I do not know of an open source router hardware that is available (at ISP scale)
Getting an X86 box and throwing freebsd and quagga on it might work for small amounts of traffic. But for ISP scale bandwidth your stuck with a few main vendors...
You're basically correct but it's worth remembering that it is possible to add a bunch of quad nics to whitebox and run bpf on it, courtesy of one of the BSDs and do real routing at an ISP scale.
That would be completely open source (and quite secure).
Probably pretty performant if you knew what you were doing.
Its never going to work in a core network routing 100s of gigs of traffic. So it's not going to work for ISPs
How so? Is there anywhere a list of hashes of software that multiple people independently from each other verified, compiled (and compared)? Because without that, "open source" adds 0 to security in the scenarios we are talking about here.
Additionally, how many packages/firmwares have you inspected the last, let's say, decade? If I were a betting man, I'd put my money on 'less than I can count on one hand' (well maybe you're in infosec and get paid for audits...)
Which open source projects have deterministic builds, with a verifiable source of signature information - be that concensus or a source of truth that is verfied by concensus, so that I can be sure that either a) the binary I download has a very high certainty of being correct, or b) the source code I downloaded and compiled from has a very high certainty of being un-modified.
With a vendor, they don't need deterministic builds so badly, as they push a single file, and an authoritative source of signatures.
I have an example of a binary where the difference between a remotely exploitable service and a fixed version of the same service is a single bit. That bit (or some other bit that also changes the code's functionality in an important way) can be changed by any person or (potentially compromised) computer in the chain of compilation, packaging, and distribution unless someone has a way to tell whether what those people and computers assert is "correct". If the binaries routinely change appreciably every time someone cuts a software release, it's going to be expensive to go back and check whether there's also a malicious change, bearing in mind that the person whose computer made the change doesn't have to have been malicious.
And in the mean time, a million monkeys have social-engineered your secretary's machine through a Farmville js trick and pwned your network 3 ways from Sunday.
The point is that should be able to read the source code like a recepie for your cake. If someone hides the recepie and tells you its a yummy cake but you know its full of exploits its not.
Main point is, if it can be hacked it will be hacked. It´s better that we can read the source code and verify that the software in our devices has not been tampered with.
Also we need an inventory of where things can be hidden in devices. Are there embedded flash areas where are they?
A modern reinstall should:
1) Reinstall the operating system from a known good source 2) Flash all firmware and flash chips with known good code.
Both operating system and firmware should be open source so you can inspect the code.