and (sorry) gbak can cause a gap > 20000, can that cause a sweep to be
started when another process 'notices' the gap?

On 11 January 2013 12:06, Nick Upson <nu@...> wrote:

> Hi,
>
> I'm trying to track down an intermittent performance issue, and I find I
> don't have sufficient grasp of sweep behavior.
>
> firebird 2.1.5 classic on centos
>
> I understand that a sweep is started (new thread is created) when OST-OIT
> > 20000 (by default). What I can't find documented is when this check is
> made. Is it made when a client starts a connection, starts a transaction,
> queries a table or something else?
>
>
>
>
> --
> Nick Upson, Telensa Ltd
>



--
Nick Upson, Telensa Ltd


[Non-text portions of this message have been removed]