[Cluster-devel] [PATCH] GFS2: force a log flush when invalidating the rindex glock

[Steven Whitehouse]

Hi,

On Mon, 2011-06-13 at 14:27 -0500, Benjamin Marzinski wrote:
> Right now, there is nothing that forces the log to get flushed when a node
> drops its rindex glock so that another node can grow the filesystem. If the
> log doesn't get flushed, GFS2 can corrupt the sd_log_le_rg list in the
> following way.
> 
> A node puts an rgd on the list in rg_lo_add(), and then the rindex glock is
> dropped so the other node can grow the filesystem. When the node reacquires the
> rindex glock, that rgd gets deleted in clear_rgrpdi() before ever being
> removed from the list by gfs2_log_flush().
> 
> This code simply forces a log flush when the rindex glock is invalidated,
> solving the problem.
> 
> Signed-off-by: Benjamin Marzinski <bmarzins at redhat.com>
> ---
>  fs/gfs2/glops.c |    4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> Index: gfs2-2.6-nmw/fs/gfs2/glops.c
> ===================================================================
> --- gfs2-2.6-nmw.orig/fs/gfs2/glops.c
> +++ gfs2-2.6-nmw/fs/gfs2/glops.c
> @@ -221,8 +221,10 @@ static void inode_go_inval(struct gfs2_g
>  		}
>  	}
>  
> -	if (ip == GFS2_I(gl->gl_sbd->sd_rindex))
> +	if (ip == GFS2_I(gl->gl_sbd->sd_rindex)) {
> +		gfs2_log_flush(gl->gl_sbd, NULL);
>  		gl->gl_sbd->sd_rindex_uptodate = 0;
> +	}
>  	if (ip && S_ISREG(ip->i_inode.i_mode))
>  		truncate_inode_pages(ip->i_inode.i_mapping, 0);
>  }
> 

Now in the -nmw git tree. Thanks,

Steve.