997 at 2:46 pm
    May 20, 1997 at 11:16 pm
  • I am sorry to bother you, but majordomo does not remove me from the postgresql lists and the list owner does not respond either, so I will have to repeatedly mail to this list that I want to get off ...
    Gunther SchadowGunther Schadow
    May 9, 1997 at 11:09 am
    May 16, 1997 at 9:00 pm
  • My name is jungwoon min from korea I'm trying to use PostgreSQL with Apache server but apparently have a configuration problem with Apache. I'm trying to retrive data from Netscape using pg95perl. ...
    JungwoonJungwoon
    May 16, 1997 at 12:01 am
    May 16, 1997 at 5:14 pm
  • Hi... There are a few things that I'd like to propose for post v6.1 in order to try and clean things up. The source/docs distribution is getting up to almost 3Meg in size, which is quite large, ...
    The Hermit HackerThe Hermit Hacker
    May 15, 1997 at 1:06 am
    May 15, 1997 at 6:35 pm
  • Thomas has a good idea here. If we don't provide documentation on the new features, it is hard for people to use them. People can use the current TODO list of changes as a roadmap to what new ...
    Bruce MomjianBruce Momjian
    May 14, 1997 at 9:56 pm
    May 15, 1997 at 4:28 pm
  • Probably it's been discussed already... But I can't find the right posting. I thought I switch to new 6.1. I dumped all db's using pg_dumpall and I undumped it as user 'postgres' using syntax: cat ...
    Michał MosiewiczMichał Mosiewicz
    May 14, 1997 at 6:48 pm
    May 14, 1997 at 7:45 pm
  • Hi hackers, I'd like to propose that we add a new option to the "create database" command to allow the specification of the database administrator. Currently the user that executes the "createdb" ...
    Keith ParksKeith Parks
    May 12, 1997 at 10:40 pm
    May 13, 1997 at 12:43 am
  • It seems that one thing which Postgres is lacking is a flexible, cross-platform means for doing security. Kerberos and identd are often not welcome at some installation sites, so resorting to "trust" ...
    Ben EllistonBen Elliston
    May 3, 1997 at 6:21 am
    May 7, 1997 at 8:53 am
  • Hi... It was proposed (by Adrian, I believe?) that we should change the directory structure of the source tree slightly so that all the 'interfaces' are under one subdirectory: ...
    The Hermit HackerThe Hermit Hacker
    May 1, 1997 at 3:58 pm
    May 3, 1997 at 11:46 pm
  • Doug is /really/ good at that sort of thing. Do you read Dilbert? Are you familiar with the term "induhviduals"? Maybe I'm too cynical, but I find that many people are uneducable. </chris Those who ...
    Christopher HicksChristopher Hicks
    May 1, 1997 at 6:43 am
    May 1, 1997 at 7:08 am
  • Has someone installed a patch to make GEQO work only with 6 tables, and changed the SET GEQO to accept a number? Are we going to do that before 6.1 is released? Just curious where we are on this. - ...
    Bruce MomjianBruce Momjian
    May 30, 1997 at 7:54 pm
    May 31, 1997 at 3:29 pm
  • Thanks for replying. I hadn't applied your patch since it appeared to be already applied. We both now know that the patch was backward, judging by later messages. Robert Bruccoleri posted another ...
    David FriendDavid Friend
    May 21, 1997 at 11:28 am
    May 25, 1997 at 12:07 pm
  • Ok. I took out all the logic to compile and run lextest. Now the gmake process gets passed that and compiles for a while until it dies again. This is the output before the compilation dies: gcc -o ...
    Arthur M. KangArthur M. Kang
    May 16, 1997 at 9:18 pm
    May 17, 1997 at 2:21 am
  • Pardon me if this has been asked before... I'm not sure if this really is a bug so I'll post it in "questions". I've been using PostgreSQL 6.0 for a few months. Last night, I decided to try one of ...
    Michael J. MaravilloMichael J. Maravillo
    May 16, 1997 at 1:38 pm
    May 17, 1997 at 12:46 am
  • Major ones...with a release less then 2 days away? Is this patch guaranteed not to throw a 'wrench in the works'? Considering that any new features have been halted weeks ago, and just fixes going ...
    The Hermit HackerThe Hermit Hacker
    May 15, 1997 at 2:29 am
    May 15, 1997 at 9:12 pm
  • Hi, unfortunately I did not receive any response from questions@ so maybe you could answer this question directly. I'm not shure if I made a mistake or stumbled over a bug... Ok, here it is: I'm ...
    Michael ReifenbergerMichael Reifenberger
    May 19, 1997 at 12:26 pm
    May 31, 1997 at 3:26 am
  • System: Solaris 2.5 / gcc 2.7.2 / Bourne Shell Flex 2.5.4 (as reported by flex --version) Postgres account is cleaned and a new .tar.gz (not the chngd file) is used for each regression run. ...
    Adrian HallAdrian Hall
    May 30, 1997 at 2:14 pm
    May 30, 1997 at 2:58 pm
  • As long as the developers aren't afraid that there will be a flood of new conficting code uploaded the day after the release, or that they can't continue developing new stuff at home easily without ...
    David FriendDavid Friend
    May 27, 1997 at 1:34 am
    May 28, 1997 at 2:17 am
  • I decided to run the regression test, and found some problems. I am getting a core dump when running the tests. The following patch fixes the core dump. There seems to be some confusion over the ...
    Bruce MomjianBruce Momjian
    May 24, 1997 at 6:18 pm
    May 25, 1997 at 2:35 pm
  • Vadim, are we keeping this. Do we need to add it to the manual pages and psql help? What are we doing with GEQO and when it gets invoked. Do we need to update the manual page to explain it? - -- ...
    Bruce MomjianBruce Momjian
    May 22, 1997 at 6:44 pm
    May 23, 1997 at 8:32 am
  • Hi hackers, Bruce, I've been playing with the new pg_dumpall command and have found what appears to be a bug in the Solaris shell /bin/sh The code fragment:- psql -l -A -q -t | tr '|' ' ' | grep -v ...
    Keith ParksKeith Parks
    May 22, 1997 at 8:50 pm
    May 23, 1997 at 12:15 am
  • I have modified pg_dumpall and psql to allow '\connect database user', and for pg_dumpall to use it. When pg_dumpall is done loading, database owners are preserved, and every table in the database is ...
    Bruce MomjianBruce Momjian
    May 21, 1997 at 3:40 am
    May 21, 1997 at 2:13 pm
  • dear hackers: if this is the wrong list, sorry and please correct me... I'm trying to get a recent 6.1 to build, preparatory to upgrading "wisql" to work with the new postgres. The instructions in ...
    De ClarkeDe Clarke
    May 19, 1997 at 8:42 pm
    May 20, 1997 at 2:56 am
  • This raises the point: do we actually know which platforms have been tested for this release? I have i686/Linux/gcc and have seen traffic or talked with folks using Sparcs. Obviously the other ...
    Thomas G. LockhartThomas G. Lockhart
    May 15, 1997 at 6:24 pm
    May 19, 1997 at 3:31 am
  • Great, version 6.1 worked far better for me than any previous version of PostGres. When I started postmaster with normal fsync behavior it ran the filesystem for a good 30 minutes before I stopped ...
    Brian BartholomewBrian Bartholomew
    May 14, 1997 at 4:25 am
    May 14, 1997 at 5:46 pm
  • Hello! Can anyone mail me his patches using TCP_NODELAY for version 6.0? I got a speed problem and hope this helps for me, too. Thanks Martin Martin.H.Ludwig@rz.ruhr-uni-bochum.de ...
    Martin H. LudwigMartin H. Ludwig
    May 13, 1997 at 6:17 pm
    May 14, 1997 at 11:05 am
  • from 970512 src/libpq/fe-connect.o, line 723: //----------------------------------------------- static void startup2PacketBuf(StartupInfo* s, PacketBuf* res) { char* tmp; /* res = ...
    Eric BuddingtonEric Buddington
    May 12, 1997 at 7:59 pm
    May 13, 1997 at 1:45 am
  • Hi, I have been looking at the parser, and came across the following: There is a rule for 'CAST attr AS type' included in 'a_expr'. 'a_expr' is included in 'res_target_el2' (which is the target list ...
    Maarten BoekholdMaarten Boekhold
    May 12, 1997 at 9:23 pm
    May 12, 1997 at 10:52 pm
May 1997
1234
Group Navigation
period‹ prev | May 1997 | next ›
Group Overview
grouppgsql-hackers @
categoriespostgresql
discussions158
posts476
users74
websitepostgresql.org...
irc#postgresql

74 users for May 1997

Bruce Momjian: 105 posts The Hermit Hacker: 63 posts Vadim B. Mikheev: 40 posts Thomas G. Lockhart: 27 posts David Friend: 25 posts Keith Parks: 23 posts Christopher Hicks: 19 posts System Administrator: 12 posts Thomas Lockhart: 10 posts D'Arcy J.M. Cain: 7 posts Darren King: 7 posts James A. Hillyerd: 7 posts Michał Mosiewicz: 7 posts Brian Bartholomew: 6 posts Brian Bartholomew: 6 posts Eric Buddington: 6 posts David Wetzel: 5 posts Sean Lyndersay: 5 posts James Hughes: 4 posts Leo Shuster: 4 posts
show more
Archives