Egg: 按文档进行线上部署一直报这个错，怎么解决？

Created on 22 Jan 2018  ·  44Comments  ·  Source: eggjs/egg

[egg-scripts] Wait Start: 1...
[egg-scripts] Wait Start: 2...
[egg-scripts] Wait Start: 3...
[egg-scripts] Wait Start: 4...
[egg-scripts] Wait Start: 5...
[egg-scripts] Wait Start: 6...
[egg-scripts] Wait Start: 7...
[egg-scripts] Wait Start: 8...
[egg-scripts] Wait Start: 9...
[egg-scripts] Wait Start: 10...
[egg-scripts] Wait Start: 11...
[egg-scripts] Wait Start: 12...
[egg-scripts] Wait Start: 13...
[egg-scripts] Wait Start: 14...
[egg-scripts] Wait Start: 15...
[egg-scripts] Wait Start: 16...
[egg-scripts] Wait Start: 17...
[egg-scripts] Wait Start: 18...
[egg-scripts] Wait Start: 19...
[egg-scripts] Wait Start: 20...
[egg-scripts] Wait Start: 21...
[egg-scripts] Wait Start: 22...
[egg-scripts] Wait Start: 23...
[egg-scripts] Wait Start: 24...
[egg-scripts] Wait Start: 25...
[egg-scripts] Wait Start: 26...
[egg-scripts] Wait Start: 27...
[egg-scripts] Wait Start: 28...
[egg-scripts] Wait Start: 29...
[egg-scripts] Wait Start: 30...
[egg-scripts] Wait Start: 31...
[egg-scripts] Wait Start: 32...
[egg-scripts] Wait Start: 33...
[egg-scripts] Wait Start: 34...
[egg-scripts] Wait Start: 35...
[egg-scripts] 2018-01-22 11:43:51,078 ERROR 838 [-/127.0.0.1/-/0ms GET /] nodejs.Error: [ClusterClient] leader does not be active in 30000ms on port:41217
[egg-scripts] at Function.waitFor (/website/production/source/node_modules/cluster-client/lib/server.js:239:15)
[egg-scripts] at waitFor.next ()
[egg-scripts] at onFulfilled (/website/production/source/node_modules/co/index.js:65:19)
[egg-scripts] at
[egg-scripts] at process._tickCallback (internal/process/next_tick.js:188:7)

question
Source
picture zrxisme

Most helpful comment

iptables 用得不多，不过我觉得你这个 policy 没有允许 loopback 的 INPUT 有点多余了。试试下面的配置。

-A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT
-A OUTPUT -j ACCEPT
-A INPUT -i lo -j ACCEPT
-A INPUT -p tcp --dport 443 -j ACCEPT
-A INPUT -p tcp --dport 80 -j ACCEPT
-A INPUT -p tcp --dport 8081 -j ACCEPT
-A INPUT -p tcp -m state --state NEW --dport 39999 -j ACCEPT
-A INPUT -p icmp -m icmp --icmp-type 8 -j ACCEPT
-A INPUT -m limit --limit 5/min -j LOG --log-prefix "iptables denied:" --log-level 7
-A INPUT -p tcp --dport 80 -i eth0 -m state --state NEW -m recent --set
-A INPUT -p tcp --dport 80 -i eth0 -m state --state NEW -m recent --update --seconds 60 --hitcount 150 -j DROP
-A INPUT -j REJECT
-A FORWARD -j REJECT
COMMIT
picture geekdada on 31 Jan 2018
🎉2

All 44 comments

启动超时，是不是网络不通导致某些插件（如 mysql）或者你自己的 beforeStart 一直没返回

atian25 picture atian25 on 22 Jan 2018

完整的报错信息如下，在本地测试npm run dev 没问题，部署到线上的时候npm start一直不行，而且很奇怪为什么会报端口:44898的错误，我侦听的是3000端口，我的防火墙只允许测试的3000端口通过

2018-01-22 11:10:08,950 ERROR 458 [-/127.0.0.1/-/1ms GET /] nodejs.Error: [ClusterClient] leader does not be active in 30000ms on port:44898
[egg-scripts] at Function.waitFor (/website/production/source/node_modules/cluster-client/lib/server.js:239:15)
[egg-scripts] at waitFor.next ()
[egg-scripts] at onFulfilled (/website/production/source/node_modules/co/index.js:65:19)
[egg-scripts] at
[egg-scripts] at process._tickCallback (internal/process/next_tick.js:188:7)
[egg-scripts]
[egg-scripts] pid: 458
[egg-scripts]
[egg-scripts] 2018-01-22 11:10:08,951 ERROR 458 nodejs.Error: [ClusterClient] leader does not be active in 30000ms on port:44898
[egg-scripts] at Function.waitFor (/website/production/source/node_modules/cluster-client/lib/server.js:239:15)
[egg-scripts] at waitFor.next ()
[egg-scripts] at onFulfilled (/website/production/source/node_modules/co/index.js:65:19)
[egg-scripts] at
[egg-scripts] at process._tickCallback (internal/process/next_tick.js:188:7)
[egg-scripts]
[egg-scripts] pid: 458
[egg-scripts]
[egg-scripts] 2018-01-22 11:10:08,951 ERROR 458 [app_worker] start error, exiting with code:1
[egg-scripts] 2018-01-22 11:10:08,951 ERROR 458 [app_worker] exit with code:1
[egg-scripts] 2018-01-22 11:10:08,956 ERROR 442 nodejs.AppWorkerDiedError: [master] app_worker#1:458 died (code: 1, signal: null, suicide: false, state: dead), current workers: []
[egg-scripts] at Master.onAppExit (/www/website/production/source/node_modules/egg-cluster/lib/master.js:384:21)
[egg-scripts] at emitOne (events.js:115:13)
[egg-scripts] at Master.emit (events.js:210:7)
[egg-scripts] at Messenger.sendToMaster (/website/production/source/node_modules/egg-cluster/lib/utils/messenger.js:122:17)
[egg-scripts] at Messenger.send (/website/production/source/node_modules/egg-cluster/lib/utils/messenger.js:87:12)
[egg-scripts] at EventEmitter.cluster.on (/website/production/source/node_modules/egg-cluster/lib/master.js:263:22)
[egg-scripts] at emitThree (events.js:140:20)
[egg-scripts] at EventEmitter.emit (events.js:216:7)
[egg-scripts] at ChildProcess.worker.process.once (internal/cluster/master.js:185:13)
[egg-scripts] at Object.onceWrapper (events.js:318:30)
[egg-scripts] name: 'AppWorkerDiedError'
[egg-scripts] pid: 442
[egg-scripts]
[egg-scripts] 2018-01-22 11:10:08,958 ERROR 442 [master] app_worker#1:458 start fail, exiting with code:1
[egg-scripts] 2018-01-22 11:10:08,959 ERROR 442 [master] exit with code:1
[egg-scripts] 2018-01-22 11:10:08,961 ERROR 448 [agent_worker] receive disconnect event on child_process fork mode, exiting with code:110
[egg-scripts] 2018-01-22 11:10:08,961 ERROR 448 [agent_worker] exit with code:110

zrxisme picture zrxisme on 22 Jan 2018

那个超时是 ClusterClient 内部通讯的，目前猜测是你在启动期的 worker 里面，做了一些比较占 CPU 的操作，导致启动超时。建议检查下代码，尤其是 beforeStart 这种地方。

或者提供一个最小可复现代码库，目前这些错误信息不足判断。

atian25 picture atian25 on 23 Jan 2018

我也遇到了同样的问题，没有写过什么beforeStart 这些，我还尝试了 重新新建了一个项目，然后直接start 也是一样，我的服务器环境是阿里云的，单核cpu 我怀疑是不是配置的问题？

jy418 picture jy418 on 23 Jan 2018

试下加上 --workers= 参数，Node 在某些云服务的 cpu.length 返回值有 bug。

atian25 picture atian25 on 23 Jan 2018

[egg-scripts] Wait Start: 35...
[egg-scripts] 2018-01-23 16:12:03,863 ERROR 3941 [-/127.0.0.1/-/0ms GET /] nodejs.Error: [ClusterClient] leader does not be active in 30000ms on port:45728
[egg-scripts] at Function.waitFor (/root/egg/node_modules/[email protected]@cluster-client/lib/server.js:239:15)
[egg-scripts] at waitFor.next ()
[egg-scripts] at onFulfilled (/root/egg/node_modules/[email protected]@co/index.js:65:19)
[egg-scripts] at
[egg-scripts] at process._tickCallback (internal/process/next_tick.js:160:7)
[egg-scripts]
[egg-scripts] pid: 3941
[egg-scripts] hostname: iZ2ze76p2wnibfx0j7dslmZ
[egg-scripts]
[egg-scripts] 2018-01-23 16:12:03,863 ERROR 3941 nodejs.Error: [ClusterClient] leader does not be active in 30000ms on port:45728
[egg-scripts] at Function.waitFor (/root/egg/node_modules/[email protected]@cluster-client/lib/server.js:239:15)
[egg-scripts] at waitFor.next ()
[egg-scripts] at onFulfilled (/root/egg/node_modules/[email protected]@co/index.js:65:19)
[egg-scripts] at
[egg-scripts] at process._tickCallback (internal/process/next_tick.js:160:7)
[egg-scripts]
[egg-scripts] pid: 3941
[egg-scripts] hostname: iZ2ze76p2wnibfx0j7dslmZ
[egg-scripts]
[egg-scripts] 2018-01-23 16:12:03,863 ERROR 3941 [app_worker] start error, exiting with code:1
[egg-scripts] 2018-01-23 16:12:03,863 ERROR 3941 [app_worker] exit with code:1
[egg-scripts] 2018-01-23 16:12:03,869 ERROR 3925 nodejs.AppWorkerDiedError: [master] app_worker#1:3941 died (code: 1, signal: null, suicide: false, state: dead), current workers: []
[egg-scripts] at Master.onAppExit (/root/egg/node_modules/[email protected]@egg-cluster/lib/master.js:384:21)
[egg-scripts] at Master.emit (events.js:159:13)
[egg-scripts] at Messenger.sendToMaster (/root/egg/node_modules/[email protected]@egg-cluster/lib/utils/messenger.js:122:17)
[egg-scripts] at Messenger.send (/root/egg/node_modules/[email protected]@egg-cluster/lib/utils/messenger.js:87:12)
[egg-scripts] at EventEmitter.cluster.on (/root/egg/node_modules/[email protected]@egg-cluster/lib/master.js:263:22)
[egg-scripts] at EventEmitter.emit (events.js:164:20)
[egg-scripts] at ChildProcess.worker.process.once (internal/cluster/master.js:185:13)
[egg-scripts] at Object.onceWrapper (events.js:254:19)
[egg-scripts] at ChildProcess.emit (events.js:159:13)
[egg-scripts] at Process.ChildProcess._handle.onexit (internal/child_process.js:209:12)
[egg-scripts] name: 'AppWorkerDiedError'
[egg-scripts] pid: 3925
[egg-scripts] hostname: iZ2ze76p2wnibfx0j7dslmZ
[egg-scripts]
[egg-scripts] 2018-01-23 16:12:03,870 ERROR 3925 [master] app_worker#1:3941 start fail, exiting with code:1
[egg-scripts] 2018-01-23 16:12:03,871 ERROR 3925 [master] exit with code:1
[egg-scripts] 2018-01-23 16:12:03,873 ERROR 3931 [agent_worker] receive disconnect event on child_process fork mode, exiting with code:110
[egg-scripts] 2018-01-23 16:12:03,873 ERROR 3931 [agent_worker] exit with code:110
[egg-scripts]
[egg-scripts] Start got error, see /root/logs/master-stderr.log
npm ERR! code ELIFECYCLE
npm ERR! errno 1
npm ERR! [email protected] start: egg-scripts start --workers=1 --daemon
npm ERR! Exit status 1
npm ERR!
npm ERR! Failed at the [email protected] start script.
npm ERR! This is probably not a problem with npm. There is likely additional logging output above.

npm ERR! A complete log of this run can be found in:
npm ERR! /root/.npm/_logs/2018-01-23T08_12_05_860Z-debug.log
刚刚测试了一下
加上了 --workers=1 还是不行
求大神支招~~~
还是新项目，没有加过任何内容的

jy418 picture jy418 on 23 Jan 2018

删除依赖重新安装下看看，貌似你的依赖都挺旧的，不要锁版本。

本地能否复现？

atian25 picture atian25 on 23 Jan 2018

刚刚删除了node_modules 重新npm i 安装之后（安装过程没有报错,也没有锁定版本）但是还是在报这个错。。。

jy418 picture jy418 on 23 Jan 2018

提供本地最小可复现仓库吧，这样隔空猜不出

atian25 picture atian25 on 23 Jan 2018

最小可复现仓库？，我试过直接在阿里云服务器上按照下面教程安装启动，代码没动，但是照样报同样的错误，实在找不到原因，一直很奇怪“ERROR 9248 nodejs.Error: [ClusterClient] leader does not be active in 30000ms on port:45188”这个是什么错误，这个端口哪里来的，而且每一次启动错误的端口都不一样?

$ npm i egg-init -g
$ egg-init egg-example --type=simple
$ cd egg-example
$ npm install
$npm start

zrxisme picture zrxisme on 23 Jan 2018

本地是否能复现？

atian25 picture atian25 on 23 Jan 2018

本地是window系统npm run dev正常启动放到服务器上npm start就不行了

zrxisme picture zrxisme on 23 Jan 2018

仓库我没搞好。。。但是确实如楼上所说。。。就是按照教程创建的一个新项目 什么都没改过 都会报这个错。。。

jy418 picture jy418 on 23 Jan 2018

本地windows 直接npm start 都是对的 都可以访问。。。到linux就不行了。。。ubuntu16

jy418 picture jy418 on 23 Jan 2018

我只是初始化项目而已，没有加任何代码

zrxisme picture zrxisme on 23 Jan 2018

能否提供一个环境让我看看？

JacksonTian picture JacksonTian on 23 Jan 2018

测试的服务器是Ubuntu 16.04 64位

zrxisme picture zrxisme on 23 Jan 2018

@jy418 @zrxisme 周末会帮你们看看，之前我在centos自己跑没问题，阿里云的我到时候看看

fenis picture fenis on 23 Jan 2018

我试过root在阿里云ubuntu正常

npm i egg-init -g
egg-init egg-example --type=simple
cd egg-example
npm install
npm start

@zrxisme 试一下root用户运行？我看你是$，再帮忙发一下node版本

feix760 picture feix760 on 23 Jan 2018

我服务侦听的是3000端口，为何下面会有个45728端口的错误，而且错误的端口还会变化，如何让线上部署的程序只启动侦听服务的3000端口或者让45728这个端口不变，因为我的iptables规则没有让45728这个端口通过，所以启动失败，但iptables规则加上这个端口后，错误的端口又变了，所以一直启动不了，只有清除iptables规则后才能启动成功，那位大佬可以解答？求帮忙

egg-scripts] 2018-01-23 16:12:03,863 ERROR 3941 [-/127.0.0.1/-/0ms GET /] nodejs.Error: [ClusterClient] leader does not be active in 30000ms on port:45728

zrxisme picture zrxisme on 26 Jan 2018

@zrxisme 我这边切阿里云切换下系统镜像看下，跟着把你iptables配置的规则发上来看看

fenis picture fenis on 28 Jan 2018

我iptables规则只开通了eggjs侦听的3000端口和服务器登陆端口

zrxisme picture zrxisme on 28 Jan 2018

不知道为什么npm start启动eggjs后除了侦听的3000端口外还会有其他端口被启动，其他端口没通过iptables规则所以启动失败，其他端口哪里来的不知道，怀疑是socket问题

zrxisme picture zrxisme on 28 Jan 2018

@zrxisme lsof -i:端口号 看下端口被什么占用了

fenis picture fenis on 28 Jan 2018

。。。端口根本就没通过iptables规则，怎么会占用呢，你自己试下配置iptables规则，除了侦听的端口以外，其他端口不予通过，然后启动egg你就可以看到报什么错误了

zrxisme picture zrxisme on 28 Jan 2018

我看过 报错的端口号之前没有被占用过，应该就是iptables配置 ，有人试试删除iptables的配置可以成功么？

jy418 picture jy418 on 28 Jan 2018

我试过，清除iptables规则，启动成功

zrxisme picture zrxisme on 28 Jan 2018

额 那这样的话 感觉有点不爽额 没有iptables 有裸奔的感觉

jy418 picture jy418 on 28 Jan 2018

对，没有根本解决问题，额外端口问题还是解决不了

zrxisme picture zrxisme on 28 Jan 2018

话说有没有试过不用egg-script启动 换成pm2启动？

jy418 picture jy418 on 28 Jan 2018

我用pm2的ecosystem.json的方式部署过，iptables规则没清除，也启动不了

zrxisme picture zrxisme on 28 Jan 2018

端口问题推测出在这个插件里cluster-client，只是找了很久都没找到解决方案

zrxisme picture zrxisme on 28 Jan 2018

@JacksonTian: 能否提供一个环境让我看看？

https://github.com/eggjs/egg/issues/2004#issuecomment-359729397

@zrxisme

atian25 picture atian25 on 29 Jan 2018

目前来看就是因为设置Iptables导致跑不起来 附上我的相关配置 8081是egg监听的端口
-A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT
-A OUTPUT -j ACCEPT
-A INPUT -p tcp --dport 443 -j ACCEPT
-A INPUT -p tcp --dport 80 -j ACCEPT
-A INPUT -p tcp --dport 8081 -j ACCEPT
-A INPUT -p tcp -m state --state NEW --dport 39999 -j ACCEPT
-A INPUT -p icmp -m icmp --icmp-type 8 -j ACCEPT
-A INPUT -m limit --limit 5/min -j LOG --log-prefix "iptables denied:" --log-level 7
-A INPUT -p tcp --dport 80 -i eth0 -m state --state NEW -m recent --set
-A INPUT -p tcp --dport 80 -i eth0 -m state --state NEW -m recent --update --seconds 60 --hitcount 150 -j DROP
-A INPUT -j REJECT
-A FORWARD -j REJECT
COMMIT

jy418 picture jy418 on 29 Jan 2018

按理说内部端口不会被拦截的吧？ @JacksonTian

atian25 picture atian25 on 31 Jan 2018
👍1

但是楼主亲测，清除iptables就可以启动了的说。。。

jy418 picture jy418 on 31 Jan 2018

@fenis 等同学试了没？能否复现？

atian25 picture atian25 on 31 Jan 2018

iptables 用得不多，不过我觉得你这个 policy 没有允许 loopback 的 INPUT 有点多余了。试试下面的配置。

-A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT
-A OUTPUT -j ACCEPT
-A INPUT -i lo -j ACCEPT
-A INPUT -p tcp --dport 443 -j ACCEPT
-A INPUT -p tcp --dport 80 -j ACCEPT
-A INPUT -p tcp --dport 8081 -j ACCEPT
-A INPUT -p tcp -m state --state NEW --dport 39999 -j ACCEPT
-A INPUT -p icmp -m icmp --icmp-type 8 -j ACCEPT
-A INPUT -m limit --limit 5/min -j LOG --log-prefix "iptables denied:" --log-level 7
-A INPUT -p tcp --dport 80 -i eth0 -m state --state NEW -m recent --set
-A INPUT -p tcp --dport 80 -i eth0 -m state --state NEW -m recent --update --seconds 60 --hitcount 150 -j DROP
-A INPUT -j REJECT
-A FORWARD -j REJECT
COMMIT
geekdada picture geekdada on 31 Jan 2018
🎉2

@zrxisme 阿里云上目前无法复现，ecs ubuntu16 node 8.9 关于防火墙这块，阿里云直接在安全组配置端口就可以，建议你直接配置一个可用端口区间

fenis picture fenis on 1 Feb 2018

看来问题属于 iptable 这块的问题了，可以自己排查下了。

egg 有 cluster-client 来做进程间的高级通信，所以会开一些端口，但这些端口只是内部通信的。

atian25 picture atian25 on 1 Feb 2018

楼主最后怎么解决的？最近在学egg，部署到服务器也遇到跟你同样的问题。。。搞了好几天了。。。没搞明白 @zrxisme

akumaLin picture akumaLin on 15 Apr 2018

@atian25 遇到同样的问题

imshenshen picture imshenshen on 17 Apr 2018

@akumaLin @imshenshen https://github.com/eggjs/egg/issues/2004#issuecomment-361878130 试过按照这个解决吗？

试试临时关掉 iptables

sudo service iptables stop
geekdada picture geekdada on 17 Apr 2018

iptables增加一条规则，允许本地回环接口(即运行本机访问本机)：
-A INPUT -s 127.0.0.1 -d 127.0.0.1 -j ACCEPT

variy picture variy on 15 Jan 2020
Was this page helpful?
0 / 5 - 0 ratings

Related issues

zlab picture zlab  ·  55Comments
atian25 picture atian25  ·  68Comments
fengmk2 picture fengmk2  ·  51Comments
popomore picture popomore  ·  59Comments
kkys4bfgp75be9p picture kkys4bfgp75be9p  ·  36Comments