Hacktoberfest 2026:维护者为十月标记出来的 issue,仍然开放、适合新手。 浏览 Hacktoberfest issue

vmm: boot and periodic log rotation copy files on async workers

未关闭
#1,270 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 1 天内回复

还没有人认领这个 Issue。

  • #1272 来自 @kvinwang —— 已关闭,未合并

评估

难度
4/5
预计耗时
3-5 天
新手友好度
48/100
Issue 类型
缺陷
描述清晰度
基本清楚
活跃度
活跃
技术栈
rust

调研方向

Locate App::rotate_oversized_logs and the VM startup rotation path, then trace their interaction with the Tokio maintenance task, fs::copy, and VM lifecycle operations. Keep copytruncate, retention cleanup, and serialized rotation behavior while moving both paths to blocking workers; add regression tests as requested and verify cancellation does not release protection prematurely.

由索引模型根据 Issue 内容生成。

描述

Problem

App::rotate_oversized_logs calls synchronous copytruncate (fs::copy, truncation, retention cleanup) directly from the Tokio maintenance task. The VM startup path also rotates logs synchronously. Large logs, many VMs, or slow storage can occupy async workers and delay unrelated control-plane work.

The default periodic threshold is 4 MB with a 5-second scan interval; files can grow beyond the threshold between scans. Actual latency depends on storage and workload and has not been benchmarked.

Fix

Offload both rotation paths to blocking workers. Keep rotations serialized with each other and VM lifecycle operations, retaining that protection if a caller is cancelled. Preserve copytruncate and retention behavior and add regression tests.

#1263 changes supervisor log flushing; its VMM change is only a comment and does not offload rotation.

主要语言
Rust
星标
555
派生
99
平均合并
1 天 3 小时
30 天内合并 PR
195

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。

Dstack-TEE/dstack 的其他 Issue

查看 Dstack-TEE/dstack 的全部 Issue

相似的 Issue

更多 Rust Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。