find the answers inline. — Bishoy
> On Dec 4, 2015, at 6:54 AM, Mountrakis, Michael <[email protected]> > wrote: > > Hi all > > The scenario that I am thinking to implement has as follows: > > Mount a volume locally to my Node1 as i-scsi: > mybox1# iscsiadm -m node --targetname "rot1:newvolume.1" --portal > "192.168.0.1:3260" --login > .. format and make file system... > mybox1# mount /dev/sdb1 /data -t ext4 > > Mount another volume locally to my Node2 as i-scsi: > mybox2# iscsiadm -m node --targetname "rot2:newvolume.1" --portal > "192.164.0.1:3260" --login > .. format and make file system... > mybox2# mount /dev/sdb1 /data -t ext4 > > I do the peering between node1/node2: > root@ mybox1:~ # gluster peer probe mybox2.mydomain > > Try to create a glusterFS volume from those mounts: > Will the following work? > root@ mybox1:~ # gluster volume create mydata replica 2 mybox1.mydomain:/data > mybox2.mydomain:/data force > > In other words: > Can I create and share a glusterFS volume from a couple of mounted iSCSI > bricks? yes you can, but the problem you might have is that when the iSCSI sessions get broken then reestablished they will be mapped with a different name (if it was /dev/sdb to might get mapped to /dev/sde or any other letter) it’s better to use ZFS with iSCSI as ZFS handles Linux device mapper mapping of the drives better than XFS. > > > Best Regards > mike > _______________________________________________ > Gluster-users mailing list > [email protected] > http://www.gluster.org/mailman/listinfo/gluster-users
_______________________________________________ Gluster-users mailing list [email protected] http://www.gluster.org/mailman/listinfo/gluster-users
