Re: [GENERAL] Need setup help for Postgresql 8.1.3 on Solaris 10

2007-01-16 Thread Gellert, Andre
Hello Richard, > > I decreased the import time from >40 minutes to 32 minutes with changing > > the parameters to: > > > > shared_buffers = 45 # kept same > > temp_buffers = 10# kept same > > > > work_mem = 32768 # now 32mb , before: 102400 = 100mb > > maintenanc

Re: [GENERAL] Need setup help for Postgresql 8.1.3 on Solaris 10

2007-01-12 Thread Gellert, Andre
Hello Richard, > > We have such a testsystem now, 4 x AMD Opteron (double core techn.) > > 885 2.6ghz , running with 32 GB Ram and fast 10.000rpm SAS discs, > > How many disks? What RAID? 4 Disks, RAID1. > Is your disk I/O saturating? Try vmstat/iostat (or whatever the Solaris > equivalent i

[GENERAL] Need setup help for Postgresql 8.1.3 on Solaris 10

2007-01-12 Thread Gellert, Andre
Hello all, I need some hints how to setup Postgresql on a brand new testsystem. Due to heavy load on the current database servers, my boss decided to test a big server as a replacement for 5 other servers. ;-) The system is used in a extranet environment - over 50 percent of the load is prod

[GENERAL] Wanted: Want to hide data by using PL/PGSQL functions

2004-07-21 Thread Gellert, Andre
Hello, I have following problem: A user "xy" shouldn't have any rights to a table, but needs data from the content of the table. My idea was to setup a PL/PGSQL procedure to fetch the data from the table, so that the user only is allowed to access the procedure. I also tried using a SQL function

Re: [GENERAL] Cache lookup failure for pg_restore?

2004-05-12 Thread Gellert, Andre
Hello, > Recently, we discover how to avoid problem with OIDs backup/restore > in tsearch2. Check > http://www.sai.msu.su/~megera/postgres/gist/tsearch/V2/ for > a little patch (regprocedure_7.4.patch.gz). It won't work on > existed tsearch2 > installation, though, but will help in future. I d

[GENERAL] Are Postgres 7.4.1 RPMs available for SuSE 9.0 ?

2004-03-04 Thread Gellert, Andre
Hi everybody, currenty we use Suse Linux 9 and the rpms with 7.4.0 downloaded from ftp://ftp.gmd.de/mirrors2/suse/ftp.suse.com/people/max/postgresql-7.4/ . I don't know how to contact the person who build this rpms , and i cannot find other resources on the web, so does anyone know a place to g

[GENERAL] Mistake in my query or Index Scan on subquery failure? (7.4)

2004-03-03 Thread Gellert, Andre
Hello, I have a problem with this simple query : explain select * from ref_artikel a where a.artnr in ( 351275 , 351346 , 293082 ) LIMIT 20 OFFSET 0 ; QUERY PLAN -

Re: [GENERAL] PHP + Postgres: More than 1000 postmasters produce

2004-02-24 Thread Gellert, Andre
Hi to all, those helped or enjoyedmy english. I waited over the day to give a summary. > your pgsql backend must be able to accommodate the max persistent > connections * the max number of apache daemons. > > the php pgsql.max_persistent setting is per apache daemon. All of your answers where

Re: [GENERAL] Reset oid , starting value=1 -- Max. Number of OID

2003-12-09 Thread Gellert, Andre
> The only way to reset oids is to destroy the entire data directory and > initdb again. In recent versions you can create tables > without oids, thus > reducing their usage. Today we have oid's > 2147483647 , what happens, if we have more than allowed ? I remember it was about 4 billions max. oi

Re: [GENERAL] Reset oid , starting value=1

2003-12-09 Thread Gellert, Andre
Hi Martijn, > > There must be a global value which sets the last oid, i need to set this to > > "1" or to fill up from 1 , so i have no oids >= 2147483647 for the near > > future.. how can i do this ? Fixing the mentioned above will take more time. > > The only way to reset oids is to destroy the

[GENERAL] Reset oid , starting value=1

2003-12-09 Thread Gellert, Andre
Hello, i have a bug in a php application based on postgres 7.2.2 , which cannot handle oid's greater than 2147483647 . So I have to reset the used oids to start from 1 again, so that we have more time to correct this bug, I tried to delete dropdb, createdb and read back the data, but the oids are