RI SSU IPv6 address not properly updated when NTCP is disabled

[anonimal]

---

By submitting this issue, I confirm the following:

• I have read and understood the contributor guide in kovri-docs.
• I have checked that the issue I am reporting can be replicated or that the feature I am suggesting is not present.
• I have checked opened or recently closed pull requests for existing solutions/implementations to my issue/suggestion.

---

A hexdump of router.info confirms that, despite IPv6 being enabled and used in SSU, our RI is not properly updated when --enable-ntcp 0 (the entire IPv6 fields are missing). When default NTCP is enabled (--enable-ntcp 1) and IPv6 is enabled, the RI is properly updated for both transports.

This issue is not related to #776 though is in a similar area.

[brbzull0]

What's the role that --host plays when --enable-ssp or --enable-ntcp is set to 0? The provided host may not match the ones in the existing RI and if --enable-ntcp 0 is used we are disabling NTCP for an existing host that didn't match the one in the --host flag.

[MoroccanMalinois]

@brbzull0 Option --host is used for socket binding, i.e. choosing wich network interface to use, just like any "classic" TCP/UDP service. It's conceptually unrelated to SSU or NTCP.

The provided host may not match the ones in the existing RI

Don't know about NTCP, so i'll only speak about SSU. Correct, but when SSU is enabled, there is a mechanism to detect the external IP of the host (First thing done after the reseed). The router will then update its RI accordingly and republish.

[anonimal]

@brbzull0 --host is essentially deprecated but cannot be fully deprecated until we implement external NTCP-only IP detection (which would also resolve #776). There may be other arguments for keeping --host but I can't think of any off the top of my head.

Sidenote: we must have at least 1 transport running and I notice that we allow disabling both transports. I'll open a new issue.

[anonimal]

NOTICE: THIS ISSUE HAS BEEN MOVED TO GitLab. Please continue the discussion there. See #1013 for details.