I am running openldap-2.4.44 (source build) using lmdb backend with configuration set in slapd.conf on a Linux machine. The configuration includes dbnosync and also checkpoint settings. The kernel vm settings are flushing dirty cache pages to disc periodically, so my question is what is the reason for the checkpoint setting if the kernel is already taking care of page flushing?
Geoff Swan wrote:
I am running openldap-2.4.44 (source build) using lmdb backend with configuration set in slapd.conf on a Linux machine. The configuration includes dbnosync and also checkpoint settings. The kernel vm settings are flushing dirty cache pages to disc periodically, so my question is what is the reason for the checkpoint setting if the kernel is already taking care of page flushing?
Probably not needed if you're satisfied with the kernel's flush frequency.
openldap-technical@openldap.org