Guan Xin: > drwxrwxrwx 24 root root 428 10=E6=9C=88 15 14:24 /mnt/0/ <- bottom, s= > quashfs > drwxr-xr-x 9 root root 200 10=E6=9C=88 15 20:11 /mnt/1/ <- top, tmpf= > s > drwx------ 4 root root 4096 10=E6=9C=88 10 12:54 /mnt/2/ <- middle > > But I still don't fully understand, if the top branch has a dir with > the same name > as that of a dir in an underlying branch, why the permission of the > dir in the top > branch does not override the permission of the same dir in the underlying > branch?
No. To keep the consistency from the point of middle fs's view, the upper permission bits should not override the lowers. As long as the middle branch prohibits such access, aufs simply follow it. Otherwise it can be a violation of a security feature which the middle branch fs has. For your information, here is the aufs behaviour for the dir permission check. - for access other than dir, aufs checks the first one only. "first" means that the file firstly aufs found. - for "write" to dir, aufs checks the first one only too, since aufs will never write to the lower branches. - for "read" to dir, aufs checks the same named dirs on all branches, since aufs reads all of them. J. R. Okajima ------------------------------------------------------------------------------ Don't let slow site performance ruin your business. Deploy New Relic APM Deploy New Relic app performance management and know exactly what is happening inside your Ruby, Python, PHP, Java, and .NET app Try New Relic at no cost today and get our sweet Data Nerd shirt too! http://p.sf.net/sfu/newrelic-dev2dev
