On Sun, Aug 19, 2012 at 6:18 AM, Emmanuel Dreyfus <manu@xxxxxxxxxx> wrote:
On Sun, Aug 19, 2012 at 05:52:56AM -0700, Anand Avati wrote:Yes, I went from 1 brick to 2 bricks. I will try 2->, but is 1->2 known
> Oh BTW, did you add-brick from 1 brick to 2 bricks? If so it might be
> understandable as with 1 brick DHT is not loaded and there will be no hash
> ranges pre-assigned. If so can you try by starting with (volume create)
> with 2 bricks and then add-brick a 3rd one?
to have issues?
I now realize that if you re-refer an inode before it is timed out in the kernel (and LOOKUP is sent), then the code path hitting fuse_resolve_gfid_cbk might reach this. We need to call dht_self_heal in dht_discover_cbk if layout requires healing.
Avati
