Avoiding memory bloat with non-external runtimes

未关闭
#60 1 条评论 3 个 reaction 已指派 0 人 在 GitHub 查看

还没有人认领这个 Issue。

评估

难度
5/5
预计耗时
一周以上
新手友好度
30/100
Issue 类型
功能
描述清晰度
需要澄清
活跃度
停滞
技术栈
ruby
领域
tooling

调研方向

Start with the ExecJS.compile and compiled-context call API described in the issue, then trace how non-external runtimes manage their contexts. Define the release lifecycle and behavior after release, including whether memory is reclaimed by MiniRacer. The issue does not name implementation files or tests, so those must be located before work can be scoped.

由索引模型根据 Issue 内容生成。

描述

I have observed that when using miniracer with our precompile assets process memory can run away big time.

This is cause sprockets often bloats the heap and at the point execjs is called there are extra v8 contexts sitting around in a bloated heap that just is too big to run a GC cause so much is free in it.

Trouble is a v8 context can easily consume 100+MB of RSS.

I was wondering if we could add to execjs a concept of "OK, I am done with the context"

Once that is called, MiniRacer could release the v8 context and reclaim all the memory. This means that usage wise if you want to reclaim memory early you would do:

context = ExecJS.compile(source)
context.call("CoffeeScript.compile", "square = (x) -> x * x", bare: true)
context.release
context.call("X") => raises error cause context is done and poison at this point. 

Clearly implementation would be optional.

Thoughts ?

主要语言
Ruby
星标
581
派生
285
PR 合并指标
30 天内没有已合并 PR

环境准备

  • 没有 Dockerfile 或 Docker Compose 文件
  • 没有 Pull Request 模板
  • 阅读贡献指南

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

rails/execjs 的其他 Issue

查看 rails/execjs 的全部 Issue

相似的 Issue

更多 Ruby Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。