Hm, it seems it needs to set your volume right. With samba4 I had issues and set the volume like this: performance.stat-prefetch: off
Volume Name: sambacluster Type: Replicate Volume ID: 4fd0da03-8579-47cc-926b-d7577dac56cf Status: Started Number of Bricks: 1 x 2 = 2 Transport-type: tcp Bricks: Brick1: s4master:/raid5hs/glusterfs/samba Brick2: s4slave:/raid5hs/glusterfs/samba Options Reconfigured: performance.quick-read: on network.ping-timeout: 5 performance.stat-prefetch: off It worked for me EDV Daniel Müller Leitung EDV Tropenklinik Paul-Lechler-Krankenhaus Paul-Lechler-Str. 24 72076 Tübingen Tel.: 07071/206-463, Fax: 07071/206-499 eMail: [email protected] Internet: www.tropenklinik.de -----Ursprüngliche Nachricht----- Von: [email protected] [mailto:[email protected]] Gesendet: Dienstag, 19. August 2014 16:30 An: [email protected] Cc: [email protected] Betreff: Re: AW: [Gluster-users] mysql on gluster storage > I know this issue if you are running proxmox for virt. machine and you > store it on gluster then you need to set "write through" in you > settings of your hdd (vmdk or else). Thanks, Daniel. I have that, and I believe it is the default setting for KVM/QEMU (other than in Proxmox, where it is "none"). I.e. the errors occur despite "writethrough". Cheers, Chris _______________________________________________ Gluster-users mailing list [email protected] http://supercolony.gluster.org/mailman/listinfo/gluster-users
