[prev in list] [next in list] [prev in thread] [next in thread] 

List:       postgresql-general
Subject:    Re: [HACKERS] Forcing current WAL file to be archived
From:       Simon Riggs <simon () 2ndquadrant ! com>
Date:       2006-07-31 22:45:35
Message-ID: 1154385935.3226.30.camel () localhost ! localdomain
[Download RAW message or body]

On Tue, 2006-08-01 at 00:40 +0300, Hannu Krosing wrote:
> Ühel kenal päeval, T, 2006-07-25 kell 17:05, kirjutas Simon Riggs:
> > On Tue, 2006-07-25 at 11:53 -0400, Tom Lane wrote:
> > > That's fine, but feature freeze is in a week and we don't even have
> > > the
> > > basic function for manually doing a log file switch.  Let's get that
> > > done first and then think about automatic switches.
> > 
> > Agreed.

So: automatic switching of xlogs....

I've written a patch to implement archive_timeout, apart from the
infrastructure required to allow archiver to use LWLocks.

If we do this, it will allow the archiver to write to shared memory and
log files in particular. People may have a robustness issue with that,
so I'd like to check before doing this.

As a result, I'm thinking: What's the minimum infrastructure I can get
away with?

I'll post to -patches what I've got, to further this discussion.

-- 
  Simon Riggs
  EnterpriseDB          http://www.enterprisedb.com


---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
       choose an index scan if your joining column's datatypes do not
       match

[prev in list] [next in list] [prev in thread] [next in thread]