i get this message and subsequent filesystem shutdown using
CVS from 20021031-09:47 (CET):
Nov 4 15:07:00 embeh kernel: xfs_force_shutdown(lvm(58,3),0x8) called from
line 1042 of file xfs_trans.c. Return address = 0xc01e2c17
Nov 4 15:07:00 embeh kernel: Filesystem "lvm(58,3)": Corruption of
in-memory data detected. Shutting down filesystem: lvm(58,3)
The problem with the two stacked layer (LVM+RAID5) seems to have gone
away (see bug 182). I tend to exclude a memory problem here given that the
machine ran fine (with XFS 1.1) before trying the new kernel.
After the fs shutdown, the machine was rebooted and the bare recovery made
the trick. The subsequent xfs_check showed no problems (so the xfs_repair
-n). The same problem happened 5 hours later to the same filesystem (out
of 10 this machine exports).
Currently the machine it's stick to XFS 1.1 since the last shutdown
(currently 16 hours without a single hitch).