Closed djgalloway closed 5 years ago
# ceph df RAW STORAGE: CLASS SIZE AVAIL USED RAW USED %RAW USED hdd 427 TiB 194 TiB 233 TiB 234 TiB 54.68 ssd 2.1 TiB 1.0 TiB 1.1 TiB 1.1 TiB 51.63 TOTAL 430 TiB 195 TiB 234 TiB 235 TiB 54.67 POOLS: POOL ID STORED OBJECTS USED %USED MAX AVAIL data 0 75 TiB 50.38M 75 TiB 45.87 30 TiB metadata 1 56 GiB 2.47M 56 GiB 0.06 22 TiB libvirt-pool 4 2.0 GiB 518 2.0 GiB 0 30 TiB djf_tmp 92 1.4 TiB 8.24M 1.4 TiB 1.60 30 TiB .rgw.root 93 1.1 KiB 4 1.1 KiB 0 30 TiB default.rgw.control 94 0 B 8 0 B 0 30 TiB default.rgw.meta 95 2.4 KiB 8 2.4 KiB 0 30 TiB default.rgw.log 96 8.0 MiB 209 8.0 MiB 0 30 TiB default.rgw.buckets.index 97 38 MiB 2 38 MiB 0 30 TiB default.rgw.buckets.data 98 744 GiB 242.06k 744 GiB 0.81 30 TiB default.rgw.buckets.non-ec 99 0 B 0 0 B 0 30 TiB device_health_metrics 100 52 MiB 146 52 MiB 0 30 TiB # ceph --version ceph version 14.2.1-198-g869a6a3 (869a6a3e1140d44523ad1e10239a9c874cce0885) nautilus (stable)
I'm not sure how far back it's supported, but it seems to me the json output of the ceph commands (ceph df -f json in this case) would be more robust than parsing the human-readable table outputs.
json
ceph
ceph df -f json