Re: Safely Killing Backends (Was: Applications that leak connections)

On Fri, Feb 04, 2005 at 05:01:43PM -0500, Jim Wilson wrote:

> Rather than getting into the raised eyebrow thing ;-), I\\\'d suggest
> checking your "qualifiers". Consider that with Postgres, if killing a
> single connection brings the whole server down, you will loose _all_
> uncommitted data. If you did not, then I would call that a bug. The
> weakness is not in the data integrity (directly), it is in the
> integrity of the server processes and their managability.

Are you saying that your applications regularly leave uncommitted
transactions for long periods of time? That sounds like bugs in your
applications to me.

Maybe I didn't get the part about lost connections. Do you mean that
you applications lose conectivity to the server, and thus the
transaction they were working with are lost? If that's the case, then
it certainly sounds dangerous to commit whatever was there; what if the
transaction was incomplete? Of course, if you can't commit it, the only
way to proceed is to roll it back.

What's with the backslashes anyway?

--
Alvaro Herrera (<alvherre[(at)]dcc(dot)uchile(dot)cl>)
Y dijo Dios: "Que sea Satanás, para que la gente no me culpe de todo a mí."
"Y que hayan abogados, para que la gente no culpe de todo a Satanás"