Ceph - Erasure coded PGs stuck unclean
Issue
- Ceph - Erasure coded PGs stuck unclean
- On an EC pool with an 8+3 jerasure profile we have a few PGs that won't become clean after losing a node from the cluster. We expect them to recover but instead it looks like CRUSH can't find a good mapping.
Environment
- Red Hat Ceph Storage 1.3.z
- Ceph version 0.94.7
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.