On 02/14/2013 01:35 PM, Gregory Farnum wrote: > On Thu, Feb 14, 2013 at 11:32 AM, Dimitri Maziuk <dmaziuk@xxxxxxxxxxxxx> wrote: >> >> # cephfs /ceph/blastdb set_layout --osd 3 >> Error setting layout: Invalid argument > Ooo, you haven't been able to specify an OSD in a very long time > indeed. So I guess the manpage is broken. Thanks. Next question: shouldn't -233.016% be some other unit than '%' -- or is there an "if(x < y) // calculate a little differently" kind of thing missing someplace? 2013-02-14 13:40:56.831105 7fc69a9f0700 0 log [INF] : pgmap v71793: 576 pgs: 413 active+clean, 150 active+recovery_wait, 1 incomplete, 3 remapped+incomplete, 9 active+recovering; 3411 MB data, 16611 MB used, 3465 GB / 3667 GB avail; 0B/s wr, 3op/s; -8808/3780 degraded (-233.016%); recovering 7 o/s, 29105KB/s 2013-02-14 13:40:57.876709 7fc69a9f0700 0 log [INF] : pgmap v71794: 576 pgs: 413 active+clean, 150 active+recovery_wait, 1 incomplete, 3 remapped+incomplete, 9 active+recovering; 3375 MB data, 16503 MB used, 3465 GB / 3667 GB avail; 0B/s wr, 8op/s; -8846/3744 degraded (-236.271%); recovering 11 o/s, 45599KB/s 2013-02-14 13:40:58.942340 7fc69a9f0700 0 log [INF] : pgmap v71795: 576 pgs: 413 active+clean, 150 active+recovery_wait, 1 incomplete, 3 remapped+incomplete, 9 active+recovering; 3355 MB data, 16467 MB used, 3465 GB / 3667 GB avail; 0B/s wr, 4op/s; -8846/3724 degraded (-237.540%) This is on libcephfs1-0.56.3-0.el6.x86_64, ceph-0.56.3-0.el6.x86_64 -- I changed the CRUSH map and then deleted everything from the directory without waiting for the cluster to stabilize. -- Dimitri Maziuk Programmer/sysadmin BioMagResBank, UW-Madison -- http://www.bmrb.wisc.edu
Attachment:
signature.asc
Description: OpenPGP digital signature
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com