Hi Isaiah,
A simple solution for multi-site redundancy is to have two nearby sites
with < 3ms latency and setup crush map [0] for datacenter level
redundancy instead of the default host level.
Performance was adequate in my testing for large number of small files
if the latency between all n
Isaiah,
I'm trying to understand your requirements for a CephFS Active-Active setup.
What do you want to achieve with a CephFS Active-Active setup ?
Once you list the exact requirements, we can discuss further on how to
achieve them.
There's also something called *CephFS Snapshot Mirroring*:
https