Files
dsh_ai1net_server/scripts/switch-C-worker.sh
T

53 lines
2.0 KiB
Bash
Raw Normal View History

#!/usr/bin/env bash
# C 步(在 106 上跑):把生产 Worker agent 装成 systemd 单元
# · env 与 47 的生产实例侧对齐(DSH_INSTANCE_* 必须一致 —— 实例是在 Worker 上 spawn 的)
# · token 走 env 文件(600)而不是命令行,避免 ps 泄露
# · 反向隧道复用演练时那把 key(其公钥已在 47 的 authorized_keys 里,restrict,port-forwarding)
set -uo pipefail
TOKEN="${WORKER_TOKEN:-dshs-worker-7f3a91c05e}"
cat > /etc/dshs-worker.env <<ENV
# DSHS 集群 Worker(2026-09-15 切换)—— 与 47 /etc/dshs.env 的**实例侧**条目保持一致
DSHS_DATA_ROOT=/var/lib/dshs
DSHS_ISOLATION_MODE=account
DSHS_DSH_BIN=/usr/bin/dsh
DSHS_BASE_UID=100000
DSH_INSTANCE_NODE_OPTIONS=--max-old-space-size=160
DSH_INSTANCE_UNIVER_SOCKET=auto
# 控制通道:Worker 主动拨 47 的反向隧道(公网入方向被云安全组挡住 ⇒ 只能这个方向)
DSHS_CLUSTER_AGENT_TOKEN=$TOKEN
[email protected]:32022
DSHS_TUNNEL_IDENTITY=/root/.ssh/tunnel_ed25519
ENV
chmod 600 /etc/dshs-worker.env
cat > /etc/systemd/system/dshs-worker.service <<UNIT
[Unit]
Description=DSHS cluster worker agent (hosts per-user dsh instances on 106)
After=network-online.target
Wants=network-online.target
[Service]
Type=simple
EnvironmentFile=/etc/dshs-worker.env
ExecStart=/usr/bin/node /opt/dshs-cluster/lib/cli.js worker --port 19000 --host 127.0.0.1 --host-id w-106 --instance-host 127.0.0.1 --log-level info
Restart=on-failure
RestartSec=3
KillMode=mixed
[Install]
WantedBy=multi-user.target
UNIT
systemctl daemon-reload
systemctl enable dshs-worker >/dev/null 2>&1
echo " 单元已装并 enable;token 长度=${#TOKEN}"
# 旧的手工 agent 若在跑先停(按端口定位)
pid=$(ss -lntpH 'sport = :19000' 2>/dev/null | grep -oP 'pid=\K[0-9]+' | head -1)
[ -n "$pid" ] && kill "$pid" && sleep 2 && echo " 已停旧的手工 agent pid=$pid"
systemctl restart dshs-worker
sleep 6
echo " dshs-worker=$(systemctl is-active dshs-worker)"
echo " healthz: $(curl -s -m 6 http://127.0.0.1:19000/healthz | head -c 200)"