Re: Blockchain and Networking

[Denys Fedoryshchenko <denys () visp net lb>]

On 2018-01-08 08:59, Peter Kristolaitis wrote:
> On 2018-01-08 12:52 AM, William Herrin wrote:
> > I'm having trouble envisioning a scenario where blockchain does that 
> > any
> > better than plain old PKI.
> >
> > Blockchain is great at proving chain of custody, but when do you need 
> > to do
> > that in computer networking?
> >
> > Regards,
> > Bill Herrin
>
> There's probably some potential in using a blockchain for things like
> configuration management.  You can authenticate who made what change
> and when (granted, we can kinda-sorta do this already with the various
> authentication and logging mechanisms, but the blockchain is an
> immutable, permanent record inherently required for the system to work
> at all).
>
> That immutable, sequenced chain of events would let you do things like
> "make my test environment look like production did last Thursday at
> 9AM" trivially by reading the blockchain up until that timestamp, then
> running a fork of the chain for the new test environment to track its
> own changes during testing.
>
> Or when you know you did something 2 months ago for client A, and you
> need your new NOC guy to now do it for client B -- the blockchain
> becomes the documentation of what was done.
>
> We can build all of the above in other ways today, of course.  But
> there's certainly something to be said for a vendor-supported solution
> that is inherent in the platform and requires no additional
> infrastructure.  Whether or not that's worth the complexities of
> managing a blockchain on networking devices is, perhaps, a whole other
> discussion.   :)
>
> - Peter
Why to reinvent git? :)
Lot of tools available also, to see diff on git commits, to see who did 
commit, and what exactly he changed.
(it is possible to cryptographically sign commits, as well, and yes, 
they are chain signed, as "blockchain")