IT Automation Digest for November 16, 2012

Canary concept for system updates (Server Density Blog). Server Density’s David Mytton shares how they roll out updates to OS packages, kernels, and Microsoft’s Patch Tuesday, where the ‘canaries’ get the updates first before the rest of the fleet. All automated, of course.
Beyond VM deployment: Monitoring your VM’s at scale (SlideShare). How do you build and maintain cloudscale monitoring infrastructure that is ‘sane and manageable’? Kris Buytaert shares his approach to deploying dynamically generated monitoring using Nagios, Graphite, Collectd, and Puppet.
How one NYC data center survived Hurricane Sandy (Ars Technica). How did Internap’s data center in downtown Manhattan cope when ‘water was rushing down the elevator shafts, filling the lobby with anywhere from one to three feet of water…’? (tip o’ the hat to Reddit).
Are We Seeing The Renaissance Of Enterprises In The Cloud? (High Scalability). …or are the analysts just working overtime? High Scalability cites research, sponsored by VMware and EMC, stating that ~1/3 of respondents plan on moving mission-critical apps to the cloud in the next year.

Puppet Labs