Hi avati, The fixed-id xlator works as designed. What I was looking to do was the opposite behavior, where files get their uids change on their way to the lower volume, rather than as they come out. I was able to do this with a lower-level storage translator thanks to Angel's suggestion. The more time I spend in the code, the more I appreciate the design. Keep up the good work! :august On Jan 20, 2008 7:17 PM, Anand Avati <avati@xxxxxxxxxxxxx> wrote: > August, > > > 2008/1/20, August R. Wohlt <glusterfs@xxxxxxxxxxx>: > > > > Hi glusterfs hackers, > > > > I am trying to write a simple xlator similar to the fixed-id one that > > comes > > with the source code. > > > > The fixed-id xlator passes all calls through to the underlying volume > > and > > then mangles the uid and gid of the stat structures on the way back to > > the > > client, so what you end up with when looking at a mount is a directory > > of > > files that appear to be owned by a fixed uid.gid, but really on the > > underlying volume, they are created as the original user, > > > This is specifically how we want it. By removing fixed-id from your volume > graph at a later point, the filesystem continues as if fixed-id was never > loaded in history. > > which can cause > > interesting permission issues. > > > I'm interested to know what the exact issue is. I would prefer to have it > fixed in such a way that the filesystem stores the file owned by the real > uid/gid rather than the masked value. > > thanks, > avati >
