Re: On "multi-master"

From: Chris Travers <chris(at)metatrontech(dot)com>
To: Andrew Sullivan <ajs(at)crankycanuck(dot)ca>
Cc: Tatsuo Ishii <ishii(at)sraoss(dot)co(dot)jp>, pgsql-general(at)postgresql(dot)org
Subject: Re: On "multi-master"
Date: 2005-10-16 00:58:20
Message-ID: 4351A5AC.6040203@metatrontech.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Andrew Sullivan wrote:

>On Fri, Oct 14, 2005 at 11:54:19PM +0900, Tatsuo Ishii wrote:
>
>
>>Enforcement? There would be plenty of ways to achieve that. For
>>example, you could set pg_hba.conf so that on ly the host where pgpool
>>is running on could connect to the host where postmaster is running
>>on.
>>
>>
>
>That just changes the problem to someone logging in from that host.
>(This isn't a theoretical problem, by the way; it's an objection that
>I've heard from people.)
>
>

Repeat after me... Data integrity, like security, is a process, not a
product.

Your objection is not an insoluble one. But it involves care in
designing the system.

If your DBA logs in to the actual DB host and issues a command, then
that is the DBA's fault, and your DBA should know better. If he/she
doesn't you have bigger problems than replication. A more valid concern
is the question of not knowing that the DBA logged in from the wrong
system to the wrong system. This can be greately reduced by
intelligently engineering the network, and setting things up so that the
system automatically logs in with a read-only account unless a username
is specified.... I think that this risk can be reduced to something
acceptable even in the world of high availability systems, but you are
right, it is not so done out of the box.

I think that this objection comes from the difference between FOSS where
you have projects as toolkits for solutions and COTS Software where the
packages are supposed to be complete solutions.

>
>
>
>>Right. It's your freedom that you do not use pgpool until you think
>>it's solid enough.
>>
>>
>
>And my managers approve it :) I should note, for the record, that
>I'm extremely impressed with pgpool. I just think we have some room
>to grow before we can say we have something to really compete with
>the commercial multimaster systems.
>
>

Now, I never said it was a direct competitor by itself with commerical
multimaster systems. I do think that with appropriate consulting
services, network engineering, etc. services it could be with a little
work, if these are the only objections you have :-)

Best Wishes,
Chris Travers
Metatron Technology Consulting

Attachment Content-Type Size
chris.vcf text/x-vcard 127 bytes

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Chris Travers 2005-10-16 01:04:54 Re: On "multi-master"
Previous Message Chris Travers 2005-10-16 00:46:03 Re: PostgreSQL Gotchas