On Tue, Mar 5, 2013 at 12:58 PM, Michael B. Smith <[email protected]> wrote: > But at the most basic level – it was a human error (as I read it). “Someone” > didn’t mark the update package as a critical update.
At the most basic level, all errors are human errors. :-) Either someone didn't follow the design, or someone didn't foresee a failure case in the design. > Arguably (and I can see this), because the package had 5 weeks before it was > required… Prolly a better approach to scheduling is not a raw priority field, but something deadline and dependency based. </captain obvious> It also occurs to me that treating the certificates as part of the software build may be a poor choice. I realize that a software install needs something to start with, but that could be a bootstrap element, and from there it could update via other mechanisms. (Think DNS root hints.) But I don't know enough about their architecture to say. It might be their "Fabric Controller" management architecture is more flexible than that blog post leads me to believe, or there might be other factors involved that I'm not aware. -- Ben ~ Finally, powerful endpoint security that ISN'T a resource hog! ~ ~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/> ~ --- To manage subscriptions click here: http://lyris.sunbelt-software.com/read/my_forums/ or send an email to [email protected] with the body: unsubscribe ntsysadmin
