Replication options in Postgres

I guess if you don't do deletes then something like selecting all the
records with an oid greater than the last replication cycle would
find the most recent additions.

Erich wrote:

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

1. In my PHP code, I have functions like
inserttransaction(values...). I could just modify inserttransaction()
so that it runs the same query (the INSERT) on two or more DB
servers. This would probably work ok.

Why not have a proxy server that your clients talk to, which replicates
the transaction across the other (independent) backend servers, and only
returns OK if all the backends return OK.

Then, theoretically, your databases should always remain concurrant. You
could dump & diff them periodically to make sure.

Ian
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.1 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE5hjM7fn9ub9ZE1xoRAtevAJ9v7Ik/wtasCyTgeCx+zsYvYQWL4QCgubgx
PE0m/X6VoY7+ESZS/p3CIlQ=
=b6Bv
-----END PGP SIGNATURE-----

Hi,

I want to alter the size of a column, say from char(40) to char(80),
but it seem that
the ALTER does not support such operation, nor does it support column
removing.

How can I do for this ?

Thanks

I want to alter the size of a column, say from char(40) to char(80),
but it seem that
the ALTER does not support such operation, nor does it support column
removing.

How can I do for this ?

I would also like to know how to do both of these things.

Michael Talbot-Wilson wrote:

I want to alter the size of a column, say from char(40) to char(80),
but it seem that
the ALTER does not support such operation, nor does it support column
removing.

How can I do for this ?

I would also like to know how to do both of these things.

I'm not aware of an easy way of doing it. But you can dump your schema
and data separately. Manually edit your schema. Reload the schema then
reload the data.

Right now the best way is probably:

create table newtable ( ... new column info ... )
insert into newtable select * from oldtable;
alter table oldtable rename to old_oldtable;
alter table newtable rename to oldtable;

In the second line, you may not be able to get away with a
* if you're doing more complicated changes of types that
can't automatically converted. And once you're done and
sure everything is working, you can delete the backup of
the old table.

Stephan Szabo
sszabo@bigpanda.com

On Tue, 1 Aug 2000, Cheng Kai wrote:

sorry, forgot to include this address...

Date: Tue, 01 Aug 2000 14:51:03 -0400
To: "Cheng Kai" <chengk@isse.kuis.kyoto-u.ac.jp>
From: mikeo <mikeo@spectrumtelecorp.com>
Subject: Re: [GENERAL] How to alter the size of a column
In-Reply-To: <002301bffb6a$63dcc0a0$9b0210ac@cembaro>
References: <200008010045.RAA22022@cyberpass.net>

<398625F7.C22A345C@nimrod.itg.telecom.com.au>

hi, i changed the size of a column using this method:

tig4=# \d cust
Table "cust"
Attribute | Type | Modifier
-------------------+-------------+------------------------------------------
cust_id | varchar(15) | not null
cut_id | varchar(6) | not null
cust_name | varchar(50) | not null
cust_division | varchar(6) |
cust_svc_start_dt | date | not null default now()
cust_svc_end_dt | date |
cust_valid | char(1) | not null default 'Y'
cust_bill_loc_id | varchar(6) | not null
wu_id | varchar(10) | not null default 'SPECTRUM'
cust_timestamp | timestamp | not null default now()
agt_id | varchar(10) | default 'DEFAULT'
rse_id | integer |
bd_id | varchar(6) | not null
cust_email | varchar(50) |
cust_stream | integer | default nextval('cust_stream_seq'::text)
br_cycle | integer |
cust_qr_reports | varchar(20) |
cust_qr_sent | timestamp |
Indices: cust_cut_idx,
cust_pkey,
cust_stream_idx

update pg_attribute set atttypmod = 19 where attname = 'cut_id' where

attrelid =