[ceph-users] Re: (Ceph Octopus) Repairing a neglected Ceph cluster - Degraded Data Reduncancy, all PGs degraded, undersized, not scrubbed in time

All good points (also replying to Frank Schilder) On 11/16/20 2:36 PM, Janne Johansson wrote:

Not trying to say you don't understand this, but rather that people who run small ceph clusters tend to start out with R=2 or K+1 EC because the larger faults are easier to imagine. TBH, I think I did kind of underestimate this with EC. The implications were clear-cut for me in the case of 2xReplication.

My wrong rationale was, I never had down time with my 2+1 EC with min_size=2, even when doing maintenance so I'm doing this right! But the min_size should of course be '3' for data integrity, for reasons that you and Frank have described so well in this thread. Thanks.