Hide Forgot
Pasting the correct output of the bug: cp /tmp/file /mnt/client/ c9ecca0f5000035fa1b06c07cbb765ba file 10932188053aae6e8902244970e126bc /mnt/client1/file
PATCH: http://patches.gluster.com/patch/2781 in master (server_writev: add proper iobuf into iobref)
PATCH: http://patches.gluster.com/patch/2781 in release-3.0 (server_writev: add proper iobuf into iobref)
When write-behind was used on the server side, there was data corruption. cp /tmp/file /mnt/client/ c9ecca0f5000035fa1b06c07cbb765ba /tmp/file c9ecca0f5000035fa1b06c07cbb765ba /mnt/client/file Server Volume file: volume posix1 type storage/posix option directory /mnt/export end-volume volume locks type features/posix-locksn subvolumes posix1 end-volume volume iot type performance/io-threads option thread-count 8 subvolumes locks end-volume volume writebehind type performance/write-behind option cache-size 4MB subvolumes iot end-volume volume brick1 type performance/read-ahead option page-count 4 option force-atime-update off subvolumes writebehind end-volume volume server type protocol/server option transport-type tcp option transport.socket.listen-port 7777 subvolumes brick1 option auth.addr.brick1.allow * end-volume Client vol file: volume client1 type protocol/client option transport-type tcp option remote-host 127.0.0.1 option transport.socket.remote-port 7777 option remote-subvolume brick1 end-volume volume writebehind type performance/write-behind option cache-size 4MB option flush-behind on subvolumes client1 end-volume