Hi, Maybe it's just a typo but your nufa volume refers to 'replicate3', which doesn't seem to exist... Anyway, I was wondering if it was possible to use nufa over replicated volumes. It seems it is. I might set up such a cluster soon. Can you tell me about the performances you get on it ? Regards, wenaideyu wenaideyu a ?crit : > hi > > I took 4 servers for glusterfs, sev1 and sev 2 were used for > replicate1, sev3 and sev4 were usered for > replicate2, replicate1 and replicate2 were combined for nufa, there is > server.vol in every server: > > server.vol > > volume posix > type storage/posix > option directory /mnt/data1 > end-volume > > volume locks > type features/locks > subvolumes posix > end-volume > > volume brick > type performance/io-threads > option thread-count 8 > subvolumes locks > end-volume > > volume server > type protocol/server > option transport-type tcp > option auth.addr.brick.allow * > subvolumes brick > end-volume > > > client.vol in sev1 > volume sev1 > type protocol/client > option transport-type tcp > option remote-host sev1 > option remote-subvolume brick > end-volume > > volume sev2 > type protocol/client > option transport-type tcp > option remote-host sev2 > option remote-subvolume brick > end-volume > > volume sev3 > type protocol/client > option transport-type tcp > option remote-host sev3 > option remote-subvolume brick > end-volume > > volume sev4 > type protocol/client > option transport-type tcp > option remote-host sev4 > option remote-subvolume brick > end-volume > > volume replicate1 > type cluster/replicate > subvolumes sev1 sev2 > end-volume > > volume replicate2 > type cluster/replicate > subvolumes sev3 sev4 > end-volume > > volume nufa > type cluster/nufa > option local-volume-name replicate1 > subvolumes replicate1 replicate3 > end-volume > > volume writebehind > type performance/write-behind > option page-size 128KB > option cache-size 1MB > subvolumes nufa > end-volume > > volume cache > type performance/io-cache > option cache-size 512MB > subvolumes writebehind > end-volume > > every server has a disk mounted on /mnt/data1, the disk has a capability > with 40GB, > so the glusterfs's total capbility is 80GB, > > client is in sev1, and mount the glusterfs on /mnt/data > > now the problem is that: when the replicate1 is nearly fullfilled, for > expample when the > replicate1 is 50MB free, i copy a file A bigger than 50MB into > /mnt/data, file A will be > created on replicate1, while the replicate1 is fullfiled, the rest of > file A can not be > write on replicate2,and the log in sev1 or sev2 is that: > > [posix.c:1736:posix_writev] posix: writev failed: No space left on device > > anybody can help me??? thanks very much > > > ------------------------------------------------------------------------ > > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users
