Re: reading redo log files

The problem with that suggestion is that you should monitor this on the time it happens. As those things often happen on weekend or night times I should prefer a method making it posible to do the analyse offline ( on monday then ).

I will certainly try your suggestion anyway, perhaps in a cron-job or so.

Jeremiah Wilton <jeremiah_at_wolfenet.com> wrote in article <Pine.OSF.3.95.981007151822.16181A-100000_at_gonzo.wolfenet.com>...
> On 7 Oct 1998, dke wrote:
> >
> > Is there something to read the contents of the redo log files.
> >
> > I think I use too much redo-log space, so I want to know who is doing
this
>
> A much better way to find out who is responsible for excessive redo
> generation is to monitor the growth of statistic# 97 (for Oracle8) in
> v$sesstat for each session connected. If one session is generating more
> redo than the others, then its value for v$sesstat statistic# 97 should
> grow conspicuously.
>
> I think the same statistic ('redo size') for Oracle7 is 85, but check
that
> in v$statname before you try it.
>
> --
> Jeremiah Wilton http://www.wolfenet.com/~jeremiah
>
>
Received on Thu Oct 08 1998 - 13:55:58 CDT