Re: [Gnumed-devel] client tcp/ip and Postgres timeouts

[James Busser]

On 16-Aug-08, at 3:34 AM, Karsten Hilbert wrote:

> Tom Lane attests to PostgreSQL not having ...
> a timeout to idle connections... He suggests running an empty
> query every so often.
>
> If dropped connections appear to become a problem I'll add
> that to GNUmed.

I was just aiming to prevent data loss as a consequence of unsaved  
data sitting in an idle GNUmed client session which the provider may  
have parked / minimized while in the meantime focusing on other work,  
maybe even in GNUmed sessions other than the one that is sitting idle.

In parallel to running an empty query every so often (if needed), may  
we desire an otherwise idle GNUmed session would

- at minimum, after some suitable period of inactivity (maybe q15-30  
minutes), attract the attention of the user?

- possibly, if the user does not acknowledge:

        auto-save progress notes?

        enter some form of "locked" mode
        -> would require the user re-provide their password
        (not that any "locked" mode is yet supported...)