Which is a different OSD. Try running ceph-volume lvm activate --all Paul -- Paul Emmerich Looking for help with your Ceph cluster? Contact us at https://croit.io croit GmbH Freseniusstr. 31h 81247 München www.croit.io Tel: +49 89 1896585 90 On Mon, Nov 4, 2019 at 7:44 AM Anton Aleksandrov <anton@xxxxxxxxxxxxxx> wrote: > > I think, that yes, LVM is online - both drives preset and LVM operates. > > lvs -o lv_tags returns: > > ceph.block_device=/dev/osd_vg/osd_lv,ceph.block_uuid=EyUggo-Ja06-MptY-Rt4Q-l6iV-0pse-7Y8rPh,ceph.cephx_lockbox_secret=,ceph.cluster_fsid=31c48e79-7724-49db-a20f-dd86e195972f,ceph.cluster_name=ceph,ceph.crush_device_class=None,ceph.encrypted=0,ceph.osd_fsid=a0aa881c-aa2d-4462-9c2f-cd289810e9e7,ceph.osd_id=23,ceph.type=block,ceph.vdo=0 > > We deployed everything using ceph-deloy tool, pretty automatic and > without doing anything extra special. > > Reason for this setup on this server is because all other OSDs have > 1x8tb, but this host has 2x4tb and just 8gb of ram, therefore we did not > want to make two separate OSDs. > > Anton > > On 04.11.2019 00:49, Paul Emmerich wrote: > > On Sun, Nov 3, 2019 at 6:25 PM Anton Aleksandrov <anton@xxxxxxxxxxxxxx> wrote: > >> Hello community. > >> > >> We run Ceph on quite old hardware with quite low traffic. Yesterday we > >> had to reboot one of the OSDs and after reboot it did not came up. The > >> error message is: > >> > >> [2019-11-02 15:05:07,317][ceph_volume.process][INFO ] Running command: > >> /usr/sbin/ceph-volume lvm trigger 22-1c0b3fd7-7d80-4de9-9594-17ac5b2bf92f > >> [2019-11-02 15:05:07,473][ceph_volume.process][INFO ] stderr --> > >> RuntimeError: could not find osd.22 with fsid > >> 1c0b3fd7-7d80-4de9-9594-17ac5b2bf92f > >> > >> This OSD has 2 disks, which are put into one logical volume (basically > >> raid0) and then used for osd storage. > > please don't do that (unless you have very good reason to) > > > >> We are quite a beginners with Ceph and this error stuck us. What should > >> we do? Change fsid (where?)? Right now cluster is in repair-state.. As > >> the last resort we would drop osd and rebuild it, but it would be very > >> important for us to understand - what and why happenned. Is it faulty > >> config or did something bad happen with the disks? > > Is the LV online? Do the LV tags look correct? Check lvs -o lv_tags > > > > > > > > Paul > > > > > >> Regards, > >> Anton. > >> > >> _______________________________________________ > >> ceph-users mailing list > >> ceph-users@xxxxxxxxxxxxxx > >> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |