On Thu, Feb 01, 2007 at 11:24:09AM +0100, Attila Nagy wrote:
Hello,
I have a setup, where I have a /data filesystem with a "big" (20G)
sparse file on it. This sparse file is set up with:
mdconfig -a -t vnode -f /data/nfsimg
and I use /dev/md0 for storing some data, shared via NFS (I needed
redundant read only NFS servers, so I just sync this image to the others
and mount read only on them).
With the above, I get the following in the syslog:
Feb 1 11:16:49 boot00a kernel:
g_vfs_done():md0[WRITE(offset=1350844416, length=16384)]error = 1
Feb 1 11:16:49 boot00a kernel:
g_vfs_done():md0[WRITE(offset=1928118272, length=16384)]error = 1
Feb 1 11:17:18 boot00a kernel:
g_vfs_done():md0[WRITE(offset=1350844416, length=16384)]error = 1
Feb 1 11:17:18 boot00a kernel:
g_vfs_done():md0[WRITE(offset=1928118272, length=16384)]error = 1
And these two lines are alternating, with the same offsets, as you can
see. I use the raw device for storing the filesystem (newfs /dev/md0).
Currently the image is mounted read-write and there is almost no
activity on that.
Any ideas about this?
errno == 1 == EPERM (Operation not permitted). Check for this.