Over the last 24 hours we have had a series of stability issues. First, the DC our mail/lists machine is hosted in had power issues ( Internap LGA9, NY, USA ); then we had a h/w failure and needed a chassis swap (
thanks to the guys in the DC who executed this within minutes ); and we have since again had network issues – although power to the DC and the server seems stable now.
Over the next 24 hours, lets just try and be a bit patient and work with this. We have good backups so if things dont stabalise in the next 24
hrs we will failover.
The primary reason we have not done this as yet ( failover ), is down to the fact that its taken a long time, in years, for us to build a reputation with large hosting facilities and email providers and are on many whitelists. Moving the host over makes it hard for us to get that sort of level of trust again really quickly. Longer term, we should perhaps spread the load through multiple machines and then ensure we have good relationships out of all these machines exit IP’s with the various vendors and service providers.
There are backup MX’s in place, so if there are issues with the primary machine, we will have mail’s go into holding queues and we should not lose anything inbound.