Copilot commented on code in PR #3044: URL: https://github.com/apache/hugegraph/pull/3044#discussion_r3330297571
########## hugegraph-server/hugegraph-dist/src/assembly/travis/test-start-hugegraph.sh: ########## @@ -0,0 +1,379 @@ +#!/bin/bash +# +# Licensed to the Apache Software Foundation (ASF) under one or more +# contributor license agreements. See the NOTICE file distributed with +# this work for additional information regarding copyright ownership. +# The ASF licenses this file to You under the Apache License, Version 2.0 +# (the "License"); you may not use this file except in compliance with +# the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# test-start-hugegraph.sh — Tests for start-hugegraph.sh foreground mode fix +# +# Baseline (unmodified code): Tests 1, 2, 4 PASS — Test 3 FAILS +# After chunk 1 fix: All 4 tests PASS +# +# Usage: ./test-start-hugegraph.sh [path-to-hugegraph-root] +# path-to-hugegraph-root: path to the extracted HugeGraph server dist directory +# defaults to current directory if not provided +# +# Note: Tests require crontab access. On macOS Catalina+ this may +# require Full Disk Access permission in System Preferences. +# Note: Assumes rocksdb backend. init-store.sh is run automatically if needed. + +set -uo pipefail + +HUGEGRAPH_ROOT="${1:-$(pwd)}" +BIN="$HUGEGRAPH_ROOT/bin" +START_SCRIPT="$BIN/start-hugegraph.sh" +STOP_SCRIPT="$BIN/stop-hugegraph.sh" +PID_FILE="$BIN/pid" +SERVER_URL="http://127.0.0.1:8080"; +STARTUP_WAIT=60 # seconds to wait for server HTTP to respond +SETTLE_WAIT=5 # seconds after kill before checking process exit +WAIT_TIMEOUT=30 # seconds for timeout on wait calls +PASS=0 +FAIL=0 +ERRORS=() + +GREEN='\033[0;32m' +RED='\033[0;31m' +YELLOW='\033[1;33m' +NC='\033[0m' + +pass() { + echo -e "${GREEN} PASS${NC} $1" + PASS=$((PASS + 1)) +} + +fail() { + echo -e "${RED} FAIL${NC} $1" + ERRORS+=("$1") + FAIL=$((FAIL + 1)) +} + +info() { + echo -e "${YELLOW} ....${NC} $1" +} + +section() { + echo "" + echo "── $1 ──" +} + +# Kill server, clear pid file, kill ports, clear crontab monitor entry +cleanup() { + info "Cleaning up..." + "$STOP_SCRIPT" >/dev/null 2>&1 || true + if [[ -s "$PID_FILE" ]]; then + kill "$(cat "$PID_FILE")" 2>/dev/null || true + fi + rm -f "$PID_FILE" + # kill anything still holding server ports so check_port doesn't fail next test + lsof -ti :8080 | xargs kill -9 2>/dev/null || true + lsof -ti :8182 | xargs kill -9 2>/dev/null || true + lsof -ti :8088 | xargs kill -9 2>/dev/null || true + sleep 3 + crontab -l 2>/dev/null | grep -v monitor-hugegraph | crontab - 2>/dev/null || true +} + +# Wait until server HTTP endpoint responds (200 or 401) or timeout +wait_for_server() { + local elapsed=0 + while (( elapsed < STARTUP_WAIT )); do + local status + status=$(curl -s -o /dev/null -w "%{http_code}" \ + "$SERVER_URL/versions" 2>/dev/null || echo "000") + if [[ "$status" == "200" || "$status" == "401" ]]; then + return 0 + fi + sleep 2 + elapsed=$((elapsed + 2)) + done + return 1 +} + +# Wait until bin/pid is non-empty or timeout +wait_for_pid_file() { + local elapsed=0 + while [[ ! -s "$PID_FILE" ]] && (( elapsed < 30 )); do + sleep 1 + elapsed=$((elapsed + 1)) + done +} + +# Wait for a background script PID to exit, with timeout +# Usage: wait_script_exit <script_pid> +# Returns the real exit code of the process, or kills it after WAIT_TIMEOUT seconds +# NOTE: must be called from the same shell that spawned script_pid so that +# bash's built-in wait can track it as a child process +wait_script_exit() { + local script_pid="$1" + # background killer fires after timeout if process hasn't exited yet + ( sleep "$WAIT_TIMEOUT" && kill "$script_pid" 2>/dev/null ) & + local killer_pid=$! + wait "$script_pid" 2>/dev/null + local exit_code=$? + # cancel the killer if process already exited + kill "$killer_pid" 2>/dev/null || true + wait "$killer_pid" 2>/dev/null || true + return $exit_code +} + +# ── preflight ───────────────────────────────────────────────────────────────── + +echo "" +echo "start-hugegraph.sh chunk 1 test suite" +echo "root: $HUGEGRAPH_ROOT" +echo "" + +if [[ ! -f "$START_SCRIPT" ]]; then + echo -e "${RED}ERROR:${NC} $START_SCRIPT not found." + echo " Pass the HugeGraph dist root as \$1" + exit 1 +fi + +if [[ ! -f "$STOP_SCRIPT" ]]; then + echo -e "${RED}ERROR:${NC} $STOP_SCRIPT not found." + exit 1 +fi + +# start-monitor.sh requires JAVA_HOME +if [[ -z "${JAVA_HOME:-}" ]]; then + if command -v /usr/libexec/java_home >/dev/null 2>&1; then + export JAVA_HOME="$(/usr/libexec/java_home 2>/dev/null)" + fi +fi + +# Warn if JAVA_HOME could not be determined +if [[ -z "${JAVA_HOME:-}" ]]; then + echo -e "${YELLOW}WARN${NC} JAVA_HOME is not set" + echo " Test 4 may fail because start-monitor.sh requires JAVA_HOME" + echo " Export JAVA_HOME before running this script" +fi + +# Run init-store.sh once if RocksDB has not been initialized yet +if [[ ! -d "$HUGEGRAPH_ROOT/rocksdb-data" ]]; then + info "RocksDB not initialized — running init-store.sh..." + if "$BIN/init-store.sh" >/dev/null 2>&1; then + pass "init-store.sh completed successfully" + else + echo -e "${RED}ERROR:${NC} init-store.sh failed. Cannot run tests." + exit 1 + fi +else + info "RocksDB already initialized, skipping init-store.sh" +fi + +cleanup + +# ── test 1: daemon mode regression ─────────────────────────────────────────── + +section "Test 1 — daemon mode regression" + +info "Starting in daemon mode (waiting up to ${STARTUP_WAIT}s for HTTP)..." +"$START_SCRIPT" -d true -t "$STARTUP_WAIT" >/dev/null 2>&1 +EXIT_CODE=$? + +if [[ $EXIT_CODE -eq 0 ]]; then + pass "script returned exit 0 in daemon mode" +else + fail "script returned non-zero exit $EXIT_CODE in daemon mode" +fi + +wait_for_pid_file +DAEMON_PID=$(cat "$PID_FILE" 2>/dev/null || echo "") + +if [[ -n "$DAEMON_PID" ]]; then + pass "bin/pid written with PID $DAEMON_PID" +else + fail "bin/pid is empty or missing after daemon start" +fi + +if [[ -n "$DAEMON_PID" ]] && ps -p "$DAEMON_PID" >/dev/null 2>&1; then + pass "Java process is running (pid $DAEMON_PID)" +else + fail "Java process not found for pid '$DAEMON_PID'" +fi + +if wait_for_server; then + pass "server HTTP endpoint responding at $SERVER_URL" +else + fail "server HTTP endpoint not responding after ${STARTUP_WAIT}s" +fi + +info "Stopping server..." +"$STOP_SCRIPT" >/dev/null 2>&1 +sleep 3 + +if [[ -n "$DAEMON_PID" ]] && ! ps -p "$DAEMON_PID" >/dev/null 2>&1; then + pass "stop-hugegraph.sh successfully killed the process" +else + fail "process still running after stop-hugegraph.sh" +fi + +cleanup + +# ── test 2: foreground mode blocks ─────────────────────────────────────────── + +section "Test 2 — foreground mode blocks until Java exits" + +info "Starting in foreground mode (backgrounded for observation)..." +"$START_SCRIPT" -d false >/dev/null 2>&1 & +SCRIPT_PID=$! + +info "Waiting up to ${STARTUP_WAIT}s for server to come up..." +if wait_for_server; then + info "Server is up" +else + info "Server did not respond — continuing to check blocking behavior" +fi + +# Primary assertion: script must still be running while Java is alive +if ps -p "$SCRIPT_PID" >/dev/null 2>&1; then + pass "script is still running (blocking correctly)" +else + fail "script exited early — foreground mode is not blocking" +fi + +# Secondary observations only (the actual bug is tested in Test 3) +wait_for_pid_file +FG_PID=$(cat "$PID_FILE" 2>/dev/null || echo "") + +if [[ -n "$FG_PID" ]]; then + info "pid file written with PID $FG_PID" +else + info "pid file not written in foreground mode (expected before chunk 1 fix)" +fi + +if [[ -n "$FG_PID" ]] && ps -p "$FG_PID" >/dev/null 2>&1; then + info "Java process running (pid $FG_PID)" +else + info "Java PID not available (expected before chunk 1 fix)" +fi + +if [[ -n "$FG_PID" ]]; then + info "Killing Java process..." + kill "$FG_PID" 2>/dev/null || true + sleep "$SETTLE_WAIT" + + if ! ps -p "$SCRIPT_PID" >/dev/null 2>&1; then + info "script exited after Java was killed" + else + info "script still running after Java was killed" + kill "$SCRIPT_PID" 2>/dev/null || true + fi +else + kill "$SCRIPT_PID" 2>/dev/null || true +fi + +cleanup + +# ── test 3: exit code propagates ───────────────────────────────────────────── + +section "Test 3 — foreground PID tracking and exit propagation" + +info "Starting in foreground mode..." +"$START_SCRIPT" -d false >/dev/null 2>&1 & +SCRIPT_PID=$! + +info "Waiting up to ${STARTUP_WAIT}s for server..." +if wait_for_server; then + info "Server is up" +else + info "Server did not respond — continuing" +fi + +wait_for_pid_file +FG_PID=$(cat "$PID_FILE" 2>/dev/null || echo "") + +if [[ -n "$FG_PID" ]]; then + pass "bin/pid written with PID $FG_PID" +else + fail "bin/pid is empty or missing in foreground mode" +fi + +if [[ -n "$FG_PID" ]] && ps -p "$FG_PID" >/dev/null 2>&1; then + pass "Java process running (pid $FG_PID)" +else + fail "Java process not found for pid '$FG_PID'" +fi + +if [[ -z "$FG_PID" ]]; then + fail "could not get PID from pid file — skipping exit code check" + kill "$SCRIPT_PID" 2>/dev/null || true +else + info "Hard-killing Java with SIGKILL (pid $FG_PID)..." + kill -9 "$FG_PID" 2>/dev/null || true + sleep "$SETTLE_WAIT" + + wait_script_exit "$SCRIPT_PID" + ACTUAL_EXIT=$? + + if [[ $ACTUAL_EXIT -ne 0 ]]; then + pass "script exited non-zero ($ACTUAL_EXIT) after SIGKILL — Docker restart will fire" + else + fail "script exited 0 after SIGKILL — Docker would NOT restart (exit code lost)" + fi + + if [[ $ACTUAL_EXIT -eq 137 ]]; then + pass "exit code is 137 (128+9 for SIGKILL) — correctly propagated" + elif [[ $ACTUAL_EXIT -ne 0 ]]; then + info "exit code was $ACTUAL_EXIT (not 137 — may be shell-wrapped, acceptable if non-zero)" + fi +fi + +cleanup + +# ── test 4: monitor registers in daemon mode ────────────────────────────────── + +section "Test 4 — -m true registers cron job in daemon mode" + +info "Clearing crontab..." +crontab -r 2>/dev/null || true Review Comment: This clears the entire user's crontab before the monitor test. The script is documented for local use as well as CI, so running it on a developer machine would delete unrelated scheduled jobs. Remove only the HugeGraph monitor entry, as `cleanup()` already does, before asserting that `-m true` re-adds it. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
