Dai Ha
b67b1585c2
CI / build (push) Successful in 1m23s
The CB-307 durable ReplyInbox needs an AMQP broker, but the one behind it
was run ad hoc and had simply vanished from the host — which takes the
whole daemon with it, since AmqpReplyInbox.open throws and Bridged.java:187
does not guard it. A missing broker is a hard startup failure, not a
degraded mode, so 'how the broker runs' is part of the system, not a local
detail.
Pinned to 2.9.1 (:latest would move the broker under a running daemon),
data on a named volume (held-but-unacked replies are the entire point of
Stage 2 — a plain 'compose down' would discard exactly what durability
protects), and restart: unless-stopped so it comes back after a reboot
instead of disappearing again.
Ports are bound to 127.0.0.1 deliberately: LavinMQ ships a default
guest/guest account, which is only acceptable while nothing off-host can
reach it.
Verified by driving the production AmqpReplyInbox against this deployment
(publish/peek/dedup/FIFO/ack, then reconnect): 8/8 including redelivery of
the unacked message. That pairing had never been exercised — the
@Tag("contract") test runs against a RabbitMQ container, and is excluded
from the default build, so mvn clean install covers the broker path zero
times.
61 lines
2.4 KiB
YAML
61 lines
2.4 KiB
YAML
# LavinMQ — the AMQP broker behind bridged's durable ReplyInbox (CB-307 Stage 2).
|
|
#
|
|
# Why this file exists: the broker was previously run ad hoc and simply vanished from the host,
|
|
# which takes bridged down with it — AmqpReplyInbox.open throws on an unreachable broker and
|
|
# Bridged.java:187 does not guard it, so a missing broker is a hard startup failure, not a
|
|
# degraded mode. This pins the version, keeps the data, and brings itself back after a reboot.
|
|
#
|
|
# Usage:
|
|
# docker compose -f deploy/lavinmq/compose.yaml up -d
|
|
# docker compose -f deploy/lavinmq/compose.yaml ps
|
|
# docker compose -f deploy/lavinmq/compose.yaml logs -f
|
|
# docker compose -f deploy/lavinmq/compose.yaml down # keeps the volume
|
|
# docker compose -f deploy/lavinmq/compose.yaml down -v # DESTROYS held replies
|
|
#
|
|
# Management UI: http://127.0.0.1:15672 (guest / guest)
|
|
#
|
|
# This is bridged's OWN broker. Do not point bridged at any other AMQP server on this host —
|
|
# notably not the `local-rabbitmq` container, which belongs to a different project and would end
|
|
# up carrying this project's queues.
|
|
|
|
name: bridged-broker
|
|
|
|
services:
|
|
lavinmq:
|
|
# Pinned deliberately: :latest silently moves the broker under a running daemon.
|
|
image: cloudamqp/lavinmq:2.9.1
|
|
container_name: bridged-lavinmq
|
|
|
|
# The failure this deployment exists to prevent — survive reboots and Docker restarts, but
|
|
# stay down if it was stopped on purpose.
|
|
restart: unless-stopped
|
|
|
|
# Loopback-bound on purpose. LavinMQ ships a default guest/guest account, which is only
|
|
# acceptable because nothing off-host can reach it. bridged connects over 127.0.0.1, and
|
|
# binding 0.0.0.0 here would expose a broker with default credentials to the network.
|
|
ports:
|
|
- "127.0.0.1:5672:5672" # AMQP — bridged.yaml broker.uri points here
|
|
- "127.0.0.1:15672:15672" # HTTP management API + UI
|
|
|
|
# The whole point of Stage 2. Held-but-unacked replies live here; without a named volume a
|
|
# `docker compose down` would discard exactly what the durable inbox exists to protect.
|
|
volumes:
|
|
- lavinmq-data:/var/lib/lavinmq
|
|
|
|
healthcheck:
|
|
test: ["CMD", "lavinmqctl", "status"]
|
|
interval: 30s
|
|
timeout: 5s
|
|
retries: 3
|
|
start_period: 10s
|
|
|
|
logging:
|
|
driver: json-file
|
|
options:
|
|
max-size: "10m"
|
|
max-file: "3"
|
|
|
|
volumes:
|
|
lavinmq-data:
|
|
name: bridged-lavinmq-data
|