On 3/22/19 8:41 PM, Quanah Gibson-Mount wrote:
--On Friday, March 22, 2019 8:35 PM +0100 Michael Ströder michael@stroeder.com wrote:
See also my response. It's the usual monitoring-shows-contextCSN-diff while replication just works. It's a bug I'm observing for years now.
It's not a bug.
syncprov-checkpoint <ops> <minutes> After a write operation has succeeded, write the contextCSN to the underlying database if <ops> write operations or more than <minutes> time have passed since the last checkpoint.
Does that really mean that contextCSN value retrieved via LDAP are not current? This would also break replication.
Up to now I understood that this option is similar to slapd-mdb's checkpoint directive. Which means, the data retrieved via LDAP is current but the DB files might not be fsynced yet.
Also in Æ-DIR the default is:
syncprov-checkpoint 1000 2
So according to the man-page you cited it means the gap must not be greater than two minutes. But it's sometimes several hours and it lasts for hours even if you write again something.
IMO it's simply a bug.
Ciao, Michael.