On Mon, 2 Feb 2015, Stephen Hindle wrote: > Hi! > > Just thought I'd mention a minor inconsistency I noticed while > re-installing our cluster... > preparing an OSD with ceph-deploy will NOT activate it on a parttion > (node1:/sda1/sda2) > preparting an OSD with ceph-deploy WILL activate it on a disk ( > node1:/dev/sdc) > > I think I noticed similiar behavior with ceph-deploy osd create, but I > assumed I just messed up, > so I didnt document it. This is probably more of an 'issue' as create is > supposed to activate it for you. > > Manually issuing a ceph-deploy osd activate works fine to bring the > partition OSDs up.
This is an artifact of how activation works: we label GPT partitions so that udev can launch ceph-osd. If you use existing partitions, they aren't GPT (probably) and aren't labeled accordingly. I think the "fix" is to make sure the existing partitions are GPT partitions, and make sure that ceph-disk will relabel is possible (I forget if it does this now...) sage -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
