dev_1.0.0_yuyr_2:重新提交 PR,增加 master/agent 以及系统集成测试 (#17)

Reviewed-on: #17
Reviewed-by: sundapeng <sundp@mail.zgclab.edu.cn>
Reviewed-by: xuxt <xuxt@zgclab.edu.cn>
This commit was merged in pull request #17.
This commit is contained in:
2025-10-11 15:04:46 +08:00
parent 8a38d3d0b2
commit 1e5e91b193
105 changed files with 6096 additions and 107 deletions

2
src/agent/tests/.gitignore vendored Normal file
View File

@@ -0,0 +1,2 @@
private/
tmp/

View File

@@ -0,0 +1,69 @@
services:
bind:
image: ${BIND_IMAGE_TAG:-argus-bind9:latest}
container_name: argus-bind-agent-e2e
volumes:
- ./private:/private
networks:
default:
ipv4_address: 172.28.0.2
environment:
- "ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}"
- "ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}"
restart: always
master:
image: argus-master:latest
container_name: argus-master-agent-e2e
depends_on:
- bind
environment:
- OFFLINE_THRESHOLD_SECONDS=6
- ONLINE_THRESHOLD_SECONDS=2
- SCHEDULER_INTERVAL_SECONDS=1
- "ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}"
- "ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}"
ports:
- "32300:3000"
volumes:
- ./private/argus/master:/private/argus/master
- ./private/argus/metric/prometheus:/private/argus/metric/prometheus
- ./private/argus/etc:/private/argus/etc
networks:
default:
ipv4_address: 172.28.0.10
restart: always
agent:
image: ubuntu:22.04
container_name: argus-agent-e2e
hostname: dev-e2euser-e2einst-pod-0
depends_on:
- master
- bind
environment:
- MASTER_ENDPOINT=http://master.argus.com:3000
- REPORT_INTERVAL_SECONDS=2
- "ARGUS_BUILD_UID=${ARGUS_BUILD_UID:-2133}"
- "ARGUS_BUILD_GID=${ARGUS_BUILD_GID:-2015}"
volumes:
- ./private/argus/agent/dev-e2euser-e2einst-pod-0:/private/argus/agent/dev-e2euser-e2einst-pod-0
- ./private/argus/agent/dev-e2euser-e2einst-pod-0/health:/private/argus/agent/dev-e2euser-e2einst-pod-0/health
- ./private/argus/etc:/private/argus/etc
- ../dist/argus-agent:/usr/local/bin/argus-agent:ro
- ./scripts/agent_entrypoint.sh:/usr/local/bin/agent-entrypoint.sh:ro
- ../scripts/agent_deployment_verify.sh:/usr/local/bin/agent_deployment_verify.sh:ro
entrypoint:
- /usr/local/bin/agent-entrypoint.sh
networks:
default:
ipv4_address: 172.28.0.20
restart: always
networks:
default:
driver: bridge
ipam:
driver: default
config:
- subnet: 172.28.0.0/16

View File

@@ -0,0 +1,23 @@
#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
SCRIPTS=(
"01_bootstrap.sh"
"02_up.sh"
"03_wait_and_assert_registration.sh"
"04_write_health_files.sh"
"08_verify_agent.sh"
"05_assert_status_on_master.sh"
"06_restart_agent_and_reregister.sh"
"07_down.sh"
)
for script in "${SCRIPTS[@]}"; do
echo "[TEST] Running $script"
"$SCRIPT_DIR/$script"
echo "[TEST] $script completed"
echo
done
echo "[TEST] Agent module E2E tests completed"

View File

@@ -0,0 +1,63 @@
#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
AGENT_ROOT="$(cd "$TEST_ROOT/.." && pwd)"
MASTER_ROOT="$(cd "$AGENT_ROOT/../master" && pwd)"
REPO_ROOT="$(cd "$AGENT_ROOT/../.." && pwd)"
PRIVATE_ROOT="$TEST_ROOT/private"
TMP_ROOT="$TEST_ROOT/tmp"
AGENT_HOSTNAME="dev-e2euser-e2einst-pod-0"
AGENT_CONFIG_DIR="$PRIVATE_ROOT/argus/agent/$AGENT_HOSTNAME"
AGENT_HEALTH_DIR="$PRIVATE_ROOT/argus/agent/$AGENT_HOSTNAME/health"
MASTER_PRIVATE_DIR="$PRIVATE_ROOT/argus/master"
METRIC_PRIVATE_DIR="$PRIVATE_ROOT/argus/metric/prometheus"
DNS_DIR="$PRIVATE_ROOT/argus/etc"
BIND_IMAGE_TAG="${BIND_IMAGE_TAG:-argus-bind9:latest}"
BIND_ROOT="$(cd "$MASTER_ROOT/../bind" && pwd)"
ensure_image() {
local image="$1"
if ! docker image inspect "$image" >/dev/null 2>&1; then
echo "[ERROR] Docker image '$image' 未找到,请先运行统一构建脚本 (例如 ./build/build_images.sh) 生成所需镜像" >&2
exit 1
fi
}
mkdir -p "$AGENT_CONFIG_DIR"
mkdir -p "$AGENT_HEALTH_DIR"
mkdir -p "$MASTER_PRIVATE_DIR"
mkdir -p "$METRIC_PRIVATE_DIR"
mkdir -p "$TMP_ROOT"
mkdir -p "$DNS_DIR"
touch "$AGENT_HEALTH_DIR/.keep"
# 中文提示:准备 bind 模块提供的 update-dns.sh,模拟生产下发
if [[ -f "$BIND_ROOT/build/update-dns.sh" ]]; then
cp "$BIND_ROOT/build/update-dns.sh" "$DNS_DIR/update-dns.sh"
chmod +x "$DNS_DIR/update-dns.sh"
else
echo "[WARN] bind update script missing at $BIND_ROOT/build/update-dns.sh"
fi
ensure_image "argus-master:latest"
ensure_image "$BIND_IMAGE_TAG"
AGENT_BINARY="$AGENT_ROOT/dist/argus-agent"
pushd "$AGENT_ROOT" >/dev/null
./scripts/build_binary.sh
popd >/dev/null
if [[ ! -x "$AGENT_BINARY" ]]; then
echo "[ERROR] Agent binary not found at $AGENT_BINARY" >&2
exit 1
fi
echo "$AGENT_BINARY" > "$TMP_ROOT/agent_binary_path"
echo "$BIND_IMAGE_TAG" > "$TMP_ROOT/bind_image_tag"
echo "[INFO] Agent E2E bootstrap complete"

View File

@@ -0,0 +1,53 @@
#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
REPO_ROOT="$(cd "$TEST_ROOT/../../.." && pwd)"
TMP_ROOT="$TEST_ROOT/tmp"
ENV_FILE="$TEST_ROOT/.env"
source "$REPO_ROOT/scripts/common/build_user.sh"
load_build_user
export ARGUS_BUILD_UID ARGUS_BUILD_GID
cat > "$ENV_FILE" <<EOF
ARGUS_BUILD_UID=$ARGUS_BUILD_UID
ARGUS_BUILD_GID=$ARGUS_BUILD_GID
EOF
if [[ ! -f "$TMP_ROOT/agent_binary_path" ]]; then
echo "[ERROR] Agent binary path missing; run 01_bootstrap.sh first" >&2
exit 1
fi
AGENT_BINARY="$(cat "$TMP_ROOT/agent_binary_path")"
if [[ ! -x "$AGENT_BINARY" ]]; then
echo "[ERROR] Agent binary not executable: $AGENT_BINARY" >&2
exit 1
fi
BIND_IMAGE_TAG_VALUE="argus-bind9:latest"
if [[ -f "$TMP_ROOT/bind_image_tag" ]]; then
BIND_IMAGE_TAG_VALUE="$(cat "$TMP_ROOT/bind_image_tag")"
fi
compose() {
if docker compose version >/dev/null 2>&1; then
docker compose "$@"
else
docker-compose "$@"
fi
}
docker container rm -f argus-agent-e2e argus-master-agent-e2e argus-bind-agent-e2e >/dev/null 2>&1 || true
docker network rm tests_default >/dev/null 2>&1 || true
pushd "$TEST_ROOT" >/dev/null
compose down --remove-orphans || true
BIND_IMAGE_TAG="$BIND_IMAGE_TAG_VALUE" compose up -d
popd >/dev/null
echo "[INFO] Master+Agent stack started"

View File

@@ -0,0 +1,65 @@
#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
TMP_ROOT="$TEST_ROOT/tmp"
API_BASE="http://localhost:32300/api/v1/master"
AGENT_HOSTNAME="dev-e2euser-e2einst-pod-0"
NODE_FILE="$TEST_ROOT/private/argus/agent/$AGENT_HOSTNAME/node.json"
mkdir -p "$TMP_ROOT"
node_id=""
for _ in {1..30}; do
sleep 2
response=$(curl -sS "$API_BASE/nodes" || true)
if [[ -z "$response" ]]; then
continue
fi
list_file="$TMP_ROOT/nodes_list.json"
echo "$response" > "$list_file"
node_id=$(python3 - "$list_file" <<'PY'
import json, sys
with open(sys.argv[1]) as handle:
nodes = json.load(handle)
print(nodes[0]["id"] if nodes else "")
PY
)
if [[ -n "$node_id" ]]; then
break
fi
done
if [[ -z "$node_id" ]]; then
echo "[ERROR] Agent did not register within timeout" >&2
exit 1
fi
echo "$node_id" > "$TMP_ROOT/node_id"
if [[ ! -f "$NODE_FILE" ]]; then
echo "[ERROR] node.json not created at $NODE_FILE" >&2
exit 1
fi
python3 - "$NODE_FILE" <<'PY'
import json, sys
with open(sys.argv[1]) as handle:
node = json.load(handle)
assert "id" in node and node["id"], "node.json missing id"
PY
detail_file="$TMP_ROOT/initial_detail.json"
curl -sS "$API_BASE/nodes/$node_id" -o "$detail_file"
python3 - "$detail_file" "$TMP_ROOT/initial_ip" <<'PY'
import json, sys, pathlib
with open(sys.argv[1]) as handle:
node = json.load(handle)
ip = node["meta_data"].get("ip")
if not ip:
raise SystemExit("meta_data.ip missing")
pathlib.Path(sys.argv[2]).write_text(ip)
PY
echo "[INFO] Agent registered with node id $node_id"

View File

@@ -0,0 +1,22 @@
#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
HEALTH_DIR="$TEST_ROOT/private/argus/agent/dev-e2euser-e2einst-pod-0/health"
cat > "$HEALTH_DIR/log-fluentbit.json" <<JSON
{
"status": "healthy",
"timestamp": "2023-10-05T12:05:00Z"
}
JSON
cat > "$HEALTH_DIR/metric-node-exporter.json" <<JSON
{
"status": "healthy",
"timestamp": "2023-10-05T12:05:00Z"
}
JSON
echo "[INFO] Health files written"

View File

@@ -0,0 +1,53 @@
#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
TMP_ROOT="$TEST_ROOT/tmp"
API_BASE="http://localhost:32300/api/v1/master"
NODE_ID="$(cat "$TMP_ROOT/node_id")"
NODES_JSON="$TEST_ROOT/private/argus/metric/prometheus/nodes.json"
success=false
detail_file="$TMP_ROOT/agent_status_detail.json"
for _ in {1..20}; do
sleep 2
if ! curl -sS "$API_BASE/nodes/$NODE_ID" -o "$detail_file"; then
continue
fi
if python3 - "$detail_file" <<'PY'
import json, sys
with open(sys.argv[1]) as handle:
node = json.load(handle)
if node["status"] != "online":
raise SystemExit(1)
health = node.get("health", {})
if "log-fluentbit" not in health or "metric-node-exporter" not in health:
raise SystemExit(1)
PY
then
success=true
break
fi
done
if [[ "$success" != true ]]; then
echo "[ERROR] Node did not report health data in time" >&2
exit 1
fi
if [[ ! -f "$NODES_JSON" ]]; then
echo "[ERROR] nodes.json missing at $NODES_JSON" >&2
exit 1
fi
python3 - "$NODES_JSON" <<'PY'
import json, sys
with open(sys.argv[1]) as handle:
nodes = json.load(handle)
assert len(nodes) == 1, nodes
entry = nodes[0]
assert entry["node_id"], entry
PY
echo "[INFO] Master reflects agent health and nodes.json entries"

View File

@@ -0,0 +1,143 @@
#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
TMP_ROOT="$TEST_ROOT/tmp"
API_BASE="http://localhost:32300/api/v1/master"
NODE_ID="$(cat "$TMP_ROOT/node_id")"
AGENT_HOSTNAME="dev-e2euser-e2einst-pod-0"
NETWORK_NAME="tests_default"
NEW_AGENT_IP="172.28.0.200"
ENTRYPOINT_SCRIPT="$SCRIPT_DIR/agent_entrypoint.sh"
ENV_FILE="$TEST_ROOT/.env"
# 中文提示:重启场景也需要同样的入口脚本,确保 DNS 注册逻辑一致
if [[ ! -f "$ENTRYPOINT_SCRIPT" ]]; then
echo "[ERROR] agent entrypoint script missing at $ENTRYPOINT_SCRIPT" >&2
exit 1
fi
if [[ ! -f "$TMP_ROOT/agent_binary_path" ]]; then
echo "[ERROR] Agent binary path missing; rerun bootstrap" >&2
exit 1
fi
AGENT_BINARY="$(cat "$TMP_ROOT/agent_binary_path")"
if [[ ! -x "$AGENT_BINARY" ]]; then
echo "[ERROR] Agent binary not executable: $AGENT_BINARY" >&2
exit 1
fi
if [[ -f "$ENV_FILE" ]]; then
set -a
# shellcheck disable=SC1090
source "$ENV_FILE"
set +a
else
REPO_ROOT="$(cd "$TEST_ROOT/../../.." && pwd)"
# shellcheck disable=SC1090
source "$REPO_ROOT/scripts/common/build_user.sh"
load_build_user
fi
AGENT_UID="${ARGUS_BUILD_UID:-2133}"
AGENT_GID="${ARGUS_BUILD_GID:-2015}"
compose() {
if docker compose version >/dev/null 2>&1; then
docker compose "$@"
else
docker-compose "$@"
fi
}
before_file="$TMP_ROOT/before_restart.json"
curl -sS "$API_BASE/nodes/$NODE_ID" -o "$before_file"
prev_last_updated=$(python3 - "$before_file" <<'PY'
import json, sys
with open(sys.argv[1]) as handle:
node = json.load(handle)
print(node.get("last_updated", ""))
PY
)
prev_ip=$(python3 - "$before_file" <<'PY'
import json, sys
with open(sys.argv[1]) as handle:
node = json.load(handle)
print(node["meta_data"].get("ip", ""))
PY
)
initial_ip=$(cat "$TMP_ROOT/initial_ip")
if [[ "$prev_ip" != "$initial_ip" ]]; then
echo "[ERROR] Expected initial IP $initial_ip, got $prev_ip" >&2
exit 1
fi
pushd "$TEST_ROOT" >/dev/null
compose rm -sf agent
popd >/dev/null
docker container rm -f argus-agent-e2e >/dev/null 2>&1 || true
AGENT_DIR="$TEST_ROOT/private/argus/agent/$AGENT_HOSTNAME"
HEALTH_DIR="$TEST_ROOT/private/argus/agent/$AGENT_HOSTNAME/health"
# 先以 sleep 方式启动容器,确保我们掌握注册时的网络状态
if ! docker run -d \
--name argus-agent-e2e \
--hostname "$AGENT_HOSTNAME" \
--network "$NETWORK_NAME" \
--ip "$NEW_AGENT_IP" \
-v "$AGENT_DIR:/private/argus/agent/$AGENT_HOSTNAME" \
-v "$HEALTH_DIR:/private/argus/agent/$AGENT_HOSTNAME/health" \
-v "$TEST_ROOT/private/argus/etc:/private/argus/etc" \
-v "$AGENT_BINARY:/usr/local/bin/argus-agent:ro" \
-v "$ENTRYPOINT_SCRIPT:/usr/local/bin/agent-entrypoint.sh:ro" \
-e MASTER_ENDPOINT=http://master.argus.com:3000 \
-e REPORT_INTERVAL_SECONDS=2 \
-e ARGUS_BUILD_UID="$AGENT_UID" \
-e ARGUS_BUILD_GID="$AGENT_GID" \
--entrypoint /usr/local/bin/agent-entrypoint.sh \
ubuntu:22.04 >/dev/null; then
echo "[ERROR] Failed to start agent container with custom IP" >&2
exit 1
fi
success=false
detail_file="$TMP_ROOT/post_restart.json"
for _ in {1..20}; do
sleep 3
if ! curl -sS "$API_BASE/nodes/$NODE_ID" -o "$detail_file"; then
continue
fi
if python3 - "$detail_file" "$prev_last_updated" "$NODE_ID" "$prev_ip" "$NEW_AGENT_IP" <<'PY'
import json, sys
with open(sys.argv[1]) as handle:
node = json.load(handle)
prev_last_updated = sys.argv[2]
expected_id = sys.argv[3]
old_ip = sys.argv[4]
expected_ip = sys.argv[5]
last_updated = node.get("last_updated")
current_ip = node["meta_data"].get("ip")
assert node["id"] == expected_id
if current_ip != expected_ip:
raise SystemExit(1)
if current_ip == old_ip:
raise SystemExit(1)
if not last_updated or last_updated == prev_last_updated:
raise SystemExit(1)
PY
then
success=true
break
fi
done
if [[ "$success" != true ]]; then
echo "[ERROR] Agent did not report expected new IP $NEW_AGENT_IP after restart" >&2
exit 1
fi
echo "[INFO] Agent restart produced successful re-registration with IP change"

View File

@@ -0,0 +1,36 @@
#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
ENV_FILE="$TEST_ROOT/.env"
compose() {
if docker compose version >/dev/null 2>&1; then
docker compose "$@"
else
docker-compose "$@"
fi
}
docker container rm -f argus-agent-e2e >/dev/null 2>&1 || true
pushd "$TEST_ROOT" >/dev/null
compose down --remove-orphans
popd >/dev/null
if [[ -d "$TEST_ROOT/private" ]]; then
docker run --rm \
-v "$TEST_ROOT/private:/target" \
ubuntu:24.04 \
chown -R "$(id -u):$(id -g)" /target >/dev/null 2>&1 || true
rm -rf "$TEST_ROOT/private"
fi
rm -rf "$TEST_ROOT/tmp"
if [[ -f "$ENV_FILE" ]]; then
rm -f "$ENV_FILE"
fi
echo "[INFO] Agent E2E environment cleaned up"

View File

@@ -0,0 +1,26 @@
#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
TEST_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
VERIFY_SCRIPT="$(cd "$TEST_ROOT/.." && pwd)/scripts/agent_deployment_verify.sh"
if ! docker ps --format '{{.Names}}' | grep -q '^argus-agent-e2e$'; then
echo "[WARN] agent container not running; skip verification"
exit 0
fi
if docker exec -i argus-agent-e2e bash -lc 'command -v curl >/dev/null && command -v jq >/dev/null'; then
echo "[INFO] curl/jq already installed in agent container"
else
echo "[INFO] Installing curl/jq in agent container"
docker exec -i argus-agent-e2e bash -lc 'apt-get update >/dev/null 2>&1 && apt-get install -y curl jq >/dev/null 2>&1' || true
fi
if docker exec -i argus-agent-e2e bash -lc 'command -v /usr/local/bin/agent_deployment_verify.sh >/dev/null'; then
docker exec -i argus-agent-e2e /usr/local/bin/agent_deployment_verify.sh
elif [[ -x "$VERIFY_SCRIPT" ]]; then
docker exec -i argus-agent-e2e "$VERIFY_SCRIPT"
else
echo "[WARN] agent_deployment_verify.sh not found"
fi

View File

@@ -0,0 +1,79 @@
#!/usr/bin/env bash
set -euo pipefail
LOG_PREFIX="[AGENT-ENTRYPOINT]"
DNS_SCRIPT="/private/argus/etc/update-dns.sh"
DNS_CONF="/private/argus/etc/dns.conf"
TARGET_DOMAIN="master.argus.com"
AGENT_UID="${ARGUS_BUILD_UID:-2133}"
AGENT_GID="${ARGUS_BUILD_GID:-2015}"
AGENT_HOSTNAME="${HOSTNAME:-unknown}"
AGENT_DATA_DIR="/private/argus/agent/${AGENT_HOSTNAME}"
AGENT_HEALTH_DIR="${AGENT_DATA_DIR}/health"
RUNTIME_GROUP="argusagent"
RUNTIME_USER="argusagent"
log() {
echo "${LOG_PREFIX} $*"
}
mkdir -p "$AGENT_DATA_DIR" "$AGENT_HEALTH_DIR"
chown -R "$AGENT_UID:$AGENT_GID" "$AGENT_DATA_DIR" "$AGENT_HEALTH_DIR" 2>/dev/null || true
chown -R "$AGENT_UID:$AGENT_GID" "/private/argus/etc" 2>/dev/null || true
if ! getent group "$AGENT_GID" >/dev/null 2>&1; then
groupadd -g "$AGENT_GID" "$RUNTIME_GROUP"
else
RUNTIME_GROUP="$(getent group "$AGENT_GID" | cut -d: -f1)"
fi
if ! getent passwd "$AGENT_UID" >/dev/null 2>&1; then
useradd -u "$AGENT_UID" -g "$AGENT_GID" -M -s /bin/bash "$RUNTIME_USER"
else
RUNTIME_USER="$(getent passwd "$AGENT_UID" | cut -d: -f1)"
fi
log "运行用户: $RUNTIME_USER ($AGENT_UID:$AGENT_GID)"
# 中文提示:等待 bind 下发的 update-dns.sh 脚本
for _ in {1..30}; do
if [[ -x "$DNS_SCRIPT" ]]; then
break
fi
log "等待 update-dns.sh 准备就绪..."
sleep 1
done
if [[ -x "$DNS_SCRIPT" ]]; then
log "执行 update-dns.sh 更新容器 DNS"
while true; do
if "$DNS_SCRIPT"; then
log "update-dns.sh 执行成功"
break
fi
log "update-dns.sh 执行失败,3 秒后重试"
sleep 3
done
else
log "未获取到 update-dns.sh,使用镜像默认 DNS"
fi
# 中文提示:记录当前 dns.conf 内容,便于排查
if [[ -f "$DNS_CONF" ]]; then
log "dns.conf 内容: $(tr '\n' ' ' < "$DNS_CONF")"
else
log "dns.conf 暂未生成"
fi
# 中文提示:尝试解析 master 域名,失败不阻塞但会打日志
for _ in {1..30}; do
if getent hosts "$TARGET_DOMAIN" >/dev/null 2>&1; then
MASTER_IP=$(getent hosts "$TARGET_DOMAIN" | awk '{print $1}' | head -n 1)
log "master.argus.com 解析成功: $MASTER_IP"
break
fi
sleep 1
done
log "启动 argus-agent"
exec su -s /bin/bash -c /usr/local/bin/argus-agent "$RUNTIME_USER"