Node.js includes a command-line debugging utility. The Node.js debugger client\nis not a full-featured debugger, but simple stepping and inspection are\npossible.
\nTo use it, start Node.js with the inspect argument followed by the path to the\nscript to debug.
$ node inspect myscript.js\n< Debugger listening on ws://127.0.0.1:9229/621111f9-ffcb-4e82-b718-48a145fa5db8\n< For help, see: https://nodejs.org/learn/getting-started/debugging\n<\nconnecting to 127.0.0.1:9229 ... ok\n< Debugger attached.\n<\n ok\nBreak on start in myscript.js:2\n 1 // myscript.js\n> 2 global.x = 5;\n 3 setTimeout(() => {\n 4 debugger;\ndebug>\nThe debugger automatically breaks on the first executable line. To instead\nrun until the first breakpoint (specified by a debugger statement), set\nthe NODE_INSPECT_RESUME_ON_START environment variable to 1.
$ cat myscript.js\n// myscript.js\nglobal.x = 5;\nsetTimeout(() => {\n debugger;\n console.log('world');\n}, 1000);\nconsole.log('hello');\n$ NODE_INSPECT_RESUME_ON_START=1 node inspect myscript.js\n< Debugger listening on ws://127.0.0.1:9229/f1ed133e-7876-495b-83ae-c32c6fc319c2\n< For help, see: https://nodejs.org/learn/getting-started/debugging\n<\nconnecting to 127.0.0.1:9229 ... ok\n< Debugger attached.\n<\n< hello\n<\nbreak in myscript.js:4\n 2 global.x = 5;\n 3 setTimeout(() => {\n> 4 debugger;\n 5 console.log('world');\n 6 }, 1000);\ndebug> next\nbreak in myscript.js:5\n 3 setTimeout(() => {\n 4 debugger;\n> 5 console.log('world');\n 6 }, 1000);\n 7 console.log('hello');\ndebug> repl\nPress Ctrl+C to leave debug repl\n> x\n5\n> 2 + 2\n4\ndebug> next\n< world\n<\nbreak in myscript.js:6\n 4 debugger;\n 5 console.log('world');\n> 6 }, 1000);\n 7 console.log('hello');\n 8\ndebug> .exit\n$\nThe repl command allows code to be evaluated remotely. The next command\nsteps to the next line. Type help to see what other commands are available.
Pressing enter without typing a command will repeat the previous debugger\ncommand.
node inspect supports a non-interactive probe mode for inspecting runtime values\nin an application via the flag --probe. Probe mode launches the application,\nsets one or more source breakpoints, evaluates one expression whenever a\nmatching breakpoint is hit, and prints one final report when the session ends\n(either on normal completion or timeout). This allows developers to perform\nprintf-style debugging without having to modify the application code and\nclean up afterwards, and it supports structured output for tool use.
$ node inspect [--json] [--preview] [--timeout=<ms>] [--port=<port>] \\\n --probe app.js:10 --expr 'x' \\\n [--probe app.js:20 --expr 'y' ...] \\\n [--] [<node-option> ...] <script.js> [args...]\n--probe <file>:<line>[:<col>]: Source location to probe. Line and column number\nare 1-based.--timeout=<ms>: A global wall-clock deadline for the entire probe session.\nThe default is 30000. This can be used to probe a long-running application\nthat can be terminated externally.--json: If used, prints a structured JSON report instead of the default text report.--preview: If used, non-primitive values will include CDP property previews for\nobject-like JSON probe values.--port=<port>: Selects the local inspector port used for the --inspect-brk\nlaunch path. Probe mode defaults to 0, which requests a random port.-- is optional unless the child needs its own Node.js flags.Additional rules about the --probe and --expr arguments:
--probe <file>:<line>[:<col>] and --expr <expr> are strict pairs. Each\n--probe must be followed immediately by exactly one --expr.--timeout, --json, --preview, and --port are global probe options\nfor the whole probe session. They may appear before or between probe pairs,\nbut not between a --probe and its matching --expr.If a single probe needs to evaluate more than one value,\nevaluate a structured value in --expr, for example --expr \"{ foo, bar }\"\nor --expr \"[foo, bar]\", and use --preview to include property previews for\nany object-like values in the output.
Probe mode only prints the final probe report to stdout, and otherwise silences\nstdout/stderr from the child process. If the child exits with an error after the\nprobe session starts, the final report records a terminal error event with the\nexit code and captured child stderr. Invalid arguments and fatal launch or\nconnect failures may still print diagnostics to stderr without a final probe\nresult.
Consider this script:
\n// cli.js\nlet maxRSS = 0;\nfor (let i = 0; i < 2; i++) {\n const { rss } = process.memoryUsage();\n maxRSS = Math.max(maxRSS, rss);\n}\nIf --json is not used, the output is printed in a human-readable text format:
$ node inspect --probe cli.js:5 --expr 'rss' cli.js\nHit 1 at cli.js:5\n rss = 54935552\nHit 2 at cli.js:5\n rss = 55083008\nCompleted\nPrimitive results are printed directly, while objects and arrays use Chrome\nDevTools Protocol preview data when available. Other non-primitive values\nfall back to the Chrome DevTools Protocol description string.\nExpression failures are recorded as [error] ... lines and do not fail\nthe overall session. If richer text formatting is needed, wrap the expression\nin JSON.stringify(...) or util.inspect(...).
When --json is used, the output shape looks like this:
$ node inspect --json --probe cli.js:5 --expr 'rss' cli.js\n{\"v\":1,\"probes\":[{\"expr\":\"rss\",\"target\":[\"cli.js\",5]}],\"results\":[{\"probe\":0,\"event\":\"hit\",\"hit\":1,\"result\":{\"type\":\"number\",\"value\":55443456,\"description\":\"55443456\"}},{\"probe\":0,\"event\":\"hit\",\"hit\":2,\"result\":{\"type\":\"number\",\"value\":55574528,\"description\":\"55574528\"}},{\"event\":\"completed\"}]}\n{\n \"v\": 1, // Probe JSON schema version.\n \"probes\": [\n {\n \"expr\": \"rss\", // The expression paired with --probe.\n \"target\": [\"cli.js\", 5] // [file, line] or [file, line, col].\n }\n ],\n \"results\": [\n {\n \"probe\": 0, // Index into probes[].\n \"event\": \"hit\", // Hit events are recorded in observation order.\n \"hit\": 1, // 1-based hit count for this probe.\n \"result\": {\n \"type\": \"number\",\n \"value\": 55443456,\n \"description\": \"55443456\"\n }\n // If the expression throws, \"error\" is present instead of \"result\".\n },\n {\n \"probe\": 0,\n \"event\": \"hit\",\n \"hit\": 2,\n \"result\": {\n \"type\": \"number\",\n \"value\": 55574528,\n \"description\": \"55574528\"\n }\n },\n {\n \"event\": \"completed\"\n // The final entry is always a terminal event, for example:\n // 1. { \"event\": \"completed\" }\n // 2. { \"event\": \"miss\", \"pending\": [0, 1] }\n // 3. {\n // \"event\": \"timeout\",\n // \"pending\": [0],\n // \"error\": {\n // \"code\": \"probe_timeout\",\n // \"message\": \"Timed out after 30000ms waiting for probes: app.js:10\"\n // }\n // }\n // 4. {\n // \"event\": \"error\",\n // \"pending\": [0],\n // \"error\": {\n // \"code\": \"probe_target_exit\",\n // \"exitCode\": 1,\n // \"stderr\": \"[Error: boom]\",\n // \"message\": \"Target exited with code 1 before probes: app.js:10\"\n // }\n // }\n }\n ]\n}\nIt is possible to watch expression and variable values while debugging. On\nevery breakpoint, each expression from the watchers list will be evaluated\nin the current context and displayed immediately before the breakpoint's\nsource code listing.
\nTo begin watching an expression, type watch('my_expression'). The command\nwatchers will print the active watchers. To remove a watcher, type\nunwatch('my_expression').
cont, c: Continue executionnext, n: Step nextstep, s: Step inout, o: Step outpause: Pause running code (like pause button in Developer Tools)setBreakpoint(), sb(): Set breakpoint on current linesetBreakpoint(line), sb(line): Set breakpoint on specific linesetBreakpoint('fn()'), sb(...): Set breakpoint on a first statement in\nfunction's bodysetBreakpoint('script.js', 1), sb(...): Set breakpoint on first line of\nscript.jssetBreakpoint('script.js', 1, 'num < 4'), sb(...): Set conditional\nbreakpoint on first line of script.js that only breaks when num < 4\nevaluates to trueclearBreakpoint('script.js', 1), cb(...): Clear breakpoint in script.js\non line 1It is also possible to set a breakpoint in a file (module) that\nis not loaded yet:
\n$ node inspect main.js\n< Debugger listening on ws://127.0.0.1:9229/48a5b28a-550c-471b-b5e1-d13dd7165df9\n< For help, see: https://nodejs.org/learn/getting-started/debugging\n<\nconnecting to 127.0.0.1:9229 ... ok\n< Debugger attached.\n<\nBreak on start in main.js:1\n> 1 const mod = require('./mod.js');\n 2 mod.hello();\n 3 mod.hello();\ndebug> setBreakpoint('mod.js', 22)\nWarning: script 'mod.js' was not loaded yet.\ndebug> c\nbreak in mod.js:22\n 20 // USE OR OTHER DEALINGS IN THE SOFTWARE.\n 21\n>22 exports.hello = function() {\n 23 return 'hello from module';\n 24 };\ndebug>\nIt is also possible to set a conditional breakpoint that only breaks when a\ngiven expression evaluates to true:
$ node inspect main.js\n< Debugger listening on ws://127.0.0.1:9229/ce24daa8-3816-44d4-b8ab-8273c8a66d35\n< For help, see: https://nodejs.org/learn/getting-started/debugging\n<\nconnecting to 127.0.0.1:9229 ... ok\n< Debugger attached.\nBreak on start in main.js:7\n 5 }\n 6\n> 7 addOne(10);\n 8 addOne(-1);\n 9\ndebug> setBreakpoint('main.js', 4, 'num < 0')\n 1 'use strict';\n 2\n 3 function addOne(num) {\n> 4 return num + 1;\n 5 }\n 6\n 7 addOne(10);\n 8 addOne(-1);\n 9\ndebug> cont\nbreak in main.js:4\n 2\n 3 function addOne(num) {\n> 4 return num + 1;\n 5 }\n 6\ndebug> exec('num')\n-1\ndebug>\nbacktrace, bt: Print backtrace of current execution framelist(5): List scripts source code with 5 line context (5 lines before and\nafter)watch(expr): Add expression to watch listunwatch(expr): Remove expression from watch listunwatch(index): Remove expression at specific index from watch listwatchers: List all watchers and their values (automatically listed on each\nbreakpoint)repl: Open debugger's repl for evaluation in debugging script's contextexec expr, p expr: Execute an expression in debugging script's context and\nprint its valueprofile: Start CPU profiling sessionprofileEnd: Stop current CPU profiling sessionprofiles: List all completed CPU profiling sessionsprofiles[n].save(filepath = 'node.cpuprofile'): Save CPU profiling session\nto disk as JSONtakeHeapSnapshot(filepath = 'node.heapsnapshot'): Take a heap snapshot\nand save to disk as JSONrun: Run script (automatically runs on debugger's start)restart: Restart scriptkill: Kill scriptscripts: List all loaded scriptsversion: Display V8's versionV8 Inspector integration allows attaching Chrome DevTools to Node.js\ninstances for debugging and profiling. It uses the\nChrome DevTools Protocol.
\nV8 Inspector can be enabled by passing the --inspect flag when starting a\nNode.js application. It is also possible to supply a custom port with that flag,\ne.g. --inspect=9222 will accept DevTools connections on port 9222.
Using the --inspect flag will execute the code immediately before debugger is connected.\nThis means that the code will start running before you can start debugging, which might\nnot be ideal if you want to debug from the very beginning.
In such cases, you have two alternatives:
\n--inspect-wait flag: This flag will wait for debugger to be attached before executing the code.\nThis allows you to start debugging right from the beginning of the execution.--inspect-brk flag: Unlike --inspect, this flag will break on the first line of the code\nas soon as debugger is attached. This is useful when you want to debug the code step by step\nfrom the very beginning, without any code execution prior to debugging.So, when deciding between --inspect, --inspect-wait, and --inspect-brk, consider whether you want\nthe code to start executing immediately, wait for debugger to be attached before execution,\nor break on the first line for step-by-step debugging.
$ node --inspect index.js\nDebugger listening on ws://127.0.0.1:9229/dc9010dd-f8b8-4ac5-a510-c1a114ec7d29\nFor help, see: https://nodejs.org/learn/getting-started/debugging\n(In the example above, the UUID dc9010dd-f8b8-4ac5-a510-c1a114ec7d29\nat the end of the URL is generated on the fly, it varies in different\ndebugging sessions.)
", "displayName": "V8 inspector integration for Node.js" } ], "displayName": "Advanced usage" } ] } ] }