Resolving an OpenLDAP service failure on a VM snapshot — fixing hostname identity mismatches, loopback address issues, olcServerID with CRC32 recalculation, and removing leftover replication config from the source server.

On Friday 6th September 2024 at 21:31, we received an alert from LogicMonitor indicating one of our production web app servers (Tomcat#3) was down, with the message: "The host Tomcat#3 (i-xxxxxxx) is down". However, shortly after receiving the alert, we attempted to SSH into the VM and confirmed that the server was fully operational. But what did it go wrong?

On 15th August 2024, an incident occurred where the CPUBusyPercent alert did not trigger for one of our Production MySQL Database VMs (DB01), despite the CPU being at 100% for seven minutes. This was unexpected since the threshold settings in LogicMonitor were supposed to trigger alerts under such conditions.