Practices of Using NFS Over RGW
summary
Tests write two 1 GB files over NFS to RGW and observe only 1 GB reclaimed while 2 GB remains until GC ends, traced to duplicate GC tags sent repeatedly so only one OMAP entry is deleted. Tags combine pool id, rados_handle_id, request_id, and the implementation leaves request_id stuck at zero after init so handle ID collisions produce identical tags. Fix generates a fresh request_id per write (PR submitted). Best fit: large files, few objects, low consistency, few renames.
AI-written summary · about