Failfast / don't fail starting

Cantor, Scott cantor.2 at osu.edu
Mon Dec 2 15:24:24 EST 2013


On 12/2/13, 3:20 PM, "Paul Hethmon" <paul.hethmon at clareitysecurity.com>
wrote:
>
>As a deployer, I like the fail fast. If I give Shib some garbage, it stops
>and I know it right then instead of later when I'm asleep and the customer
>calls me to complain.

That's certainly the argument on the other side, and I'm not advocating we
don't maintain that option.

>For automatic re-loads of configuration, my view is that if the new
>configuration does not load, then it should revert to the last good
>configuration.

I would rephrase as "the new configuration should not be activated in
place of the last one, if it fails".

But same effect.

-- Scott




More information about the dev mailing list