[Gluster-users] Recovering out of sync nodes from input/output error

[Jeff Darcy]

On Fri, 13 Apr 2012 17:02:41 +0300
Alex Florescu <alex.florescu at tripsolutions.co.uk> wrote:

> On Fri, Apr 13, 2012 at 4:32 PM, Jeff Darcy wrote:
> > Try changing "trusted.gluster.dht" to "trusted.glusterfs.dht" as had
> > originally been suggested.  Alternatively, you could just use "-m ."
> > instead of "-n trusted.gluster.dht" to dump *all* xattrs, and see
> > what we get.
> 
> No .dht attribute, only afr as you can see.

Oops, yeah, this time it's an AFR problem, not DHT.  I really need to
cut back on the magic mushrooms.

> > The reason you continue to get I/O errors is probably that the
> > xattrs on the *parent directory* still indicate pending operations
> > on both sides.
> 
> Indeed, this is as you said.

I'm still looking into what needs to change so that at least these
xattrs will get cleared, even if we can't (or shouldn't) do the actual
self-heal ourselves.