Hi all,
I just did an update of a storage cluster here, or rather, I've done one
node out of three updating to Infernalis from Hammer.
I shut down the daemons, as per the guide, then did a recursive chown of
the /var/lib/ceph directory, then struck the following when re-starting:
> 2016-01-05 07:32:09.114197 7f5b41d0f940 0 ceph version 9.2.0
> (bb2ecea240f3a1d525bcb35670cb07bd1f0ca299), process ceph-osd, pid 2899
> 2016-01-05 07:32:09.123740 7f5b41d0f940 0
> filestore(/var/lib/ceph/osd/ceph-0) backend xfs (magic 0x58465342)
> 2016-01-05 07:32:09.124047 7f5b41d0f940 0
> genericfilestorebackend(/var/lib/ceph/osd/ceph-0) detect_features: FIEMAP
> ioctl is disabl
> ed via 'filestore fiemap' config option
> 2016-01-05 07:32:09.124053 7f5b41d0f940 0
> genericfilestorebackend(/var/lib/ceph/osd/ceph-0) detect_features:
> SEEK_DATA/SEEK_HOLE is
> disabled via 'filestore seek data hole' config option
> 2016-01-05 07:32:09.124066 7f5b41d0f940 0
> genericfilestorebackend(/var/lib/ceph/osd/ceph-0) detect_features: splice is
> supported
> 2016-01-05 07:32:09.156182 7f5b41d0f940 0
> genericfilestorebackend(/var/lib/ceph/osd/ceph-0) detect_features: syncfs(2)
> syscall full
> y supported (by glibc and kernel)
> 2016-01-05 07:32:09.156301 7f5b41d0f940 0
> xfsfilestorebackend(/var/lib/ceph/osd/ceph-0) detect_features: extsize is
> supported and y
> our kernel >= 3.5
> 2016-01-05 07:32:09.232801 7f5b41d0f940 0
> filestore(/var/lib/ceph/osd/ceph-0) mount: enabling WRITEAHEAD journal mode:
> checkpoint i
> s not enabled
> 2016-01-05 07:32:09.253440 7f5b41d0f940 -1
> filestore(/var/lib/ceph/osd/ceph-0) mount failed to open journal /dev/sdc5:
> (13) Permissi
> on denied
> 2016-01-05 07:32:09.263646 7f5b41d0f940 -1 osd.0 0 OSD:init: unable to mount
> object store
> 2016-01-05 07:32:09.263656 7f5b41d0f940 -1 ESC[0;31m ** ERROR: osd init
> failed: (13) Permission deniedESC[0m
Things did not co-operate until I chown'ed /dev/sdc5 (and /dev/sdc6) to
ceph:ceph. (-R in /var/lib/ceph was not sufficient). Even adding ceph
to the 'disk' group (who owns /dev/sdc5) oddly enough, was not sufficient.
I have that node running, and will do the others, but I am concerned
about what happens after a reboot. Is it necessary to configure udev to
chown /dev/sdc[56] at boot or is there some way to fix ceph's permissions?
--
_ ___ Stuart Longland - Systems Engineer
\ /|_) | T: +61 7 3535 9619
\/ | \ | 38b Douglas Street F: +61 7 3535 9699
SYSTEMS Milton QLD 4064 http://www.vrt.com.au
_______________________________________________
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
