On Wed, Feb 02, 2022 at 03:18:34PM -0500, Sean Caron wrote: > Hi Dave, > > It counted up to inode 13555712 and then crashed with the error: > > malloc_consolidate(): invalid chunk size That sounds like heap corruption or something similar - that's a much more difficult problem to track down. Can you either run gdb on the core file it left and grab a stack trace of where it crashed, or run metadump again from gdb so that it can capture the crash and get a stack trace that way? > Immediately before that, it printed: > > xfs_metadump: invalid block number 4358190/50414336 (1169892770398976) > in bmap extent 0 in symlink ino 98799839421 I don't think that would cause any problems - it just aborts processing the extent records in that block and moves on to the next valid one that is found. Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx
