> IF something did go wrong, they would at times call IBM and ask for service before the customer had any knowledge of an issue. I always assumed their call ho

> IF something did go wrong, they would at times call IBM and ask for service before the customer had any knowledge of an issue. I always assumed their call home service was triggered by SNMP but may be wrong? Regardless, IBM exposed a lot of metrics via SNMP so it was always an easy way to query it for metrics and/or accept traps from their devices/OS’s for failures