228 lines
5.1 KiB
Bash
228 lines
5.1 KiB
Bash
#!/bin/bash
|
|
# Nomad 多数据中心节点自动配置脚本
|
|
# 数据中心: ${datacenter}
|
|
|
|
set -e
|
|
|
|
# 日志函数
|
|
log() {
|
|
echo "[$(date '+%Y-%m-%d %H:%M:%S')] $1" | tee -a /var/log/nomad-setup.log
|
|
}
|
|
|
|
log "开始配置 Nomad 节点 - 数据中心: ${datacenter}"
|
|
|
|
# 更新系统
|
|
log "更新系统包..."
|
|
apt-get update -y
|
|
apt-get upgrade -y
|
|
|
|
# 安装必要的包
|
|
log "安装必要的包..."
|
|
apt-get install -y \
|
|
curl \
|
|
wget \
|
|
unzip \
|
|
jq \
|
|
podman \
|
|
htop \
|
|
net-tools \
|
|
vim
|
|
|
|
# 启动 Podman
|
|
log "启动 Podman 服务..."
|
|
systemctl enable podman
|
|
systemctl start podman
|
|
usermod -aG podman ubuntu
|
|
|
|
# 安装 Nomad
|
|
log "安装 Nomad ${nomad_version}..."
|
|
cd /tmp
|
|
wget -q https://releases.hashicorp.com/nomad/${nomad_version}/nomad_${nomad_version}_linux_amd64.zip
|
|
unzip nomad_${nomad_version}_linux_amd64.zip
|
|
mv nomad /usr/local/bin/
|
|
chmod +x /usr/local/bin/nomad
|
|
|
|
# 创建 Nomad 用户和目录
|
|
log "创建 Nomad 用户和目录..."
|
|
useradd --system --home /etc/nomad.d --shell /bin/false nomad
|
|
mkdir -p /opt/nomad/data
|
|
mkdir -p /etc/nomad.d
|
|
mkdir -p /var/log/nomad
|
|
chown -R nomad:nomad /opt/nomad /etc/nomad.d /var/log/nomad
|
|
|
|
# 获取本机 IP 地址
|
|
if [ "${bind_addr}" = "auto" ]; then
|
|
# 尝试多种方法获取 IP
|
|
BIND_ADDR=$(curl -s http://169.254.169.254/latest/meta-data/local-ipv4 2>/dev/null || \
|
|
curl -s http://metadata.google.internal/computeMetadata/v1/instance/network-interfaces/0/ip -H "Metadata-Flavor: Google" 2>/dev/null || \
|
|
ip route get 8.8.8.8 | awk '{print $7; exit}' || \
|
|
hostname -I | awk '{print $1}')
|
|
else
|
|
BIND_ADDR="${bind_addr}"
|
|
fi
|
|
|
|
log "检测到 IP 地址: $BIND_ADDR"
|
|
|
|
# 创建 Nomad 配置文件
|
|
log "创建 Nomad 配置文件..."
|
|
cat > /etc/nomad.d/nomad.hcl << EOF
|
|
datacenter = "${datacenter}"
|
|
region = "dc1"
|
|
data_dir = "/opt/nomad/data"
|
|
|
|
bind_addr = "$BIND_ADDR"
|
|
|
|
%{ if server_enabled }
|
|
server {
|
|
enabled = true
|
|
bootstrap_expect = ${bootstrap_expect}
|
|
encrypt = "${nomad_encrypt_key}"
|
|
}
|
|
%{ endif }
|
|
|
|
%{ if client_enabled }
|
|
client {
|
|
enabled = true
|
|
|
|
host_volume "podman-sock" {
|
|
path = "/run/podman/podman.sock"
|
|
read_only = false
|
|
}
|
|
}
|
|
%{ endif }
|
|
|
|
ui {
|
|
enabled = true
|
|
}
|
|
|
|
addresses {
|
|
http = "0.0.0.0"
|
|
rpc = "$BIND_ADDR"
|
|
serf = "$BIND_ADDR"
|
|
}
|
|
|
|
ports {
|
|
http = 4646
|
|
rpc = 4647
|
|
serf = 4648
|
|
}
|
|
|
|
plugin "podman" {
|
|
config {
|
|
volumes {
|
|
enabled = true
|
|
}
|
|
}
|
|
}
|
|
|
|
telemetry {
|
|
collection_interval = "10s"
|
|
disable_hostname = false
|
|
prometheus_metrics = true
|
|
publish_allocation_metrics = true
|
|
publish_node_metrics = true
|
|
}
|
|
|
|
log_level = "INFO"
|
|
log_file = "/var/log/nomad/nomad.log"
|
|
EOF
|
|
|
|
# 创建 systemd 服务文件
|
|
log "创建 systemd 服务文件..."
|
|
cat > /etc/systemd/system/nomad.service << EOF
|
|
[Unit]
|
|
Description=Nomad
|
|
Documentation=https://www.nomadproject.io/
|
|
Requires=network-online.target
|
|
After=network-online.target
|
|
ConditionFileNotEmpty=/etc/nomad.d/nomad.hcl
|
|
|
|
[Service]
|
|
Type=notify
|
|
User=nomad
|
|
Group=nomad
|
|
ExecStart=/usr/local/bin/nomad agent -config=/etc/nomad.d/nomad.hcl
|
|
ExecReload=/bin/kill -HUP \$MAINPID
|
|
KillMode=process
|
|
Restart=on-failure
|
|
LimitNOFILE=65536
|
|
|
|
[Install]
|
|
WantedBy=multi-user.target
|
|
EOF
|
|
|
|
# 启动 Nomad 服务
|
|
log "启动 Nomad 服务..."
|
|
systemctl daemon-reload
|
|
systemctl enable nomad
|
|
systemctl start nomad
|
|
|
|
# 等待服务启动
|
|
log "等待 Nomad 服务启动..."
|
|
sleep 10
|
|
|
|
# 验证安装
|
|
log "验证 Nomad 安装..."
|
|
if systemctl is-active --quiet nomad; then
|
|
log "✅ Nomad 服务运行正常"
|
|
log "📊 节点信息:"
|
|
/usr/local/bin/nomad node status -self || true
|
|
else
|
|
log "❌ Nomad 服务启动失败"
|
|
systemctl status nomad --no-pager || true
|
|
journalctl -u nomad --no-pager -n 20 || true
|
|
fi
|
|
|
|
# 配置防火墙（如果需要）
|
|
log "配置防火墙规则..."
|
|
if command -v ufw >/dev/null 2>&1; then
|
|
ufw allow 4646/tcp # HTTP API
|
|
ufw allow 4647/tcp # RPC
|
|
ufw allow 4648/tcp # Serf
|
|
ufw allow 22/tcp # SSH
|
|
fi
|
|
|
|
# 创建有用的别名和脚本
|
|
log "创建管理脚本..."
|
|
cat > /usr/local/bin/nomad-status << 'EOF'
|
|
#!/bin/bash
|
|
echo "=== Nomad 服务状态 ==="
|
|
systemctl status nomad --no-pager
|
|
|
|
echo -e "\n=== Nomad 集群成员 ==="
|
|
nomad server members 2>/dev/null || echo "无法连接到集群"
|
|
|
|
echo -e "\n=== Nomad 节点状态 ==="
|
|
nomad node status 2>/dev/null || echo "无法获取节点状态"
|
|
|
|
echo -e "\n=== 最近日志 ==="
|
|
journalctl -u nomad --no-pager -n 5
|
|
EOF
|
|
|
|
chmod +x /usr/local/bin/nomad-status
|
|
|
|
# 添加到 ubuntu 用户的 bashrc
|
|
echo 'alias ns="nomad-status"' >> /home/ubuntu/.bashrc
|
|
echo 'alias nomad-logs="journalctl -u nomad -f"' >> /home/ubuntu/.bashrc
|
|
|
|
log "🎉 Nomad 节点配置完成！"
|
|
log "📍 数据中心: ${datacenter}"
|
|
log "🌐 IP 地址: $BIND_ADDR"
|
|
log "🔗 Web UI: http://$BIND_ADDR:4646"
|
|
log "📝 使用 'nomad-status' 或 'ns' 命令查看状态"
|
|
|
|
# 输出重要信息到 motd
|
|
cat > /etc/update-motd.d/99-nomad << EOF
|
|
#!/bin/bash
|
|
echo ""
|
|
echo "🚀 Nomad 节点信息:"
|
|
echo " 数据中心: ${datacenter}"
|
|
echo " IP 地址: $BIND_ADDR"
|
|
echo " Web UI: http://$BIND_ADDR:4646"
|
|
echo " 状态检查: nomad-status"
|
|
echo ""
|
|
EOF
|
|
|
|
chmod +x /etc/update-motd.d/99-nomad
|
|
|
|
log "节点配置脚本执行完成" |