We had a case where having it to anything different then zero was simply not acceptable :) As the create of a row, and the immediate update of the row afterwards were happening on different nodes. Not necessary a good thing, but that's how it works :) And the only solution was to set commit delay to zero. On 6/10/05, Connor McDonald <mcdonald.connor@xxxxxxxxx> wrote: > In the absense of currently having a RAC environment to play around > with, does anyone have any "gut feel" responses to the impact of > dropping the commit propagation: >=20 > a) to any value under its default > b) to zero >=20 > (yeah, I know this is a "how long is a piece of string" question) >=20 > Cheers > Connor >=20 > --=3D20 > Connor McDonald > =3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D= 3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D > =3D3D=3D3D > email: connor_mcdonald@xxxxxxxxx > web: http://www.oracledba.co.uk >=20 > "Semper in excremento, sole profundum qui variat" > -- > //www.freelists.org/webpage/oracle-l >=20 --=20 Christo Kutrovsky Database/System Administrator The Pythian Group -- //www.freelists.org/webpage/oracle-l