CEPH Reduced Data Avability

[Teapot]

Hello,
We have 6 node cluster with every node 6 SSD disk.

When I delete one OSD from node1 and node3 , it says "reduced data avability" and write operations is stopping.

When I delete disk and re-added its fixed problem.

Is it normal ? How many disk can be loose same time ?
Cluster size : 3
Min size : 2
Used Space : %20

 

[spirit]

with size=3 + min_size=2, you can loose 1 full node.

with size=3 + min_size=1 you can loose 2 full nodes


min_size is a protection, setting objets in readonly if you have less than x replicat.

 

[Teapot]

with size=3 + min_size=2, you can loose 1 full node.

with size=3 + min_size=1 you can loose 2 full nodes


min_size is a protection, setting objets in readonly if you have less than x replicat.

Totally can I lose 6 SSD driver of different nodes without any data corruption is it ok?
but I still don't understand why I'm having problems with 2 disks.
thanks.