-
Notifications
You must be signed in to change notification settings - Fork 67
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Remove unnecessary published ports #38
Comments
caddy is bound to the host network: searx-docker/docker-compose.yaml Line 11 in 92cdb70
The reason: filtron needs the orginal IP address, and Docker won't send the original IPv6 address if caddy is not bound to the host network (it works with IPv4 addresses). See moby/moby#17666 You can have a look to this PR #26 (and for a workaround https://github.com/robbertkl/docker-ipv6nat ). So in this configuration, the filtron and morty ports are exposed. |
Just disable dockers userland proxy which will become a default in the future.
Which I think is a really bad default because docker bypasses ufw by default. |
* Use docker image caddy:2-alpine * Caddyfile: remove "limits 10KB" * Caddyfile: URL /filtron/rules removes (filtron API still availabled on http://localhost:4041/rules ) * caddy storage are docker volumes (caddy-data and caddy-config). start.sh and stop.sh have been modified to keep these volumes. * .env: Remove SEARX_PROTOCOL, SEARX_TLS, FILTRON_USER and FILTRON_PASSWORD variables. * docker-compose.yml: filtron and morty listen on 127.0.0.1 (related to #38) * Fix #37: settings ```SEARX_HOSTNAME=localhost:8888``` works as expected (https connection)
Why are the ports of some services published to the host despite not needing them? Caddy needs them for sure but why do filtron or morty need them?
The text was updated successfully, but these errors were encountered: