Closed derroman closed 6 years ago
Hi, this issue is most common when your DNS isn't set up correctly or it is not yet fully propagated. Does nslookup sub.mydomain.com
return the correct IP address?
Yes, it does. The IP address returned is correct.
(BTW: thanks for the ultra fast response!)
What about on server itself? This is a failure on a pre-flight check, where
the container tries to access http://sub.mydomain.com:80
. When it fails,
it means either the DNS is not propagated, or the port 80 is blocked
somehow. (Local mode doesn't do the check).
On Sun, Jan 14, 2018 at 4:26 PM, roman notifications@github.com wrote:
Yes, it does. The IP address returned is correct.
(BTW: thanks for the ultra fast response!)
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/SteveLTN/https-portal/issues/114#issuecomment-357519191, or mute the thread https://github.com/notifications/unsubscribe-auth/AAfAZ2tR2LLS1HJaMe4cybsWaYpe41Jwks5tKhyxgaJpZM4RdoLr .
mh... there is an automated redirect to https (port 443) when I hit port 80 (http)
That sounds normal. But somehow port 80 of the given domain is not accessible from within the container. If you just set up the DNS, wait for a while for it to fully propagate. I'm afraid I can't help you with more than that.
On Sun, Jan 14, 2018 at 4:36 PM, roman notifications@github.com wrote:
mh... there is an automated redirect to https (port 443) when I hit port 80 (http)
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/SteveLTN/https-portal/issues/114#issuecomment-357519927, or mute the thread https://github.com/notifications/unsubscribe-auth/AAfAZ6ViUiDX6KJHbad3m7oAVH4KEjIzks5tKh8ZgaJpZM4RdoLr .
Okay, I've used the same config locally, created a subdomain to my local computer (opening ports and stuff in my local router, etc) --- it worked out like a charm. Registering works.
But unfortunately I have absolutely no idea, why the same docker-compose config does not work on this specific server.
I will investigate a little bit more testing, and will report if I find the error.
I've found the issue! The reason is my firewall configuration:
I've created the following json file (couple of months ago), to prevent docker from bypassing any UFW firewall rules.
/etc/docker/daemon.json
{
"iptables": false
}
Together with UFW this worked out fine (Source: https://svenv.nl/unixandlinux/dockerufw/). UFW enables ports 80 and 443 (both TCP) Everything was okay, until I wanted to start using https-portal.
What does not work:
What does work:
But without the config json, preventing docker from bypassing UWF rules, it seems not correct.
okay. So, we can close this for now: I've followed the instructions by svenv.nl again and now everything works fine. Thanks.
Glad you solved this!
On Sun, Jan 14, 2018 at 8:36 PM, roman notifications@github.com wrote:
okay. So, we can close this for now: I've followed the instructions by svenv.nl again and now everything works fine. Thanks.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/SteveLTN/https-portal/issues/114#issuecomment-357535824, or mute the thread https://github.com/notifications/unsubscribe-auth/AAfAZ_U9_Ru1RHrUAtrozj5va8CwesiJks5tKlcggaJpZM4RdoLr .
Hi! I really love this little cool helper - when I run my config with "local" everything runs fine, but as soon as I switch to "staging" and therefore start talking to let's encrypt, if fails as follows:
Here's the part of my docker-compose config:
Am I missing something? Any help is highly appreciated. Thanks!