I sent this to the users list yesterday, but it's really is more of a developer question so I'm reposting it here: One of the main limitations of using CephFS is the requirement to reduce the number of active MDS daemons to one during upgrades. As far as I can tell this has been a known problem since Luminous (~2017). This issue essentially requires downtime during upgrades for any CephFS cluster that needs more than one active MDS at all times. I saw there were some improvements to the upgrade process with 16.2.6 (you no longer have to stop the standby MDSes), but it has me wondering if there are any plans to fix this limitation soon? Thanks, Bryan _______________________________________________ Dev mailing list -- dev@xxxxxxx To unsubscribe send an email to dev-leave@xxxxxxx
