site_archiver@lists.apple.com Delivered-To: darwin-kernel@lists.apple.com Well, yes, by definition. S+E -- Quinn "The Eskimo!" <http://www.apple.com/developer/> Apple Developer Relations, Developer Technical Support, Core OS/Hardware At 10:28 -0700 23/10/06, Rick Mann wrote: Would the open file show up in the VFS data structures? The problem is that "lsof" traces file descriptors from the process down. So it iterates the process list, then iterates the descriptor table within each process that it finds, then works out what those descriptors are. <x-man-page://2/unmount> works in a completely different way [1]. VFS knows which vnodes are associated with which mount points. When you try to unmount a volume, VFS iterates that list reclaiming each vnode (that is, detaching it from the underlying file system node). If this fails, the unmount fails. It typically fails because one of the vnodes has a long term ("use count") reference. A vnode gets a use count reference whenever it's attached to a new descriptor, but it can also get one in other ways (like the file being opened by kernel code). [1] The following is a gross simplification; my apologies to the VFS experts out there. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.a... This email sent to site_archiver@lists.apple.com