On 02/06/2012 10:12 AM, Whit Blauvelt wrote: > Thanks Pranith. I take it then that if there is corruption of a file due to, > say, a degraded disk on one side of a replica pair that gluster has no way > to spot and heal this? There's nothing built in (such as a stored CRC value) > by which it could conclude "Brick A's file is corrupted, so I'll restore it > from Brick B"? That's correct for the time being. The way we know that a copy of a file needs self-heal is by looking at the flags on *other* copies of the file, so if something happens to the file itself that's effectively invisible. We've built some of the infrastructure that will eventually be needed to support storing CRCs along with data, because it's similar to infrastructure we need for other things, but the feature itself is likely to remain a long way off.
