Looks like the orchestation side support for this got brought into pacific with the rest of the drive group stuff, but the actual underlying feature in ceph-volume (from https://github.com/ceph/ceph/pull/40659) never got a pacific backport. I've opened the backport now https://github.com/ceph/ceph/pull/53581 and I think another pacific release is planned so we can hopefully have it fixed there eventually, but it's definitely broken as of now. Sorry about that. On Thu, Sep 21, 2023 at 7:54 AM Boris Behrens <bb@xxxxxxxxx> wrote: > I have a use case where I want to only use a small portion of the disk for > the OSD and the documentation states that I can use > data_allocation_fraction [1] > > But cephadm can not use this and throws this error: > /usr/bin/podman: stderr ceph-volume lvm batch: error: unrecognized > arguments: --data-allocate-fraction 0.1 > > So, what I actually want to achieve: > Split up a single SSD into: > 3-5x block.db for spinning disks (5x 320GB or 3x 500GB regarding if I have > 8TB HDDs or 16TB HDDs) > 1x SSD OSD (100G) for RGW index / meta pools > 1x SSD OSD (100G) for RGW gc pool because of this bug [2] > > My service definition looks like this: > > service_type: osd > service_id: hdd-8tb > placement: > host_pattern: '*' > crush_device_class: hdd > spec: > data_devices: > rotational: 1 > size: ':9T' > db_devices: > rotational: 0 > limit: 5 > size: '1T:2T' > encrypted: true > block_db_size: 320000000000 > --- > service_type: osd > service_id: hdd-16tb > placement: > host_pattern: '*' > crush_device_class: hdd > spec: > data_devices: > rotational: 1 > size: '14T:' > db_devices: > rotational: 0 > limit: 1 > size: '1T:2T' > encrypted: true > block_db_size: 500000000000 > --- > service_type: osd > service_id: gc > placement: > host_pattern: '*' > crush_device_class: gc > spec: > data_devices: > rotational: 0 > size: '1T:2T' > encrypted: true > data_allocate_fraction: 0.05 > --- > service_type: osd > service_id: ssd > placement: > host_pattern: '*' > crush_device_class: ssd > spec: > data_devices: > rotational: 0 > size: '1T:2T' > encrypted: true > data_allocate_fraction: 0.05 > > > [1] > > https://docs.ceph.com/en/pacific/cephadm/services/osd/#ceph.deployment.drive_group.DriveGroupSpec.data_allocate_fraction > [2] https://tracker.ceph.com/issues/53585 > > -- > Die Selbsthilfegruppe "UTF-8-Probleme" trifft sich diesmal abweichend im > groüen Saal. > _______________________________________________ > ceph-users mailing list -- ceph-users@xxxxxxx > To unsubscribe send an email to ceph-users-leave@xxxxxxx > _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |