From: Anthony D'Atri Date: Wed, 26 Feb 2025 20:37:59 +0000 (-0500) Subject: src/common/options: Clarify mon_target_pg_per_osd in mgr.yaml.in X-Git-Url: http://git.apps.os.sepia.ceph.com/?a=commitdiff_plain;h=fecbb3aa07ba4183d927adb4239874c8dddb9f5c;p=ceph.git src/common/options: Clarify mon_target_pg_per_osd in mgr.yaml.in Signed-off-by: Anthony D'Atri --- diff --git a/src/common/options/mgr.yaml.in b/src/common/options/mgr.yaml.in index 5095710afdff9..931f5bf496059 100644 --- a/src/common/options/mgr.yaml.in +++ b/src/common/options/mgr.yaml.in @@ -311,11 +311,13 @@ options: - name: mon_target_pg_per_osd type: uint level: advanced - desc: Automated PG management creates this many PGs per OSD - long_desc: When creating pools, the automated PG management logic will attempt to - reach this target. In some circumstances, it may exceed this target, up to the + desc: Target number of PG replicas per OSD + long_desc: The placement group (PG) autoscaler will arrange for approximately this number of PG + replicas on each OSD as shown by the PGS column of a ``ceph osd df`` + report. In some circumstances it may exceed this target, up to the ``mon_max_pg_per_osd`` limit. Conversely, a lower number of PGs per OSD may be - created if the cluster is not yet fully utilised + set if the cluster is not yet fully utilized or when the sum of power-of-two + per-pool ``pg_num`` values does not permit a perfect fit. default: 100 min: 1 # min pgs per osd for reweight-by-pg command