mirror of
https://github.com/acme-dns/acme-dns.git
synced 2026-04-27 04:45:48 +03:00
[GH-ISSUE #246] Is the listen config parameter supposed to listen to localhost? #122
Labels
No labels
Documentation
Documentation
bug
enhancement
feature request
feature request
help wanted
pull-request
question
security
security
testing
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
starred/acme-dns#122
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @tal-zvon on GitHub (Oct 29, 2020).
Original GitHub issue: https://github.com/acme-dns/acme-dns/issues/246
The sample config file says:
Should that not be
0.0.0.0:53? Why does it only need to listen to requests on the local host? Am I missing something?Or is that just an example, and you're meant to replace
127.0.0.1with0.0.0.0?@HausnerR commented on GitHub (Dec 1, 2020):
IMHO it's a bug in README and example config.
I successfully configured acme-dns making DNS server listening on
0.0.0.0:53and making API listen only in local subnet.So only I have access to API to register new endpoints and Let's encrypt can reach DNS server to confirm domain owner.
@tal-zvon commented on GitHub (Dec 1, 2020):
That makes sense. The README and example config do not. Sounds like a documentation bug to me (or whatever that would be called).
@joohoi commented on GitHub (Dec 1, 2020):
It’s actually intentional. On many configurations systemd resolved is
listening at 127.0.1.53:53 which makes acme-dns unable to bind to 0.0.0.0:53
On Tue 1. Dec 2020 at 21.02, terminator14 notifications@github.com wrote:
@tal-zvon commented on GitHub (Dec 1, 2020):
Would not binding to 127.0.0.1:53 (as the documentation suggests) be the equivalent to binding to any other random IP? Just like binding to 1.2.3.4:53, it would technically allow the program to start (since the binding would succeed), but wouldn't allow LetsEncrypt to reach it. As far as I can tell, there's no point in listening to 127.0.0.1:53 since LetsEncrypt requests will never come from localhost, and will never hit that. Am I wrong?
If that's true, you might as well say
listen = "w.x.y.z:53", and just have a note to tell people to type in the IP of the interface they want to listen on. At least users would know that's something they need to figure out, rather than type in 127.0.0.1:53 as is and hope it works.@joohoi commented on GitHub (Dec 1, 2020):
Yeah, you are right. It would probably serve the users better as far as the documentation goes. Do you want to create a PR to fix it or should I do it?
@tal-zvon commented on GitHub (Dec 1, 2020):
If you've got the time, go ahead.
Thanks