Hello Fernando, This discussion appears to be talking about the same key points consistently, so I will keep this as short as possible.
Sure, but the costly thing that may be sponsored is the connectivity, not the IP addresses which can be reduced if the provider doesn't have them all available. Resources can still be used by an organisation, even if they they are not the most expensive operational cost to running a [insert use type here]. can just adjust their requirements to fit into what is available As has been explained previously, it is not reasonable to require an attendee to an event to make adjustments to their network back home in order to use a service at a conference/event which they might attend once or twice a year. I cannot agree with the statement that neither CGNAT or 464XLAT isn't viable for larges conferences As has also been explained, it would be a very bad idea from a security risk perspective to whitelist an address or range of addresses that 100's or 1000's of people sit behind. CGNAT and 464XLAT has the potential to cause issues, which goes against the first line of the Technical Requirements for the APRICOT Summit Network which reads "delegates need to carry on with their day job reliably and without impediment". The same can be said for other conferences which are not run by APNIC or APRICOT. I wanted to understand in which way being behind a CGNAT+IPv6 that is a problem. I already answered this in my first reply to your message: "it's not wise or smart to whitelist an IP address that is in use by 100's of people, whereas by allocating public addresses directly to end devices it allows them to whitelist their specific device addresses. It is akin to whitelisting the public IP address of the WiFi network at your favourite cafe, which IMO is a big no-no." It doesn't sound reasonable to me to think that people who work mostly in the Internet industry are unable to have a VPN Server in their companies to do their day job. What is the alternative ? They call to a colleague, inform the Public IPv4 they received and the person allows in it in the firewall? It is beyond the scope of RIR policies to tell network operators what access methods they must use while in attendance at their conferences. Accordingly, they need to ensure that all network operators are able to access their networks regardless of what access methods are available. If this means that an operator whitelists an IP address on their firewall so they can SSH into a jumpbox, then this is what they must do. Yes it is not for us to tell people who to operate their networks, but it is up to us to avoid wasting valuable resources that may be better used for something else provided there are reasonable technical solutions. I agree that it is not for us to tell operators how to run their networks, and that we must not waste resources. Having said that, it is often subjective as to what is defined as "wastage" and what is not. I still don't see justification to have a policy to reserve addresses for this type of usage that is not critical You are correct, this policy is not "critical" in the sense that it is not being used for critical infrastructure or the community must reach a consensus on it. It however (in my view) is rather important to provide access to resources for non-profit users where they may have a temporary use-case. A /21 IPv4 prefix is a rather small prefix in the scheme of things. Regards, Christopher Hawker
_______________________________________________ SIG-policy - https://mailman.apnic.net/[email protected]/ To unsubscribe send an email to [email protected]
