http://oss.sgi.com/bugzilla/show_bug.cgi?id=186
c.pascoe@xxxxxxxxxxxxxx changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|xfs_force_shutdown on a lvm |xfs_force_shutdown using NFS
|device |
------- Additional Comments From c.pascoe@xxxxxxxxxxxxxx 2002-10-27 23:07
-------
Done some more investigation - the system is shutting down the filesystem as
xfs_iget is returning EIO (there's only one place it does this) to
xfs_trans_iget, which then propagates back to xfs_ialloc and xfs_dir_alloc.
Then, once we're back in xfs_create, xfs_mkdir or xfs_symlink, we then jump to
the abort_return tag, and the filesystem gets shut down by the subsequent
xfs_trans_cancel. I have seen all three of the above functions in kdb
backtraces when this has occurred.
Hope this helps...
Regards,
Chris
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
|