In Ceph clusters prior to Firefly (0.80), the cluster does not reach a 'HEALTH_OK' status even though the OSDs are marked down/out and recovery has stopped, why?
Issue
- In clusters running Ceph versions prior to Firefly (0.80), a situation can occur where the cluster will cease all recovery, but the cluster will not yet be at the HEALTH_OK status with all PGs active+clean.
Environment
-
Inktank Ceph Enterprise
-
Ceph versions prior to Firefly ( 0.80 )
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.