I have an exchange 2007 SP1 CCR cluster and the active node log files grow out of control at about 10GB per hour with no database growth, No accounts seem to grow either and there are no indications in message tracking that this amount of data is being transmitted during the symptom time. I've ran the client monitor and saw nothing unusual. A backup clears the logs naturally. The symtoms stop as randomly as they start with no indication as to what is generating or how we can stop the sudden growth. There is enough space on the disk for this to happen for a long period of time (150GB ~15 hours) but my client is worried that it might occur for a longer period. The longest observed is around 9 hours. Restarting one of the hub transport servers (nlb clustered) seems to stop the growth but it doesnt matter which one (of the two) is restarted. This could simply be coincidence. Any help is greatly appreciated.