On Mon, Nov 28, 2011 at 7:43 PM, Emmanuel Dreyfus <manu@xxxxxxxxxx> wrote: > On Mon, Nov 28, 2011 at 07:15:49PM +0530, Anand Avati wrote: >> You need not worry as long as the numbers are stable. > > I am not sure they are stable when bricks are added: it cause a ESTALL, > which is recovered on the next attempt, but still, I suspect there is > something wrong. Are brick addition supposed to run without any hitch > for clients? Yes, that was the very intention of making the inode number a derivation of the GFID instead of the backend inode number, so that brick addition/replacement does not impact its stability. Avati
