Docker resolver breaks podman

Hours/days of "...but it used to work!"

Simplest fix: to /etc/nginx/conf.d/default.conf

resolver 127.0.0.11 10.89.1.1 valid=10s ipv6=off;

Commit a4a7a59b from Mar 20, 2026 added the Docker resolver:

 fix(sandbox): path isolation, DooD volume mounting, hints, and nginx DNS    

(#129, #136)

 - Nginx DNS: add Docker resolver (127.0.0.11) with dynamic upstream
   variable to handle backend container IP changes

It was added to handle Docker container restarts with new IPs. Works on
Docker, breaks on podman.

Simpler bug report: "The Docker DNS resolver is unnecessary and breaks podman.
Nginx resolves goclaw hostname at startup - the explicit resolver only helps
if goclaw restarts alone, which doesn't happen in a compose stack."

A drop in for nginx conf.d to fix it

auto:/code/goclaw/options/podman$ cat nginx-podman-resolver.conf 
# Podman DNS resolver override — loaded after default.conf
# Podman uses network gateway IP instead of Docker's 127.0.0.11
resolver 10.89.1.1 valid=10s ipv6=off;

networks:
  default:
    driver: bridge

services:
  goclaw-ui:
    volumes:
      - ${GOCLAW_DIR}/options/podman/nginx-podman-resolver.conf:/etc/nginx/conf.d/zz-podman-dns.conf:ro

Start with the Nginx resolver in /etc/nginx/conf.d/default.conf and the Podman override at options/podman/nginx-podman-resolver.conf, then inspect the goclaw-ui volume in the compose configuration. Reproduce startup with Podman and verify that the UI can resolve goclaw; done means the configuration works with Podman without breaking Docker behavior.