Skip site navigation (1) Skip section navigation (2)

Re: wal logs in pg_xlog

From: Nasir Iqbal Danish <nasir(at)amaana(dot)com>
To: Michael Fuhr <mike(at)fuhr(dot)org>
Cc: pgsql-admin(at)postgresql(dot)org
Subject: Re: wal logs in pg_xlog
Date: 2005-06-28 12:52:34
Message-ID: 42C14812.7060607@amaana.com (view raw or flat)
Thread:
Lists: pgsql-admin
Michael Fuhr wrote:

>On Mon, Jun 27, 2005 at 11:04:50AM +0500, Nasir Iqbal Danish wrote:
>  
>
>>I am using postgres 8.0.3, and wanted to use Point In Time Recovery enable.
>>could any one tell me about how to indentify the current WAL file, also 
>>let me know if there is any fix patteren for WAL log file generations.
>>documentaion says it will generate ever increasing number starting from 
>>some thing like "0000000100000000000000000". but I have observed 
>>different behaviour.
>>    
>>
>
>What behavior have you observed?
>
>  
>
the files in pg_xlog like 000010000000000005, 000010000000000006, 
000010000000000007, 000010000000000008, (in short term call these 5,6,7,8).
with file 8 being most recent and 5 being 2 months back, with respect to 
time stamp. after I did some work and then observed log files, now 5 and 
6 were showing me latest time stamps. I was confused which is the latest 
one.
I also want to know when new WAL segment files is created, is it 16MB or 
this will start from 0 bytes and size inceases as soon as db server will 
manupulate.

Please provide me if u can some script to identify which one is latest 
file so that I can copy this to some other place.
One more question. Is there a way to build or configure the database 
server to replicate pg_xlogs ?
-Nasir

In response to

Responses

pgsql-admin by date

Next:From: Peter EisentrautDate: 2005-06-28 13:10:53
Subject: Re: Direct I/O versus OS-Caching
Previous:From: KÖPFERL RobertDate: 2005-06-28 11:55:21
Subject: Direct I/O versus OS-Caching

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group