osd: deletes are performed inline during pg log processing
With a large number of deletes in a client workload, this can easily saturate a disk and cause very high latency, since it does not go through the op queue or do any batching.
To fix this, make recovery handle these deletes instead of performing them during peering.
- Status changed from New to In Progress
- Project changed from Ceph to RADOS
- Category set to Peering
- Component(RADOS) OSD added
- Status changed from In Progress to Resolved
- Status changed from Resolved to Pending Backport
- Backport set to jewel
- Copied to Backport #22406: jewel: osd: deletes are performed inline during pg log processing added
- Status changed from Pending Backport to Resolved
Also available in: Atom