On Tue, Oct 15, 2013 at 09:45:27PM +0800, Roy Zhang wrote: > Hi, > I met a problem that cannot mount xfs, log as below. > I got know xfs_repair -L will fix this situation, I want to know how > and why cause it, it's a bug in xfs or hdd? Is there any patch to fix > it? > > Thanks > Roy > > Oct 9 18:26:52 mcluster-alpha-node3 kernel: [11840.232985] XFS > (dm-3): Mounting Filesystem > Oct 9 18:26:52 mcluster-alpha-node3 kernel: [11840.327773] XFS > (dm-3): Internal error xlog_clear_stale_blocks(2) at line 1353 of file > fs/xfs/xfs_log_recover.c. Caller 0xffffffffa01f894d The head and tail of the log are confused - different cycle numbers but the tail is behind the head. That implies that there are 3 cycle numbers visible in the log, when here should only be 2, which would mean that some log write did not make it to disk correctly. You'll need to provide a copy of the log (xfs_logprint can get that for you) and the information about your system described here: http://xfs.org/index.php/XFS_FAQ#Q:_What_information_should_I_include_when_reporting_a_problem.3F Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs
