[ceph-users] Recovery throughput inversely linked with rbd_cache_xyz?

Hello, A couple days ago I increased the rbd cache size from the default to 256MB/osd on a small 4 node, 6 osd/node setup in a test/lab setting.  The rbd volumes are all vm images with writeback cache parameters and steady if only a few mb/sec writes going on. Logging mostly.    I noticed the recovery throughput went down 10x - 50x .  Using Ceph nautilus.  Am I seeing a coincidence or should recovery throughput tank when rbd cache sizes go up?  The underlying pools are mirrored on three disks each on a different nodes. Thanks! Harry Coin