Ceph - ceph-mgr dashboard does not match 'ceph df' output
Environment
- Red Hat Ceph Storage 3.x
Issue
-
There is a difference in "Usage" for the ceph-mgr dashboard when compared to the "Used" & "MAX AVAIL" columns from the 'ceph df' output.
-
Information From ceph-mgr Dashboard:
Pools
Name PG status Usage Activity
volumes 128 active+clean 0 / 32.1G 0 rd, 0 wr
images 128 active+clean 0 / 32.1G 0 rd, 0 wr
backups 128 active+clean 0 / 32.1G 0 rd, 0 wr
vms 128 active+clean 0 / 32.1G 0 rd, 0 wr
rbd 32 active+clean 2.76k / 32.1G 204 rd, 0 wr
- From 'ceph df' output:
user@hostname ~]# ceph df
GLOBAL:
SIZE AVAIL RAW USED %RAW USED
98705M 97489M 1216M 1.23
POOLS:
NAME ID USED %USED MAX AVAIL OBJECTS
volumes 19 0 0 30673M 0
images 20 0 0 30673M 0
backups 21 0 0 30673M 0
vms 22 0 0 30673M 0
rbd 24 2760 0 30673M 12
Resolution
- This is currently being tracked by Red Hat downstream BZ 1611669.
- If you hitting this bug and need more information please contact Red Hat support.
Root Cause
- The difference in the 'ceph df' output and the ceph-mgr dashboard is a difference in rounding & calculation method used.
- In summary, 'ceph df' uses binary math, where as the dashboard is using decimal math. The more accurate of the two is the 'ceph df' output because of the binary math.
- The larger the cluster size, the more inflated this difference will be.
This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.
Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.
