Centrica has pinned last week's 36-hour freeze of its Hive app estate – the one that coincided with the plunging temperatures in Britain – on a server fault it claims happened, er, a day after the outage actually began.
On Tuesday 2 April, users were forced to manually alter home thermostats and surveillance controls after the HiveHome website, dashboard and mobile app crashed and burned. Nearly all device services were listed as not functioning, save for Hive Leak.
Efforts to bring things back online started at 9:44 UTC on the day but it wasn't until the following afternoon (Wednesday 3 April) that normal service resumed for some, but not all customers.
Centrica emailed users a post screw-up report at pub-o-clock on Friday evening, which was suitably light on technical details and if correct, implied a glitch in the space time continuum.
"First of all, we're really sorry if this has affected you," Hive managing director Claire Miles begun earnestly.
"We want to assure you that we've been working around the clock to fix all issues and bring all affected devices back online. We were able to reset most of your devices remotely by Wednesday afternoon".
Hapless punters still unable to use the app-based service were given an eight-point list of checks to make: this included verifying that Hive Hub was still connected to the mains power socket or indeed switched on.
Centrica then gave a brief explanation of why the app went AWOL and managed to balls that up by saying a server wobble on Wednesday had caused the outage on the prior day.
"The incident was caused by a server, owned by one of our external service providers, going offline on Wednesday morning. While this is the first time an incident like this has taken place, we have taken extra steps with our supplier to make sure this does not happen again," Miles added.
It is the first time El Reg has heard of this too. We called Centrica to gain some clarity on the situation but the press office had not responded to our call for comment at the time of writing. ®