6Jul 09
Datacentre disaster lessons
Posted by Renai LeMay, Chris Duckett and Suzanne Tindal @ 16:03 1 comments
Play or download the Patch Monday podcast below:
Embed podcast:
As a system administrator, the health and status of your datacentre is at the forefront of your mind. But how often do you think about the needs beyond server status and bandwidth?
In this episode of Patch Monday, we examine what happens when the ability of generators or the status of air conditioners fall by the wayside.
First up, we look at the TransGrid outage that caused Sydney-based Equinix to have a datacentre outage, yet other Sydney-based datacentres remained unaffected.
This is followed up by the horror story of a Queensland Health datacentre, which started with an electricity brown-out and cascaded until 75 per cent of applications were down due to an air-conditioning failure.
What's your worst datacentre experience? Let us know below in the talkback.
3%
2%
Capacity is one thing but in these cases it was systems not tested. It comes down to the data centre facility managers to be managing capacity and testing. Macquarie Telecom was offline due to a cooling system failure due to water to the building being cut off. Where are there alerting systems to know water levels were critical?