Re: Data loss, vacuum, transaction wrap-around

On Fri, 18 Feb 2005 22:35:31 -0500, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> pgsql(at)mohawksoft(dot)com writes:
> > I think there should be a 100% no data loss fail safe.
>
> Possibly we need to recalibrate our expectations here. The current
> situation is that PostgreSQL will not lose data if:
>
> 1. Your disk drive doesn't screw up (eg, lie about write complete,
> or just plain die on you).
> 2. Your kernel and filesystem don't screw up.
> 3. You follow the instructions about routine vacuuming.
> 4. You don't hit any bugs that we don't know about.
>
I'm not an expert but a happy user. My opinion is:
1) there is nothing to do with #1 and #2.
2) #4 is not a big problem because of the velocity developers fix
those when a bug is found.

3) All databases has some type of maintenance routine, in informix for
example we have (update statistics, and there are others for oracle)
of course they are for performance reasons, but vacuum is too for that
and additionally give us the XID wraparound.
So, to have a maintenance routine in PostgreSQL is not bad. *Bad* is
to have a DBA(1) with no clue about the tool is using. Tools that do
to much are an incentive in hire *no clue* people.

(1) DBA: DataBase Administrator or DataBase Aniquilator???

regards,
Jaime Casanova