Re: How do you change the size of the WAL files?

From: "David Johnston" <polobo(at)yahoo(dot)com>
To: "'Ron Somaraju'" <RSomaraju(at)masergy(dot)com>, "'Tom Lane'" <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: "'Scott Marlowe'" <scott(dot)marlowe(at)gmail(dot)com>, <pgsql-general(at)postgresql(dot)org>
Subject: Re: How do you change the size of the WAL files?
Date: 2012-01-10 15:33:34
Message-ID: 015d01cccfad$37d99540$a78cbfc0$@yahoo.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

-----Original Message-----
From: pgsql-general-owner(at)postgresql(dot)org
[mailto:pgsql-general-owner(at)postgresql(dot)org] On Behalf Of Ron Somaraju
Sent: Monday, January 09, 2012 8:21 PM
To: Tom Lane
Cc: Scott Marlowe; pgsql-general(at)postgresql(dot)org
Subject: Re: [GENERAL] How do you change the size of the WAL files?

Once again, pros and cons should be left to users discretion because one may
have latest and greatest hardware and network resources. For example a SSD
on a fiber channel on a high speed network.

----------------------------------------------------------

And determining whether such a run-time configuration is feasible should be
left to programmer's discretion since they have the best chance of knowing
all the different parts of the system that relate to the feature/ability in
question. You are right in that everything should be end-user configurable
but maybe there are reasons that is not possible or desirable in specific
situations. Regardless, the designers still have to pick reasonable
defaults since the configuring 500 settings just to install the software is
not realistic or desirable in its own right.

While your concerns and reasoning are well-founded currently the capability
to dynamically adjust the WAL file size is not present and so the question
becomes whether you can convince the community to add such functionality in
a timely enough fashion or whether it is important enough to you to contract
one of the service providers to research and make the necessary
modifications. Simply being right doesn't mean that the current (wrong)
state is going to go away - especially since it isn't broken but rather is
not as flexible as it possibly could be.

David J.

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message David Johnston 2012-01-10 15:48:10 Re: string = any()
Previous Message Andy Colson 2012-01-10 15:32:48 Re: string = any()