On Mar 10, 2010, at 10:26 AM, Ian Rogers wrote: > > As far as I can tell from the docs, gluster has a very naive algorithm for picking which brick to read from and write to. > > For reading it scans the "subvolumes" entry left to right, finding which brick has the file with the most recent create time the largest "modification count". It then uses the left most one. > > For writing - either user initiated or to self-heal out of date files - it just writes to all subvolumes that are available. Self-heal is not that simplistic. Because replicate writes a "changelog" before and after every write (any kind of modification), it can derive the node that is most up-to-date during self-heal. I understand that many people have questions about how self-heal works, and I'm in the process of documenting the internals of replication. I'll post it here in a few days. Hopefully it will answer all your questions. ------------------------------ Vikas Gorur Engineer - Gluster, Inc. +1 (408) 770 1894 ------------------------------
