On Wed, 31 Mar 2004, Jan Harkes wrote: > > My question is - why would Venus ever loose/flush/invalidate the data in > > the cache, while it definitely was "produced" locally and still valid? > > Just a guess, the last operation on that file was a reintegration?
Yes, it probably was. > probable something in the client that is not correctly updating the > local version vector, which results in a forced refetch of a file that > we were the last writer for. > At the moment this is nothing more than a theory though. It would explain the behaviour. Regards, -- Ivan
