Hi, On Sun, Mar 31, 2002 at 02:35:18PM +0100, Russell King wrote: > Subject line says it all. When one of the linux.org.uk machines got > rebooted, it performed a full fs check of the main 160GB ext3 > partition claiming it contained errors. During the check it found: > > Deleted inode 142279 has zero dtime. FIXED. > > The kernel message log contained lots of: > > EXT3-fs error (device md(9,1)) in ext3_new_inode: error 28 > > We had used up all the available inodes. Since the above message is > generated from __ext3_std_error, it marks the filesystem bad. This > doesn't sound like proper behaviour. Right. Fixed in ext3 cvs ... > Kernel: Red Hat 2.4.9-21 ... and in the rawhide kernel. --Stephen
