On Mon, 2005-05-16 at 10:02 -0600, Josh Coates wrote: > >A file is basically just a collection of > >inodes, and the file name is tied to the first inode. > > just for the sake of squashing misinformation, the statement above is a > false. i'm sure lots of people caught that, but for record i figured it'd > be a good thing to correct. > > traditional unix file systems have 1 inode per file. inodes contain meta > data, and pointers to data blocks that contain the actual data (and for very > large files, it points to blocks of pointers to blocks) - and for very small > files, the inodes contain actual file data.
Ah, I see the difference. For some reason I had thought all the data blocks were called inodes. Mea culpa. Where that idea came from I have no idea. I blame it on violent video games. Corey
signature.asc
Description: This is a digitally signed message part
.===================================. | This has been a P.L.U.G. mailing. | | Don't Fear the Penguin. | | IRC: #utah at irc.freenode.net | `==================================='
