Files
mtusi2026lab6/evidence/stand/from-scratch.md
T
ricoromaandClaude Opus 5.5 a4497e3f32 feat: add OpenTelemetry SDK, collector and telemetry stores
Collector (contrib), Prometheus, Loki, Tempo and Grafana with provisioning
from git; SDK in API and worker with level 3 metrics, traces and logs
carrying trace_id/span_id; telemetry off by default outside the stand;
broker healthcheck runs as the broker user so a fresh stand starts once.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-11 10:19:50 +03:00

7.5 KiB
Raw Blame History

driver.py stand-from-scratch --runs 2 (change otel-telemetry, задача 11.3)

  • Основа: commit 9b35c25 (propose), рабочее дерево не закоммичено (apply change otel-telemetry). Основная сессия решает, переснимать ли после коммита.
  • Дата: 2026-11-11, старт 06:27:32Z, ~2 мин; команда из корня: python3 .claude/skills/run-mtusi-messenger/driver.py stand-from-scratch --runs 2.
  • Сценарий container-run «Стенд с нуля одной командой» (design, решения 19 и 21). Проект compose mtusi-lab6-scratch (рабочий mtusi-lab6 команда отказывается брать — код 2), свободные порты хоста для APP_OUT_PORT, POSTGRES_OUT_PORT, VALKEY_OUT_PORT, S3_OUT_PORT, RABBITMQ_OUT_PORT, GRAFANA_OUT_PORT аргументами make run_docker, COMPOSE_PROJECT_NAME окружением. Перед первым прогоном и после каждого — docker compose -p mtusi-lab6-scratch down -v --remove-orphans изнутри команды (в том числе при Ctrl+C): volumes проекта пусты перед прогоном и после него. .env и config.toml — рабочего стенда (секреты те же, данные — свои volumes).
  • Рабочий стенд на время проверки остановлен make stop (два стенда — 14 ГиБ лимитов на ВМ 7,75 ГиБ); его 17 контейнеров до и после — с теми же ID, затем docker compose start.
  • До change на пустом volume брокер падал на eacces (.erlang.cookie от root, KNOWN_ISSUES.md) и make run_docker кончался кодом 1; healthcheck теперь gosu rabbitmq rabbitmq-diagnostics -q check_port_connectivity.
working stand mtusi-lab6: 17 containers
down -v mtusi-lab6-scratch: code 0
run 1: make run_docker took 13 s
down -v mtusi-lab6-scratch: code 0
run 2: make run_docker took 12 s
down -v mtusi-lab6-scratch: code 0

run 1: project volumes before                      expected 0               got 0
run 1: make run_docker exit code                   expected 0               got 0
run 1: api state                                   expected running:0       got running:0
run 1: cleaner state                               expected running:0       got running:0
run 1: grafana state                               expected running:0       got running:0
run 1: haproxy state                               expected running:0       got running:0
run 1: loki state                                  expected running:0       got running:0
run 1: migrate state                               expected exited:0        got exited:0
run 1: otel-collector state                        expected running:0       got running:0
run 1: postgres state                              expected running:0       got running:0
run 1: prometheus state                            expected running:0       got running:0
run 1: rabbitmq state                              expected running:0       got running:0
run 1: relay state                                 expected running:0       got running:0
run 1: s3 state                                    expected running:0       got running:0
run 1: s3-bootstrap state                          expected exited:0        got exited:0
run 1: tempo state                                 expected running:0       got running:0
run 1: valkey state                                expected running:0       got running:0
run 1: worker state                                expected running:0       got running:0
run 1: rabbitmq RestartCount                       expected 0               got 0
run 1: otel-collector RestartCount                 expected 0               got 0
run 1: prometheus RestartCount                     expected 0               got 0
run 1: loki RestartCount                           expected 0               got 0
run 1: tempo RestartCount                          expected 0               got 0
run 1: grafana RestartCount                        expected 0               got 0
run 1: eacces / permission denied in logs          expected 0               got 0
run 1: grafana datasources provisioned and healthy expected yes             got yes
run 1: dashboard provisioned                       expected 200 True        got 200 True
run 1: make test through its LB                    expected 0 PASS          got 0 PASS
run 1: volumes left                                expected 0               got 0
run 2: project volumes before                      expected 0               got 0
run 2: make run_docker exit code                   expected 0               got 0
run 2: api state                                   expected running:0       got running:0
run 2: cleaner state                               expected running:0       got running:0
run 2: grafana state                               expected running:0       got running:0
run 2: haproxy state                               expected running:0       got running:0
run 2: loki state                                  expected running:0       got running:0
run 2: migrate state                               expected exited:0        got exited:0
run 2: otel-collector state                        expected running:0       got running:0
run 2: postgres state                              expected running:0       got running:0
run 2: prometheus state                            expected running:0       got running:0
run 2: rabbitmq state                              expected running:0       got running:0
run 2: relay state                                 expected running:0       got running:0
run 2: s3 state                                    expected running:0       got running:0
run 2: s3-bootstrap state                          expected exited:0        got exited:0
run 2: tempo state                                 expected running:0       got running:0
run 2: valkey state                                expected running:0       got running:0
run 2: worker state                                expected running:0       got running:0
run 2: rabbitmq RestartCount                       expected 0               got 0
run 2: otel-collector RestartCount                 expected 0               got 0
run 2: prometheus RestartCount                     expected 0               got 0
run 2: loki RestartCount                           expected 0               got 0
run 2: tempo RestartCount                          expected 0               got 0
run 2: grafana RestartCount                        expected 0               got 0
run 2: eacces / permission denied in logs          expected 0               got 0
run 2: grafana datasources provisioned and healthy expected yes             got yes
run 2: dashboard provisioned                       expected 200 True        got 200 True
run 2: make test through its LB                    expected 0 PASS          got 0 PASS
run 2: volumes left                                expected 0               got 0
working stand mtusi-lab6: same container IDs       expected yes             got yes
PASS: make run_docker on empty volumes passed 2 times at the first attempt

Код выхода — 0. «running:0» — состояние и код выхода контейнера (docker compose ps --format json) через 30 с после make run_docker; у двух реплик API состояние одно.