[#49] swarm test重启测试通过

2025-11-19 17:26:26 +08:00
parent 1d38304936
commit d4e0dc1511
19 changed files with 697 additions and 116 deletions
--- a/deployment_new/templates/server/compose/docker-compose.yml
+++ b/deployment_new/templates/server/compose/docker-compose.yml
@@ -5,18 +5,9 @@ networks:
    external: true

 services:
-  bind:
-    image: ${BIND_IMAGE_TAG:-argus-bind9:${PKG_VERSION}}
-    container_name: argus-bind-sys
-    networks: [argus-sys-net]
-    volumes:
-      - ../private:/private
-    restart: unless-stopped
-
  master:
    image: ${MASTER_IMAGE_TAG:-argus-master:${PKG_VERSION}}
    container_name: argus-master-sys
-    depends_on: [bind]
    environment:
      - OFFLINE_THRESHOLD_SECONDS=6
      - ONLINE_THRESHOLD_SECONDS=2
@@ -29,7 +20,10 @@ services:
      - ../private/argus/master:/private/argus/master
      - ../private/argus/metric/prometheus:/private/argus/metric/prometheus
      - ../private/argus/etc:/private/argus/etc
-    networks: [argus-sys-net]
+    networks:
+      argus-sys-net:
+        aliases:
+          - master.argus.com
    restart: unless-stopped

  es:
@@ -47,7 +41,10 @@ services:
    ports:
      - "${ES_HTTP_PORT:-9200}:9200"
    restart: unless-stopped
-    networks: [argus-sys-net]
+    networks:
+      argus-sys-net:
+        aliases:
+          - es.log.argus.com

  kibana:
    image: ${KIBANA_IMAGE_TAG:-argus-kibana:${PKG_VERSION}}
@@ -63,27 +60,10 @@ services:
    ports:
      - "${KIBANA_PORT:-5601}:5601"
    restart: unless-stopped
-    networks: [argus-sys-net]
-
-  ftp:
-    image: ${FTP_IMAGE_TAG:-argus-metric-ftp:${PKG_VERSION}}
-    container_name: argus-ftp
-    restart: unless-stopped
-    environment:
-      - TZ=Asia/Shanghai
-      - FTP_BASE_PATH=/private/argus/ftp
-      - FTP_PASSWORD=${FTP_PASSWORD:-NASPlab1234!}
-      - DOMAIN=${FTP_DOMAIN:-ftp.metric.argus.com}
-      - ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}
-      - ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}
-    ports:
-      - "${FTP_PORT:-21}:21"
-      - "${FTP_DATA_PORT:-20}:20"
-      - "${FTP_PASSIVE_HOST_RANGE:-21100-21110}:21100-21110"
-    volumes:
-      - ../private/argus/metric/ftp:/private/argus/ftp
-      - ../private/argus/etc:/private/argus/etc
-    networks: [argus-sys-net]
+    networks:
+      argus-sys-net:
+        aliases:
+          - kibana.log.argus.com

  prometheus:
    image: ${PROM_IMAGE_TAG:-argus-metric-prometheus:${PKG_VERSION}}
@@ -99,7 +79,10 @@ services:
    volumes:
      - ../private/argus/metric/prometheus:/private/argus/metric/prometheus
      - ../private/argus/etc:/private/argus/etc
-    networks: [argus-sys-net]
+    networks:
+      argus-sys-net:
+        aliases:
+          - prom.metric.argus.com

  grafana:
    image: ${GRAFANA_IMAGE_TAG:-argus-metric-grafana:${PKG_VERSION}}
@@ -122,7 +105,10 @@ services:
      - ../private/argus/metric/grafana:/private/argus/metric/grafana
      - ../private/argus/etc:/private/argus/etc
    depends_on: [prometheus]
-    networks: [argus-sys-net]
+    networks:
+      argus-sys-net:
+        aliases:
+          - grafana.metric.argus.com

  alertmanager:
    image: ${ALERT_IMAGE_TAG:-argus-alertmanager:${PKG_VERSION}}
@@ -133,7 +119,10 @@ services:
    volumes:
      - ../private/argus/etc:/private/argus/etc
      - ../private/argus/alert/alertmanager:/private/argus/alert/alertmanager
-    networks: [argus-sys-net]
+    networks:
+      argus-sys-net:
+        aliases:
+          - alertmanager.alert.argus.com
    ports:
      - "${ALERTMANAGER_PORT:-9093}:9093"
    restart: unless-stopped
@@ -151,19 +140,25 @@ services:
      - EXTERNAL_KIBANA_PORT=${WEB_PROXY_PORT_8083:-8083}
    volumes:
      - ../private/argus/etc:/private/argus/etc
-    networks: [argus-sys-net]
+    networks:
+      argus-sys-net:
+        aliases:
+          - web.argus.com
    restart: unless-stopped

  web-proxy:
    image: ${WEB_PROXY_IMAGE_TAG:-argus-web-proxy:${PKG_VERSION}}
    container_name: argus-web-proxy
-    depends_on: [bind, master, grafana, prometheus, kibana, alertmanager]
+    depends_on: [master, grafana, prometheus, kibana, alertmanager]
    environment:
      - ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}
      - ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}
    volumes:
      - ../private/argus/etc:/private/argus/etc
-    networks: [argus-sys-net]
+    networks:
+      argus-sys-net:
+        aliases:
+          - proxy.argus.com
    ports:
      - "${WEB_PROXY_PORT_8080:-8080}:8080"
      - "${WEB_PROXY_PORT_8081:-8081}:8081"
@@ -172,4 +167,3 @@ services:
      - "${WEB_PROXY_PORT_8084:-8084}:8084"
      - "${WEB_PROXY_PORT_8085:-8085}:8085"
    restart: unless-stopped
-
--- a/deployment_new/templates/server/docs/INSTALL_SERVER_zh.md
+++ b/deployment_new/templates/server/docs/INSTALL_SERVER_zh.md
@@ -71,7 +71,7 @@ export SWARM_MANAGER_ADDR=<本机管理IP>
 - 等待“六项就绪”：
  - Master `/readyz`=200、ES `/_cluster/health`=200、Prometheus TCP 可达、Grafana `/api/health`=200、Alertmanager `/api/v2/status`=200、Kibana `/api/status` level=available；
 - 将各服务 overlay IP 写入 `private/argus/etc/<域名>`，Reload Bind9 与 Nginx；
- 写出 `cluster-info.env`（含 `BINDIP/FTPIP/SWARM_JOIN_TOKEN_{WORKER,MANAGER}/SWARM_MANAGER_ADDR`）；
+- 写出 `cluster-info.env`（含 `SWARM_JOIN_TOKEN_{WORKER,MANAGER}/SWARM_MANAGER_ADDR`；compose 架构下不再依赖 BINDIP/FTPIP）；
 - 生成 `安装报告_YYYYMMDD-HHMMSS.md`（端口、健康检查摘要与提示）。

 看到什么才算成功：
@@ -79,7 +79,7 @@ export SWARM_MANAGER_ADDR=<本机管理IP>
 - `安装报告_…md` 中各项 HTTP 检查为 200/available；
 - `cluster-info.env` 包含五个关键键：
  - `SWARM_MANAGER_ADDR=...`
-  - `BINDIP=10.x.x.x` `FTPIP=10.x.x.x`
+  - `SWARM_MANAGER_ADDR=...` `SWARM_JOIN_TOKEN_*=...`
  - `SWARM_JOIN_TOKEN_WORKER=SWMTKN-...`
  - `SWARM_JOIN_TOKEN_MANAGER=SWMTKN-...`

--- a/deployment_new/templates/server/scripts/install.sh
+++ b/deployment_new/templates/server/scripts/install.sh
@@ -88,23 +88,15 @@ for i in $(seq 1 "$RETRIES"); do
 done
 [[ $ok -ge 6 ]] || err "部分服务未就绪（可稍后重试 selfcheck）"

-# Resolve overlay IPs
-bind_c=argus-bind-sys; ftp_c=argus-ftp
-BINDIP=$(docker inspect -f '{{ (index .NetworkSettings.Networks "'$NET_NAME'").IPAddress }}' "$bind_c" 2>/dev/null || true)
-FTPIP=$(docker inspect -f '{{ (index .NetworkSettings.Networks "'$NET_NAME'").IPAddress }}' "$ftp_c" 2>/dev/null || true)
-info "解析 overlay IP: BINDIP=${BINDIP:-<empty>} FTPIP=${FTPIP:-<empty>}"
-
 # Swarm join tokens
 TOKEN_WORKER=$(docker swarm join-token -q worker 2>/dev/null || echo "")
 TOKEN_MANAGER=$(docker swarm join-token -q manager 2>/dev/null || echo "")

-# cluster-info.env
+# cluster-info.env（compose 场景下不再依赖 BINDIP/FTPIP）
 CI="$PKG_ROOT/cluster-info.env"
-info "写入 cluster-info.env (manager/token/IP)"
+info "写入 cluster-info.env (manager/token)"
 {
  echo "SWARM_MANAGER_ADDR=${SWARM_MANAGER_ADDR:-}"
-  echo "BINDIP=${BINDIP:-}"
-  echo "FTPIP=${FTPIP:-}"
  echo "SWARM_JOIN_TOKEN_WORKER=${TOKEN_WORKER:-}"
  echo "SWARM_JOIN_TOKEN_MANAGER=${TOKEN_MANAGER:-}"
 } > "$CI"
@@ -131,10 +123,6 @@ RPT="$PKG_ROOT/安装报告_${ts}.md"
  echo "- JOIN_TOKEN_WORKER=${TOKEN_WORKER:-}"
  echo "- JOIN_TOKEN_MANAGER=${TOKEN_MANAGER:-}"
  echo
-  echo "## Overlay IPs"
-  echo "- BINDIP=${BINDIP:-}"
-  echo "- FTPIP=${FTPIP:-}"
-  echo
  echo "## 健康检查（简要）"
  echo "- master/readyz=$(code http://127.0.0.1:${MASTER_PORT:-32300}/readyz)"
  echo "- es/_cluster/health=$(code http://127.0.0.1:${ES_HTTP_PORT:-9200}/_cluster/health)"
@@ -146,30 +134,4 @@ RPT="$PKG_ROOT/安装报告_${ts}.md"
 info "已生成报告: $RPT"

 info "安装完成。可将 cluster-info.env 分发给 Client-GPU 安装方。"
-
-# 写入域名→overlay IP 并热更新 Bind/Nginx
-ETC_DIR="$PKG_ROOT/private/argus/etc"; mkdir -p "$ETC_DIR"
-declare -A MAP
-MAP[web-frontend]=web.argus.com
-MAP[argus-grafana]=grafana.metric.argus.com
-MAP[argus-prometheus]=prom.metric.argus.com
-MAP[argus-kibana-sys]=kibana.log.argus.com
-MAP[argus-alertmanager]=alertmanager.alert.argus.com
-MAP[argus-master-sys]=master.argus.com
-changed=0
-for cname in "${!MAP[@]}"; do
-  domain="${MAP[$cname]}"; fpath="$ETC_DIR/$domain"
-  ip=$(docker inspect -f '{{ (index .NetworkSettings.Networks "'$NET_NAME'").IPAddress }}' "$cname" 2>/dev/null || true)
-  [[ -z "$ip" ]] && { echo "[DNS-FIX][WARN] $domain: container $cname no overlay IP yet"; continue; }
-  cur=$(cat "$fpath" 2>/dev/null || echo "")
-  if [[ "$cur" != "$ip" ]]; then
-    echo "$ip" > "$fpath"; echo "[DNS-FIX][SET] $domain = $ip (was: ${cur:-<empty>})"; changed=1
-  else
-    echo "[DNS-FIX][OK] $domain already $ip"
-  fi
-done
-if [[ $changed -eq 1 ]]; then
-  docker exec argus-bind-sys /usr/local/bin/reload-bind9.sh >/dev/null 2>&1 || docker exec argus-bind-sys rndc reload >/dev/null 2>&1 || true
-  sleep 1
-fi
 docker exec argus-web-proxy nginx -t >/dev/null 2>&1 && docker exec argus-web-proxy nginx -s reload >/dev/null 2>&1 || true