Thank you all for your time. From my point of view the case is settled. "noatime" did a nice job and as a side-effect i just readjusted my understanding of journaling-fs to reality ;-)
It would also be nice to get the source of the reads that are forcing atime-updates, but i did not yet find a way to do so. I suspect lsof to be of help here or kerneltop. Both are wild guesses. so hints are still welcome... Matthias Pfeifer -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
