At 01:07 AM 3/17/2009, Stas Oskin wrote: >Hi. > >As far as I understand, in case one of the servers had downtime and >comes back, GlusterFS automatically synchronizes the files on it >when they are accessed. > >Would such synchronization increase the time the user waits for the >file? Meaning the file would be first synchronized, and only then >read to user - or it's read immediately and synchronization is done >in background? I believe how it works, from my observations: IF the correct version of the file exists on the server from which you request the file, then it's served directly and synced in the background. IF the correct version does NOT exist on the server from which you request the file, then it is synced first then served. >Regards. >_______________________________________________ >Gluster-users mailing list >Gluster-users at gluster.org >http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users