Skip to content

ddns-scripts: add Dynway provider - #30385

Open
knutmichaelhaugland wants to merge 1 commit into
openwrt:masterfrom
wayscloudas:dynway-provider-definition
Open

ddns-scripts: add Dynway provider#30385
knutmichaelhaugland wants to merge 1 commit into
openwrt:masterfrom
wayscloudas:dynway-provider-definition

Conversation

@knutmichaelhaugland

Copy link
Copy Markdown

Add a DynDNS2 provider definition for dynway.eu.

The service supports IPv4 and IPv6 through dynupdate.wayscloud.services/nic/update with HTTP Basic authentication and uses the standard good and nochg response values. The existing HTTPS option upgrades the endpoint to HTTPS.

Add a DynDNS2 provider definition for dynway.eu. Both IPv4 and IPv6
use dynupdate.wayscloud.services/nic/update with HTTP Basic
authentication and standard good/nochg status responses.

Bump PKG_RELEASE so existing installations receive the provider
definition.

Signed-off-by: Knut Michael Haugland <hostmaster@wayscloud.net>

@openwrt-ai openwrt-ai left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 new commit. Nothing here blocks a merge — the single inline note is an optional question about the URL scheme.

Checks that came back clean:

  • PKG_RELEASE 2 → 3 is correct: the new JSON changes what ddns-scripts-services ships, and PKG_VERSION is unchanged.
  • No Makefile install hunk is needed — Package/ddns-scripts-services/install globs ./files/usr/share/ddns/default/*, and dynway.eu.json is not one of the "special services" that get rm'd into a dedicated subpackage.
  • The list entry is correctly placed: dynway.eu sorts between dynv6.com and easydns.com, and the file is the "service on demand" manifest fetched at runtime by ddns.sh from raw.githubusercontent — so it is right that it is not installed by the Makefile, and right that a plain (non-dedicated-package) provider appears in it.
  • The JSON parses, uses tabs like its neighbours, and matches the dyndns.org.json DynDNS2 template exactly, "answer": "good|nochg" included (56 other definitions use that same answer string, and it is matched with grep -i -E, so the alternation is fine).
  • Commit subject scope ddns-scripts: matches the touched files, and the body accurately describes the diff.
  • All three FormalityCheck runs are green on 48323b0.

The identical myip=[IP] on both the ipv4 and ipv6 entries mirrors dyndns.org.json and is called out explicitly in the commit message, so I did not flag it — just confirm the endpoint really does accept an IPv6 literal in myip rather than needing a separate myipv6 parameter.


Generated by Claude Code

{
"name": "dynway.eu",
"ipv4": {
"url": "http://[USERNAME]:[PASSWORD]@dynupdate.wayscloud.services/nic/update?hostname=[DOMAIN]&myip=[IP]",

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Optional / non-blocking, but worth a decision: with http:// the Basic-auth credentials sit in the URL userinfo and go out in cleartext by default. The PR description notes that the HTTPS option upgrades the endpoint, but that option is opt-in — use_https=0 in dynamic_dns_updater.sh:152 — and the rewrite at dynamic_dns_functions.sh:844 only fires when the user enables it. So the out-of-the-box path for a fresh dynway.eu config sends the password unencrypted.

Both patterns exist in the tree, so this isn't a convention violation: ~29 definitions use http://[USERNAME]:[PASSWORD]@, but a dozen hardcode HTTPS — the closest sibling being now-dns.com.json, which is byte-for-byte this shape with https://. The tradeoff for hardcoding is that HTTPS needs a CA bundle (or no_cert_check) on the device, whereas http:// works on a bare install.

Since dynway.eu is a brand-new provider with no legacy HTTP-only clients to keep working: does dynupdate.wayscloud.services serve HTTPS with a publicly-trusted certificate? I couldn't reach the host from this environment to confirm, so I'm asking rather than prescribing. If it does, https:// here (and on line 8) would make the default configuration safe. Same question applies to the ipv6 URL below.


Generated by Claude Code

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants