Files
fips/testing/static/docker-compose.yml
Johnathan Corgan e4a854f6b0 Stop concurrent local CI runs from clobbering each other's containers
Two runs on one host destroyed each other's containers, producing
mid-test "No such container" failures that look like real defects. The
automated builder runs a full local CI on the same box every few minutes,
so the machine is contended almost always and this has red-ed both a hand
run and an automated gate.

Two independent causes are fixed here. Container names were hardcoded, and
docker names are global rather than scoped by compose project, so two runs
collided on the same name; every name now takes an optional suffix that
the harness sets from the run id. And the cleanup sweep matched a label
shared by every run, so one run's teardown force-removed another's
containers; resources now also carry a per-run label and the sweep can be
narrowed to it.

A third hazard turned up that was not in the original report: the sidecar
suite passes explicit compose project names, which override the run-scoped
project and put it outside the shared prefix entirely. Its project names,
network, and derived container references are now scoped too.

Both are default-off. With the suffix unset, names render exactly as they
do today and a bare compose invocation is unchanged, which is what keeps
the hosted CI and the documentation correct. A cleanup run with no run id
still reaps everything, which is what a manual "clear the box" wants.

The literal-name sweep was not sufficient: six scripts build container
names dynamically from node labels, and two suites create their own
containers outside compose. Those are handled at their construction sites.

Verified: syntax check on all modified scripts; compose validation on
every modified file with the suffix both set and unset; and a synthetic
two-run reproduction that shows the old cleanup destroying a bystander run
and the new one leaving it alone.

Known gap: subnets are still hardcoded, so two concurrent full runs will
still collide on address-pool overlap. That fix reaches into topology
configs, chaos scenarios, diagrams and production source, so it is left
for its own change rather than half-done here.
2026-07-19 06:41:25 +00:00

580 lines
18 KiB
YAML

networks:
fips-net:
driver: bridge
labels:
- "com.corganlabs.fips-ci=1"
ipam:
config:
- subnet: 172.20.0.0/24
gateway-lan:
driver: bridge
labels:
- "com.corganlabs.fips-ci=1"
enable_ipv6: true
ipam:
config:
- subnet: 172.20.1.0/24
- subnet: fd02::/64
x-fips-common: &fips-common
image: ${FIPS_TEST_IMAGE:-fips-test:latest}
cap_add:
- NET_ADMIN
devices:
- /dev/net/tun:/dev/net/tun
sysctls:
- net.ipv6.conf.all.disable_ipv6=0
restart: "no"
env_file:
- ./generated-configs/npubs.env
environment:
- RUST_LOG=info
# Passthrough for A/B benchmarking — set on the host shell before
# `docker compose up` to override the worker-pool defaults.
- FIPS_ENCRYPT_WORKERS=${FIPS_ENCRYPT_WORKERS:-}
- FIPS_DECRYPT_WORKERS=${FIPS_DECRYPT_WORKERS:-}
- FIPS_PERF=${FIPS_PERF:-0}
- FIPS_PERF_INTERVAL_SECS=${FIPS_PERF_INTERVAL_SECS:-5}
services:
# ── Mesh topology ──────────────────────────────────────────────
node-a:
<<: *fips-common
profiles: ["mesh"]
container_name: fips-node-a${FIPS_CI_NAME_SUFFIX:-}
hostname: node-a
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/mesh/node-a.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.10
node-b:
<<: *fips-common
profiles: ["mesh"]
container_name: fips-node-b${FIPS_CI_NAME_SUFFIX:-}
hostname: node-b
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/mesh/node-b.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.11
node-c:
<<: *fips-common
profiles: ["mesh"]
container_name: fips-node-c${FIPS_CI_NAME_SUFFIX:-}
hostname: node-c
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/mesh/node-c.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.12
node-d:
<<: *fips-common
profiles: ["mesh"]
container_name: fips-node-d${FIPS_CI_NAME_SUFFIX:-}
hostname: node-d
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/mesh/node-d.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.13
node-e:
<<: *fips-common
profiles: ["mesh"]
container_name: fips-node-e${FIPS_CI_NAME_SUFFIX:-}
hostname: node-e
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/mesh/node-e.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.14
# ── Mesh-public topology (mesh + external public node) ────────
pub-a:
<<: *fips-common
profiles: ["mesh-public"]
container_name: fips-node-a${FIPS_CI_NAME_SUFFIX:-}
hostname: node-a
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/mesh-public/node-a.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.10
pub-b:
<<: *fips-common
profiles: ["mesh-public"]
container_name: fips-node-b${FIPS_CI_NAME_SUFFIX:-}
hostname: node-b
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/mesh-public/node-b.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.11
pub-c:
<<: *fips-common
profiles: ["mesh-public"]
container_name: fips-node-c${FIPS_CI_NAME_SUFFIX:-}
hostname: node-c
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/mesh-public/node-c.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.12
pub-d:
<<: *fips-common
profiles: ["mesh-public"]
container_name: fips-node-d${FIPS_CI_NAME_SUFFIX:-}
hostname: node-d
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/mesh-public/node-d.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.13
pub-e:
<<: *fips-common
profiles: ["mesh-public"]
container_name: fips-node-e${FIPS_CI_NAME_SUFFIX:-}
hostname: node-e
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/mesh-public/node-e.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.14
# ── Chain topology (A-B-C-D-E) ────────────────────────────────
chain-a:
<<: *fips-common
profiles: ["chain"]
container_name: fips-node-a${FIPS_CI_NAME_SUFFIX:-}
hostname: node-a
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/chain/node-a.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.10
chain-b:
<<: *fips-common
profiles: ["chain"]
container_name: fips-node-b${FIPS_CI_NAME_SUFFIX:-}
hostname: node-b
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/chain/node-b.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.11
chain-c:
<<: *fips-common
profiles: ["chain"]
container_name: fips-node-c${FIPS_CI_NAME_SUFFIX:-}
hostname: node-c
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/chain/node-c.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.12
chain-d:
<<: *fips-common
profiles: ["chain"]
container_name: fips-node-d${FIPS_CI_NAME_SUFFIX:-}
hostname: node-d
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/chain/node-d.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.13
chain-e:
<<: *fips-common
profiles: ["chain"]
container_name: fips-node-e${FIPS_CI_NAME_SUFFIX:-}
hostname: node-e
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/chain/node-e.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.14
# ── Rekey integration test (mesh + aggressive rekey timers) ──
rekey-a:
<<: *fips-common
profiles: ["rekey"]
container_name: fips-node-a${FIPS_CI_NAME_SUFFIX:-}
hostname: node-a
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey/node-a.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.10
rekey-b:
<<: *fips-common
profiles: ["rekey"]
container_name: fips-node-b${FIPS_CI_NAME_SUFFIX:-}
hostname: node-b
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey/node-b.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.11
rekey-c:
<<: *fips-common
profiles: ["rekey"]
container_name: fips-node-c${FIPS_CI_NAME_SUFFIX:-}
hostname: node-c
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey/node-c.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.12
rekey-d:
<<: *fips-common
profiles: ["rekey"]
container_name: fips-node-d${FIPS_CI_NAME_SUFFIX:-}
hostname: node-d
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey/node-d.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.13
rekey-e:
<<: *fips-common
profiles: ["rekey"]
container_name: fips-node-e${FIPS_CI_NAME_SUFFIX:-}
hostname: node-e
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey/node-e.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.14
# ── Rekey + accept_connections=false on node b ──────────────────
# Exercises the auto_connect-initiator-with-accept-off regression
# class. Same 5-node mesh, but node b's generated config has
# `transports.udp.accept_connections: false` (injected by
# rekey-test.sh's inject-config when REKEY_ACCEPT_OFF_NODES=b).
rekey-accept-off-a:
<<: *fips-common
profiles: ["rekey-accept-off"]
container_name: fips-node-a${FIPS_CI_NAME_SUFFIX:-}
hostname: node-a
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug,fips::node::handlers::handshake=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey-accept-off/node-a.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.10
rekey-accept-off-b:
<<: *fips-common
profiles: ["rekey-accept-off"]
container_name: fips-node-b${FIPS_CI_NAME_SUFFIX:-}
hostname: node-b
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug,fips::node::handlers::handshake=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey-accept-off/node-b.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.11
rekey-accept-off-c:
<<: *fips-common
profiles: ["rekey-accept-off"]
container_name: fips-node-c${FIPS_CI_NAME_SUFFIX:-}
hostname: node-c
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug,fips::node::handlers::handshake=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey-accept-off/node-c.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.12
rekey-accept-off-d:
<<: *fips-common
profiles: ["rekey-accept-off"]
container_name: fips-node-d${FIPS_CI_NAME_SUFFIX:-}
hostname: node-d
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug,fips::node::handlers::handshake=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey-accept-off/node-d.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.13
rekey-accept-off-e:
<<: *fips-common
profiles: ["rekey-accept-off"]
container_name: fips-node-e${FIPS_CI_NAME_SUFFIX:-}
hostname: node-e
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug,fips::node::handlers::handshake=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey-accept-off/node-e.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.14
# ── Rekey topology with udp.outbound_only=true on node b ────
# Exercises the udp.outbound_only rekey-msg1 admission regression
# observed in production 2026-04-30. Same 5-node mesh as
# rekey-accept-off, but node-b's generated config has
# `transports.udp.outbound_only: true` (forces ephemeral bind +
# accept_connections false) AND node-b's peer-c address is
# rewritten to the docker hostname (`node-c:2121`) so the
# `addr_to_link` lookup misses on the inbound numeric source addr.
# Injected by rekey-test.sh's inject-config when
# REKEY_OUTBOUND_ONLY_NODES=b.
rekey-outbound-only-a:
<<: *fips-common
profiles: ["rekey-outbound-only"]
container_name: fips-node-a${FIPS_CI_NAME_SUFFIX:-}
hostname: node-a
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug,fips::node::handlers::handshake=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey-outbound-only/node-a.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.10
rekey-outbound-only-b:
<<: *fips-common
profiles: ["rekey-outbound-only"]
container_name: fips-node-b${FIPS_CI_NAME_SUFFIX:-}
hostname: node-b
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug,fips::node::handlers::handshake=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey-outbound-only/node-b.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.11
rekey-outbound-only-c:
<<: *fips-common
profiles: ["rekey-outbound-only"]
container_name: fips-node-c${FIPS_CI_NAME_SUFFIX:-}
hostname: node-c
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug,fips::node::handlers::handshake=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey-outbound-only/node-c.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.12
rekey-outbound-only-d:
<<: *fips-common
profiles: ["rekey-outbound-only"]
container_name: fips-node-d${FIPS_CI_NAME_SUFFIX:-}
hostname: node-d
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug,fips::node::handlers::handshake=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey-outbound-only/node-d.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.13
rekey-outbound-only-e:
<<: *fips-common
profiles: ["rekey-outbound-only"]
container_name: fips-node-e${FIPS_CI_NAME_SUFFIX:-}
hostname: node-e
environment:
- RUST_LOG=info,fips::node::handlers::rekey=debug,fips::node::handlers::handshake=debug
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/rekey-outbound-only/node-e.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.14
# ── TCP chain topology (A-B-C) ───────────────────────────────
tcp-a:
<<: *fips-common
profiles: ["tcp-chain"]
container_name: fips-node-a${FIPS_CI_NAME_SUFFIX:-}
hostname: node-a
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/tcp-chain/node-a.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.10
tcp-b:
<<: *fips-common
profiles: ["tcp-chain"]
container_name: fips-node-b${FIPS_CI_NAME_SUFFIX:-}
hostname: node-b
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/tcp-chain/node-b.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.11
tcp-c:
<<: *fips-common
profiles: ["tcp-chain"]
container_name: fips-node-c${FIPS_CI_NAME_SUFFIX:-}
hostname: node-c
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/tcp-chain/node-c.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.12
# ── Gateway integration test (gateway + server + non-FIPS client) ─
gw-gateway:
<<: *fips-common
profiles: ["gateway"]
container_name: fips-gw-gateway${FIPS_CI_NAME_SUFFIX:-}
hostname: gw-gateway
# Privileged required: gateway must enable IPv6 on eth1 (second network,
# attached after container start) and manage nftables NAT rules.
privileged: true
environment:
- RUST_LOG=info
- FIPS_TEST_MODE=gateway
sysctls:
- net.ipv6.conf.all.disable_ipv6=0
- net.ipv6.conf.default.disable_ipv6=0
- net.ipv6.conf.all.forwarding=1
- net.ipv6.conf.all.proxy_ndp=1
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/gateway/node-a.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.10
gateway-lan:
ipv4_address: 172.20.1.10
ipv6_address: fd02::10
gw-server:
<<: *fips-common
profiles: ["gateway"]
container_name: fips-gw-server${FIPS_CI_NAME_SUFFIX:-}
hostname: gw-server
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/gateway/node-b.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.11
# Second mesh destination — gives gw-client-2 a distinct npub to target
# so the gateway allocates a separate virtual-IP mapping per LAN client.
# Mirrors gw-server; not on gateway-lan.
gw-server-2:
<<: *fips-common
profiles: ["gateway"]
container_name: fips-gw-server-2${FIPS_CI_NAME_SUFFIX:-}
hostname: gw-server-2
volumes:
- ../docker/resolv.conf:/etc/resolv.conf:ro
- ./generated-configs/gateway/node-c.yaml:/etc/fips/fips.yaml:ro
networks:
fips-net:
ipv4_address: 172.20.0.12
gw-client:
image: ${FIPS_TEST_APP_IMAGE:-fips-test-app:latest}
profiles: ["gateway"]
container_name: fips-gw-client${FIPS_CI_NAME_SUFFIX:-}
hostname: gw-client
cap_add:
- NET_ADMIN
sysctls:
- net.ipv6.conf.all.disable_ipv6=0
volumes:
- ./configs/gateway-resolv.conf:/etc/resolv.conf:ro
networks:
gateway-lan:
ipv4_address: 172.20.1.20
ipv6_address: fd02::20
restart: "no"
env_file:
- ./generated-configs/npubs.env
# Second LAN client — exercises concurrent multi-client mappings.
# Same image and gateway-lan attachment as
# gw-client; the gateway must allocate a distinct virtual IP for it.
gw-client-2:
image: ${FIPS_TEST_APP_IMAGE:-fips-test-app:latest}
profiles: ["gateway"]
container_name: fips-gw-client-2${FIPS_CI_NAME_SUFFIX:-}
hostname: gw-client-2
cap_add:
- NET_ADMIN
sysctls:
- net.ipv6.conf.all.disable_ipv6=0
volumes:
- ./configs/gateway-resolv.conf:/etc/resolv.conf:ro
networks:
gateway-lan:
ipv4_address: 172.20.1.21
ipv6_address: fd02::21
restart: "no"
env_file:
- ./generated-configs/npubs.env