aliyun-node/node.js-troubleshooting-guide

Node.js 应用线上/线下故障、压测问题和性能调优指南手册（一期更新结束）

Node.js 应用故障排查手册

本手册主要的目的是帮助广大的 Node.js 开发者应对开发和线上部署中遇到的问题：

• 线上/线下应用出现故障时如何更快地定位问题
• 应用部署前压测中遇到的性能不足问题调优

目录

• 序章：本书定位与大纲

• 第一部分：预备篇 - 案发现场的蛛丝马迹

  • 常规排查的指标
  • 核心转储（Core dump）

• 第二部分：工具篇 - 玩转排查利器

  • 正确打开 Chrome devtools
  • Node.js 性能平台使用指南

• 第三部分：实践篇之一 - 专治疑难杂症

  • 利用 CPU 分析调优吞吐量
  • 冗余配置传递引发的内存溢出
  • 综合性 GC 问题和优化

• 第四部分：实践篇之二 - 更极端的难题

  • 类死循环导致进程阻塞
  • 雪崩型内存泄漏问题

• 附录一：更多 Node.js 性能优化实践分享

  • 后端服务性能优化总结（来自 Teambition）

// TODO: 未完待续

相关专栏

本手册也会在云栖社区同步更新，我们的云栖社区 Node.js 性能平台 专栏有更多 Node.js 相关的技术分享，欢迎订阅，地址：https://yq.aliyun.com/teams/210

知乎 Egg.js 核心团队运营的专栏也有许多 Egg.js/Node.js 各路大神的分享，欢迎关注，地址：https://zhuanlan.zhihu.com/eggjs

贡献方式

本手册目的是帮助 Node.js 开发者能更好地来应对开发过程中的各种问题，能更多的建立起使用 Node.js 踏足服务端领域的信心，也希望能让天下没有难用的 Node.js。我们非常欢迎大家有合适的案例和最佳实践来和大家一起分享。

参与方法如下：

• Fork 本仓库至你自己的 Github 仓库列表中
• Clone 你 Fork 出来的仓库至本地开发
• 添加你的案例或者最佳实践至合适的位置
• 添加 README.md 中对应的描述，提交至你的远程仓库
• 在 PR 页面选择 New Pull Request，继续选择 compare across forks，在列表中选中你的 Fork，然后创建新的 PR

我们会在 Review 后选择合并至本仓库内，贡献者也会在 README.md 中的贡献者列表中展示。

贡献列表

以下贡献数据来自 git summary:

• hyj1991 86.7%% ( 13 commits )
• richardwei195 6.7% ( 1 commits )
• Hui Hui 6.7% ( 1 commits )

LICENSE

本书采用 保持署名—非商用 创意共享 4.0 许可证。

只要保持原作者署名和非商用，您可以自由地阅读、分享、修改本书。

详细的法律条文请参见 创意共享 网站。

node.js-troubleshooting-guide's Issues

docker部署的nodejs应用，如何使用阿里云NodeJs性能平台？

docker部署的nodejs应用，如何使用阿里云NodeJs性能平台？

请问咱们的性能平台啥时候能支持到nodejs16.20.0以上

请问咱们的性能平台啥时候能支持到nodejs16.20.0以上

[TODO] 基于 Vuepress/Docsify 实现文章静态展示页

诊断报告解析失败是什么原因？

文件下载后，是一个正常的 json 文件。

比如：

{
  "header": "Node.js Performance Platform Diagnostic Report",
  "Event": "SIGRTMIN",
  "location": "DumpNodeReport",
  "FileName": "/tmp/diagnostic-report-xxxxx.diag",
  "DumpEventTime": "2019/07/04 14:05:27",
  "ModuleLoadTime": "1900/01/00 00:00:00",
  "ProcessID": "4691", 
  "NodeJsVersion": "10.15.3",
  "alinodeVersion": "4.7.2",
  "execPath": "/usr/local/bin/node",
  "glibc": "",
  "OsVersion": "Linux 3.10.0-957.5.1.el7.x86_64 #1 SMP Fri Feb 1 14:54:57 UTC 2019(glibc: )",
  "Machine": "uni-account-bff-5c5d868ffb-2gtjn x86_64",
  "JavaScriptStack": {
    "DumpEvent": "uvSignal",
    "VMState": "unknown",
    "ErrorCode": "3",
    "ErrorMessage": "eventloop idle, no stack trace",
    "stacks": []
  },
  "NativeStack": {
    "ErrorCode": "1",
    "ErrorMessage": "Native stack trace not supported on Linux platforms w/o GLIBC.",
    "stacks": []
  },
  "JsHeapGcInfo": {
    "read_only_space": {
      "memory": 524288 ,
      "committed": 43904 ,
      "capacity": 515584 ,
      "used": 35200 ,
      "available": 480384
    },
    "new_space": {
      "memory": 1048576 ,
      "committed": 975088 ,
      "capacity": 1031168 ,
      "used": 957728 ,
      "available": 73440
    },
    "old_space": {
      "memory": 125538304 ,
      "committed": 124466664 ,
      "capacity": 122828992 ,
      "used": 121442496 ,
      "available": 1386496
    },
    "code_space": {
      "memory": 1572864 ,
      "committed": 1549184 ,
      "capacity": 1501248 ,
      "used": 1297472 ,
      "available": 203776
    },
    "map_space": {
      "memory": 2109440 ,
      "committed": 1743152 ,
      "capacity": 2065920 ,
      "used": 1402632 ,
      "available": 663288
    },
    "large_object_space": {
      "memory": 3198976 ,
      "committed": 3198976 ,
      "capacity": 1394578824 ,
      "used": 1853320 ,
      "available": 1392725504
    },
    "total": {
      "heap_size": 133992448 ,
      "committed": 131976968 ,
      "used": 126988848 ,
      "available": 1395532888 ,
      "heap_limit": 1526909922
    }
  },
  "resource": {
    "total": { 
      "cpu_usr_time": 19.500450 ,
      "kernel_cpu_time": 6.734033 ,
      "average_cpu_percentage": 1.67931e-06 ,
      "max_rss": 157548544 ,
      "page_faults": { "io_required": 1, "no_io_required": 178634 },
      "fs_activity": { "read": 16968, "write": 1096 }
    },
    "eventloop_thread": { 
      "cpu_usr_time": 8.994876 ,
      "kernel_cpu_time": 1.105283 ,
      "average_cpu_percentage": 6.46526e-07 ,
      "fs_activity": { "read": 16968, "write": 296 }
    }
  },
  "libuvHandleSummary": [
    {"hasRef": false, "isActive": true, "type": "async", "address": "0x556b3d1fb140", "details": ""},
    {"hasRef": false, "isActive": true, "type": "check", "address": "0x7fffed5eb6f8", "details": ""},
    {"hasRef": true, "isActive": false, "type": "idle", "address": "0x7fffed5eb770", "details": ""},
    {"hasRef": false, "isActive": false, "type": "prepare", "address": "0x7fffed5eb7e8", "details": ""},
    {"hasRef": false, "isActive": false, "type": "check", "address": "0x7fffed5eb860", "details": ""},
    {"hasRef": false, "isActive": true, "type": "async", "address": "0x556b3ba02c00", "details": ""},
    {"hasRef": true, "isActive": false, "type": "timer", "address": "0x556b3ba19280", "details": "repeat: 0, timeout expired: 98309344 ms ago"},
    {"hasRef": false, "isActive": true, "type": "async", "address": "0x556b3ba19da0", "details": ""},
    {"hasRef": true, "isActive": true, "type": "pipe", "address": "0x556b3d20ac78", "details": "send buffer size: 212992, recv buffer size: 212992, file descriptor: 3, write queue size: 0, readable, writable"},
    {"hasRef": true, "isActive": false, "type": "pipe", "address": "0x556b3d214718", "details": "send buffer size: 0, recv buffer size: 0, stdout, write queue size: 0, writable"},
    {"hasRef": true, "isActive": false, "type": "pipe", "address": "0x556b3d2148d8", "details": "send buffer size: 0, recv buffer size: 0, stderr, write queue size: 0, writable"},
    {"hasRef": true, "isActive": false, "type": "timer", "address": "0x556b3d253b00", "details": "repeat: 0, timeout in: 139981366982024 ms"},
    {"hasRef": true, "isActive": false, "type": "timer", "address": "0x556b3d253b98", "details": "repeat: 0, timeout in: 93918977684736 ms"},
    {"hasRef": false, "isActive": true, "type": "async", "address": "0x7f4ff23a2780", "details": ""},
    {"hasRef": true, "isActive": false, "type": "prepare", "address": "0x7f4ff23a26a0", "details": ""},
    {"hasRef": true, "isActive": true, "type": "timer", "address": "0x556b3d2c1ef0", "details": "repeat: 0, timeout in: 851 ms"},
    {"hasRef": false, "isActive": true, "type": "signal", "address": "0x556b3d1ff4f0", "details": "signum: 15 (SIGTERM)"},
    {"hasRef": true, "isActive": true, "type": "tcp", "address": "0x556b3d340938", "details": "uni-account-bff-5c5d868ffb-2gtjn:59320 connected to 172.16.151.19:6379, send buffer size: 12582912, recv buffer size: 12582912, file descriptor: 37, write queue size: 0, readable, writable"},
    {"hasRef": false, "isActive": true, "type": "async", "address": "0x556b3d341528", "details": ""},
    {"hasRef": true, "isActive": true, "type": "tcp", "address": "0x556b3d44f718", "details": "localhost:42584 connected to localhost:45378, send buffer size: 12582912, recv buffer size: 12582912, file descriptor: 38, write queue size: 0, readable, writable"},
    {"hasRef": true, "isActive": true, "type": "timer", "address": "0x556b3d289dd0", "details": "repeat: 0, timeout in: 14370 ms"},
    {"hasRef": false, "isActive": true, "type": "timer", "address": "0x556b3d285b10", "details": "repeat: 0, timeout in: 39081 ms"},
    {"hasRef": false, "isActive": true, "type": "timer", "address": "0x556b3d2e34e0", "details": "repeat: 0, timeout in: 4967 ms"},
    {"dummy_handle": ""}
  ],
... 省略

default & 64MB screenshots of GC are the same

The screenshots in https://github.com/aliyun-node/Node.js-Troubleshooting-Guide/blob/master/0x08_%E5%AE%9E%E8%B7%B5%E7%AF%87_%E7%BB%BC%E5%90%88%E6%80%A7%20GC%20%E9%97%AE%E9%A2%98%E5%92%8C%E4%BC%98%E5%8C%96.md show exactly the same GC performance in the default settings & at 64MB. The article says the performance has increased significantly, but with the same image being used in both, it's hard to see what the improvement really is.