nova/releasenotes/notes/rbd-enhance-get-pool-info-14afc8eccab49dcf.yaml
melanie witt 44254ca865 rbd: use MAX_AVAIL stat for reporting bytes available
Currently, the reporting of bytes available works well for recommended
Ceph deployments that run one OSD per disk [1]. However, for users who
are running multiple OSDs on a single disk, the current reporting will
reflect bytes available * number of replicas.

We can enhance the bytes available reporting method to accomodate
unrecommended Ceph deployments by using the MAX_AVAIL stat obtainable
via the 'ceph df' command. The MAX_AVAIL stat takes the number of
configured replicas into consideration and will reflect the correct
number of bytes available even when Ceph is deployed in a way the
documentation recommends against.

For most users, this change should make no difference. It will only be
a help for users who are running unrecommended Ceph deployments.

[1] http://docs.ceph.com/docs/luminous/start/hardware-recommendations/#hard-disk-drives

Change-Id: I96faff6d3b9747514441d83c629fdd1cface1eb5
2019-06-19 20:34:36 +00:00

10 lines
401 B
YAML

---
other:
- |
The reporting for bytes available for RBD has been enhanced to accomodate
`unrecommended
<http://docs.ceph.com/docs/luminous/start/hardware-recommendations/#hard-disk-drives>`_
Ceph deployments where multiple OSDs are running on a single disk. The new
reporting method takes the number of configured replicas into consideration
when reporting bytes available.