Fix EAI_AGAIN: work around Docker's internal-network DNS bug with a static IP
- egress-proxy gets a fixed IP (172.28.5.10) on a new 172.28.5.0/24 subnet for the 'egress' network. - sde-meeting-toolkit gets a static extra_hosts entry for 'egress-proxy' instead of relying on Docker's embedded DNS, which breaks for internal-network names on a container that's also on a normal network (confirmed via a real smoke test run - EAI_AGAIN). - Update CHANGELOG with the root cause and the subnet-collision caveat.
This commit is contained in:
@@ -2,6 +2,11 @@
|
||||
|
||||
All notable changes to this project are logged here, newest first. This file starts from the point the toolkit was handed off and set up in this repo.
|
||||
|
||||
## 2026-09-29 (4)
|
||||
|
||||
- Fixed (found via real smoke test on the user's own Docker): `sde-meeting-toolkit` couldn't resolve `egress-proxy` at all (`getaddrinfo EAI_AGAIN egress-proxy`) once `egress` was marked `internal: true`. Root cause is a known Docker bug: a container on both a normal network (`proxy`) and an `internal: true` network (`egress`) can lose embedded DNS resolution for names on the internal side. Fixed by giving `egress-proxy` a fixed IP (`172.28.5.10` on a new `172.28.5.0/24` subnet) and adding a static `extra_hosts` entry on `sde-meeting-toolkit` instead of relying on DNS for that one name.
|
||||
- Note: `172.28.5.0/24` is an arbitrary private subnet choice. If it collides with another Docker network already on a host, change it (and the matching `ipv4_address`/`extra_hosts` entries) to an unused one - Docker's error message names the conflict if this happens.
|
||||
|
||||
## 2026-09-29 (3)
|
||||
|
||||
- Added: outbound network restriction. `sde-meeting-toolkit` no longer has any direct route to the internet. Its one outbound call (the AI draft button, to `api.anthropic.com`) is routed through a new `egress-proxy` container (`egress-proxy/`, tinyproxy) that only allows that one destination and refuses everything else, even from inside a compromised app container.
|
||||
|
||||
@@ -45,6 +45,14 @@ services:
|
||||
networks:
|
||||
- proxy
|
||||
- egress
|
||||
# A container attached to both a normal network ("proxy") and an
|
||||
# internal: true network ("egress") can hit a real Docker bug where
|
||||
# embedded DNS stops resolving names on the internal side (EAI_AGAIN),
|
||||
# confirmed while testing this. Sidestep it with a static hosts entry
|
||||
# instead of relying on DNS for this one name - egress-proxy's fixed
|
||||
# IP is set below via ipv4_address.
|
||||
extra_hosts:
|
||||
- "egress-proxy:172.28.5.10"
|
||||
depends_on:
|
||||
- egress-proxy
|
||||
# Hardening for a container that's reachable from the public internet
|
||||
@@ -61,8 +69,12 @@ services:
|
||||
container_name: sde-meeting-toolkit-egress-proxy
|
||||
restart: unless-stopped
|
||||
networks:
|
||||
- egress
|
||||
- default
|
||||
egress:
|
||||
# Fixed IP so sde-meeting-toolkit can reach this by static hosts
|
||||
# entry (extra_hosts above) instead of DNS, working around the
|
||||
# internal-network DNS bug. Must match that entry if ever changed.
|
||||
ipv4_address: 172.28.5.10
|
||||
default: {}
|
||||
cap_drop:
|
||||
- ALL
|
||||
security_opt:
|
||||
@@ -80,3 +92,6 @@ networks:
|
||||
# out except through egress-proxy, which also sits on "default" and IS
|
||||
# meant to have one.
|
||||
internal: true
|
||||
ipam:
|
||||
config:
|
||||
- subnet: 172.28.5.0/24
|
||||
|
||||
Reference in New Issue
Block a user