Re: Databases and servers

From: Jeff Janes <jeff(dot)janes(at)gmail(dot)com>
To: Karl Martin Skoldebrand <KS0C77263(at)techmahindra(dot)com>
Cc: "pgsql-general(at)lists(dot)postgresql(dot)org" <pgsql-general(at)lists(dot)postgresql(dot)org>
Subject: Re: Databases and servers
Date: 2019-08-20 21:26:01
Message-ID: CAMkU=1x=MTXA7NqLLkQVN=o0qGjY57G7JuACdSgvR6EUaAg-Bg@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Tue, Aug 20, 2019 at 6:33 AM Karl Martin Skoldebrand <
KS0C77263(at)techmahindra(dot)com> wrote:

> Hi,
>
> I just discovered that a client has done this:
>
> They have two web applications A1 and A2. They have seperate
> hostnames/URLs. Both have a production and a test database A1p and A1t/ A2p
> and A2t.
>
> What they've done is have both A1p and A2p on the same actual databaser
> server and A1t and A2t on the same server.
>
> Are these two PostgreSQL instances running on the same hardware, or two
databases within a single PostgreSQL instance?

> So, I'm thinking - if a bug in application A1 crashes the application and
> database badly it will risk bringing down both services A1 and A2.
>
Is this a common occurrence? Of all the occurrences of downtime in recent
memory (or better yet, from incidence documentation), what were the causes
of them? Is this near the top of the list?

Also, are the two apps completely independent, or are they used together
such that one being down makes the other one not very useful?

> The same risk would be evident on a successful security breach.
>
On the other hand, more servers means more moving parts, means more
opportunities for mistakes in configuration or maintenance that let
breaches happen.

> I would prefer to A1p and A2p on seperate servers, maybe keeping A1t and
> A2t on the same. (This is what seems to be happening when the database
> servers are being repladed).
>
I don't know what that last part means.

> What is the general thought on the current setup?
>
In my experience, people acting on mere conjectures about what might cause
downtime in the future and how to prevent it have caused more downtime than
they have prevented.

Cheers,

Jeff

>

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message chiru r 2019-08-20 21:30:44 Re: Two Postgres master process are showing - one is on and off
Previous Message Peter J. Holzer 2019-08-20 20:58:21 Re: Databases and servers