How to configure Ceph to permenantly mark flapping OSDs out of the cluster?
Issue
- Due to network or hardware issues some OSDs started flapping intermediately affecting the cluster performance. Is there any configuration to better handle this situation to permanently mark the OSD down and out
- How are the OSDs getting marked down permanently during flapping?
Environment
- Red Hat Ceph Storage 4.x
- Non-containerized environment
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.