Just wanted to ask what level of IO is required to hit this bug. We
received it during a backup using RSYNC from a set of FC volumes, to one
FC volume >1TB in size.
The backup volume was the one shutdown. I may finally have people
convinced that we need to update the XFS code, but I wanted to get a
feel for how often we might see this. TIA
<error>
Nov 27 14:24:53 Myserver kernel: xfs_force_shutdown(sd(65,32),0x8)
called from line 4065 of file xfs_bmap.c. Return address = 0xc019d00d
Nov 27 14:24:53 Myserver kernel: Corruption of in-memory data detected.
Shutting down filesystem: sd(65,32)
Nov 27 14:24:53 Myserver kernel: Please umount the filesystem, and
rectify the problem(s)
</error>
--
Austin Gonyou <austin@xxxxxxxxxxxxxxx>
Coremetrics, Inc.
|