<blockquote>It's not proven yet that IBM was at fault here. This has happened before; Air New Zealand, if I remember correctly - the CEO went online blaming IBM shortly after the systems crashed, it turned out to be their own fault.</blockquote>
The company I worked for at the time was hosted in the same datacentre when that UPS failed that your talking about. We only had only one issue as for the most part our HA systems kicked in. We had our HA cluster nodes running of different UPS feeds. One system was not migrated to the cluster yet....
I'm told Air NZ had their failover/hot standby system offline at the time.
footnote: That datacentre was old, tier 1, with a history of such problems, and IBM no longer use it having built a new one of higher grade nearby.