Closed ginkel closed 4 years ago
Just noticed this might be related to #48
Correct, this is currently not supported, as the Docker cli has no support for IPv6 addresses in the -p
argument. We'll need to have a separate mapping for that, but I'm not yet sure if #48 is the way to go.
Currently the only options is to specify the IPv6 addresses manually with .-p [fe80::1]:443:443
Currently the only options is to specify the IPv6 addresses manually with
-p [fe80::1]:443:443
.
# docker run -it --rm -p '[fe80::1]:80:80' nginx
docker: Error response from daemon: driver failed programming external connectivity on endpoint sad_ride (5cf2fd827d91d07f88a6a6a69c35abea47f5912722eead82b70ecdfcffefdd83): listen tcp [fe80::1]:80: bind: invalid argument.
Things seem to look better if I use the public IPv6 address of the host, though
@ginkel
Sorry, if you want to use fe80::/64
you need to specify the interface. The following should work:
docker run -it --rm -p '[::1]:80:80' nginx
@robbertkl We can close this.
Hi there,
I recently added an extra IP address to some of my servers so I can have multiple containers listening on the same port. Is there a way to expose one of them via IPv6. Previously, I used to have
-p 443:443
, which worked fine, now I have-p 1.2.3.4:443:443
, which no longer gets exposed via IPv6.Any hints are much appreciated!