So this isn’t about routing really, rather about optimizing standby routers for recovery.
A few things make me nervous.
First, the description of the work involved seems to imply that your setup really needs more automated tooling. Nontrivial, but you’ve already mentioned typos, and that this is for large operations.
Second, using IPv4 for your management network is wasteful and needlessly complicated. Even if your customer traffic is all IPv4, there’s really no reason to use legacy protocols for internal routing.
None of this is real, everything I said was hypothetical to demonstrate the point.
I get what you’re trying to say, but what you’re saying is in favor of unnumbered compatible routing protocols.
I do not presently work in a provider or datacenter scale environment, and of the few that I’ve seen that I’ve been able to “peek behind the curtain” so to speak, the issues I’m pointing at are very real.
So this isn’t about routing really, rather about optimizing standby routers for recovery.
A few things make me nervous.
First, the description of the work involved seems to imply that your setup really needs more automated tooling. Nontrivial, but you’ve already mentioned typos, and that this is for large operations.
Second, using IPv4 for your management network is wasteful and needlessly complicated. Even if your customer traffic is all IPv4, there’s really no reason to use legacy protocols for internal routing.
None of this is real, everything I said was hypothetical to demonstrate the point.
I get what you’re trying to say, but what you’re saying is in favor of unnumbered compatible routing protocols.
I do not presently work in a provider or datacenter scale environment, and of the few that I’ve seen that I’ve been able to “peek behind the curtain” so to speak, the issues I’m pointing at are very real.