Hi List; I am attempting to get a test ceph cluster up and running. I am using ceph-0.50 across all nodes. I am attempting to get the number of pgs to be around 100 per osd as per the documentation at:
http://ceph.com/docs/master/dev/placement-group/ I have attempted to increase the pgs via several methods: 1) Increase at mkcephfs time by passing --osd-pg-bits 7. This does not work as this option does not exist on 0.50 re: http://www.mail-archive.com/[email protected]/msg06923.html 2) Increase the number of pgs via ceph.conf (osd pg bits = 7) and create the cluster. This does not work either as the cluster comes up with 6 pgs bits per osd still. re: http://ceph.com/docs/master/config-cluster/osd-config-ref/ So I'm a bit lost! I cannot find a way to increase pgs at pool creation time. Option 1 does not work at all. Option 2 seems to be ignored. I have 68 OSD's. At 7 bits per OSD I should have 8704 (68 * 2^7) pgs, but I still end up with the default of 4352 (68 * 2^6). I'm hoping I'm missing something simple, but if anyone is able to point me in the correct direction, that would be appreciated! Regards, Tren -- 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
