Hi List,
I've got an LDAP architecture with one read-write OpenLDAP (primary) and some read-only OpenLDAP (replica).
I load the cn=monitor backend on the primary? is it sufficient to have the cn=monitor backend on the slave too or should I activate it on the replicas ?
Regards,
Kevin C
This totally depends on whether cn=config is replicated from the primary to the read-only servers.
If cn=config is specific to each host you need to configure the monitoring backend on each host.
But you could try this easily. Just configure the backend on the primary and try to access cn=monitor on the replica.
Best,
Uwe
Am 21.11.23 um 15:27 schrieb Kevin Cousin:
Hi List,
I've got an LDAP architecture with one read-write OpenLDAP (primary) and some read-only OpenLDAP (replica).
I load the cn=monitor backend on the primary? is it sufficient to have the cn=monitor backend on the slave too or should I activate it on the replicas ?
Regards,
Kevin C
--On Tuesday, November 21, 2023 3:27 PM +0100 Kevin Cousin kco@ik.me wrote:
Hi List,
I've got an LDAP architecture with one read-write OpenLDAP (primary) and some read-only OpenLDAP (replica).
I load the cn=monitor backend on the primary? is it sufficient to have the cn=monitor backend on the slave too or should I activate it on the replicas ?
I generally enable cn=config on all my ldap servers, so I can gather their individual statistics. Lots of monitoring software out there supports cn=monitor as well (Datadog, Grafana, etc)
--Quanah
openldap-technical@openldap.org