The LDAP replica (LDAP2) for hosted email became unresponsive which resulted in cascading issues throughout the email environment. We made some configuration changes to point to the LDAP master (LDAP1) which allowed servers to return to normal.
The root cause was an Out of Memory issue where the cached memory caused the server to run out of usable memory resulting in a crash. To remedy, we had to perform a hard reboot on the physical server at the data center.
To prevent the issue, going forward, we made some configuration adjustments to ensure proper redundancy is in place and if one LDAP goes down, it will not interrupt service. We also added some additional monitoring to catch issues like this going forward.