> Perhaps I'm misunderstanding the issue here, but why would VRRP _need_ > to be running so early in boot that we haven't yet mounted /var and > /var/run?
I'm not saying there is one (that's for the project team to answer). Instead, I'm saying that an implementation detail like having an internal rendezvous file in a location that isn't mounted until mid-way through boot is a spurious reason to introduce extra dependencies for a service. Put another way, if the project team wants to argue (for conceptual reasons) that VRRP should not be available until late in boot, then so be it. But instead they seem to be arguing for introducing an artificial dependency to satisfy an unnecessary implementation restriction. I'm also worried we're deep-ending on a nit. -- meem _______________________________________________ networking-discuss mailing list [email protected]
