On Wed, 27 Oct 2004, Alan Cox wrote: > On Wed, Oct 27, 2004 at 11:21:30AM -0700, Dan Hollis wrote: > > Same here, we've had numerous catastrophic filesystem failures with ext3 > > in production systems. > Should only occur on drive failure, things like bad ram, and IDE drives > when you hard reboot and don't have them set to writethrough caching. Happened on power failures or on situations that called for the red button (eg runaway processes etc). > What releases ? 2.4.x kernels, which we also use reiserfs on. reiserfs didn't fail under the exact same situations on the exact same hardware. we have a few legacy systems still on ext3 but we are upgrading them to reiserfs as we get the chance. fwiw we also had bad experiences with xfs. -Dan