egg线上运行一段时间后停了,也没有重启。

Created on 9 May 2018 · 9Comments · Source: eggjs/egg

Node Version:8.11.0
Egg Version:2.6.0
Plugin Name:
Plugin Version:
Platform:
Mini Showcase Repository:

用的egg-script启动的，运行了不到一天然后停了，也没有重启。之前没有出现过这种情况，升级了服务器后出现的。
请问这种情况是什么原因造成的？该如何解决忽然停止不重启的问题？谢谢。

[8045:0x3bebbd0] 67560204 ms: Mark-sweep 1871.0 (1939.3) -> 1871.0 (1939.3) MB, 1184.8 / 0.2 ms  allocation failure GC in old space requested
[8045:0x3bebbd0] 67561539 ms: Mark-sweep 1871.0 (1939.3) -> 1871.0 (1908.3) MB, 1335.0 / 0.3 ms  last resort GC in old space requested
[8045:0x3bebbd0] 67562873 ms: Mark-sweep 1871.0 (1908.3) -> 1871.0 (1908.3) MB, 1333.6 / 0.1 ms  last resort GC in old space requested

Source

Cheese-Yu

Most helpful comment

更新。在alinode团队的帮助下，发现原因。由于项目架构的问题，频繁(每秒2-3次,持续几十分钟)通过app.messenger.broadcast进行大数据通信，导致的内存积压，egg是不推荐使用这种模式通信大数据的。目前已经调整成使用redis做数据同步了。十分感谢alinode!

Cheese-Yu on 14 May 2018

👍4

All 9 comments

Translation of this issue:

The egg line stopped after running for a while and did not restart.

Node Version: 8.11.0
Egg Version: 2.6.0
Plugin Name:
Plugin Version:
Platform:
Mini Showcase Repository:

With egg-script started, it took less than a day to run and stopped without rebooting. There has not been such a situation before, after upgrading the server.
What is the cause of this situation? How to solve the problem of suddenly stopping the restart? Thank you.

[8045:0x3bebbd0] 67560204 ms: Mark-sweep 1871.0 (1939.3) -> 1871.0 (1939.3) MB, 1184.8 / 0.2 ms allocation failure GC in old space requested
[8045:0x3bebbd0] 67561539 ms: Mark-sweep 1871.0 (1939.3) -> 1871.0 (1908.3) MB, 1335.0 / 0.3 ms last resort GC in old space requested
[8045:0x3bebbd0] 67562873 ms: Mark-sweep 1871.0 (1908.3) -> 1871.0 (1908.3) MB, 1333.6 / 0.1 ms last resort GC in old space requested