Eugene,I am glad that you were able to find the problem using Brandon's query against v$sesstat. The one thing I will point out is that the numbers from that query are cumulative from the date of instance startup. Using it, you could wrongly identify a long-lived session that generates lots of redo, but very slowly over long periods of time. That's why the approach with Snapper is better.
You could also divide the value from v$sesstat by how long the instance has been up. It is not as reliable as Snapper, but it is more accurate than looking at grand totals.
Regards, Jeremiah Wilton Blue Gecko, Inc. http://www.bluegecko.net On Dec 1, 2009, at 10:38 AM, Eugene Pipko wrote:
Brandon,Thank you very much for the query. I was able to find the offending session and killed it. It looks like everything is back to normal (for now).Tanel, I will definitely check “snapper” sometime soon. Thank you.