[HACKERS] For Review: Allow WAL information to recover corrupted pg_controldata patch

Hi All I had added an option -r to pg_resetxlog to enable the tool can rebuild the corrupted pg_control file from the old xlog files. here is the patch. Sorry I had tried to attached it to the mail, but it failed, I dont know why, here is the link:http://www.geocities.com/yuanjia_pg/pg_resetxlog.d

Re: [HACKERS] Enhanced containment selectivity function

Hi, Moving it in contrib/ltree would be more difficult to me because it depends on other functions declared in selfuncs.c (get_restriction_variable, etc). I'd be willing to consider exporting those functions from selfuncs.c. In the meanwhile here is the latest patch which uses both mcv and

[HACKERS] unexpected pageaddr on startup/recovery

Hackers, I had a system crash today. When Postgresql started I had the following in my pg.log file. 2005-08-06 14:14:26 [3352] LOG: database system was interrupted at 2005-08-06 11:57:28 EDT 2005-08-06 14:14:26 [3352] LOG: checkpoint record is at 5E5/9CAEA594 2005-08-06 14:14:26 [3352] LOG:

Re: [HACKERS] unexpected pageaddr on startup/recovery

"Jim Buttafuoco" <[EMAIL PROTECTED]> writes: > 2005-08-06 14:14:26 [3352] LOG: database system was not properly shut down; > automatic recovery in progress > 2005-08-06 14:14:26 [3352] LOG: redo starts at 5E5/9C6796A0 > 2005-08-06 14:17:17 [3352] LOG: unexpected pageaddr 5E3/A7BFA000 in log fil

[HACKERS] Stefan Simkovics' paper and pgsql version 8.1

Hi, I would like to use postgresql for a couple of database experiments I've been thinking about. I found Stefan Simkovic's paper "Enhancement of the ANSI SQL Implementation of PostgreSQL." It is a great guide, although I'm wondering: 1. How relevant is this paper now (with release 8.1 in the wor

Re: [HACKERS] Stefan Simkovics' paper and pgsql version 8.1

On Sat, Aug 06, 2005 at 05:44:14PM -0400, Shahbaz wrote: > I would like to use postgresql for a couple of database experiments > I've been thinking about. I found Stefan Simkovic's paper > "Enhancement of the ANSI SQL Implementation of PostgreSQL." It is a > great guide, although I'm wondering:

Re: [HACKERS] Stefan Simkovics' paper and pgsql version 8.1

Alvaro Herrera <[EMAIL PROTECTED]> writes: > Interestingly, the paper is mentioned in the bibliography of the > official PostgreSQL documentation. Apparently there's a good primer on > relational theory, SQL and PostgreSQL history. Most of that material is still in the source tree, though it seem

Re: [HACKERS] Stefan Simkovics' paper and pgsql version 8.1

sql.sgml just seems to be an intro to sql/relational. The Simkovics paper is absolutely perfect for someone like me who wants to try some experiments but can't commit too much time! Does any one think time is ripe for an update to the paper??? ;) Actually I always wanted a whole book on postgres

Re: [HACKERS] unexpected pageaddr on startup/recovery

thanks -- Original Message --- From: Tom Lane <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Cc: "pgsql-hackers" Sent: Sat, 06 Aug 2005 17:24:46 -0400 Subject: Re: [HACKERS] unexpected pageaddr on startup/recovery > "Jim Buttafuoco" <[EMAIL PROTECTED]> writes: > > 2005-08-06 14:14:2

[HACKERS] gettime() - a timeofday() alternative

Hi all, I propose to add an internal function gettime() that transparently returns the current system time, as a timestamptz with maximum precision. Calling gettime() would be a more elegant approach than calling timeofday() and converting it to a timestamp, and avoids some of the potential probl