百科.dev
全部条目在线工具AI 编程趋势榜开源项目技术资讯提交条目
登录
< 返回工具列表
A

autoclave

> 编程语言
开源

重复运行程序,直到它们出现问题,并准备好连接调试器

200 stars0 点赞0 次浏览
访问官网GitHub

工具介绍

重复运行程序,直到它们出现问题,并准备好连接调试器

autoclave: a pressure cooker for programs

Summary

autoclave repeatedly executes a command line until its process exits with a non-zero status, is stopped/terminated by a signal, a user-specified timeout (-t) runs out, or a user-specified number of runs (-r) have passed without any failures.

It can log the contents of stdout (-l) and/or stderr (-e), and can rotate the log files so a fixed number of logs are kept (-c).

On failure, a handler program (-x) can be called with information about the child process. That way, a debugger can be opened if the stressed program times out or dumps core. For particularly long-running stress tests, the handler may also be a great place to send a notification.

For more detailed usage info, see the examples below the man page, and this blog post.

Examples

Repeatedly run buggy_program until it crashes:

$ autoclave buggy_program

Same, but print run and failure counts:

$ autoclave -v buggy_program

Same, but if it succeeds 10 times, return EXIT_SUCCESS:

$ autoclave -v -r 10 buggy_program

Same, but ensure there is at least 500 mesc between runs, sleeping if necessary:

$ autoclave -v -r 10 -m 500 buggy_program

Run without any delay:

$ autoclave -v -r 10 -m 0 buggy_program

Repeatedly run buggy_program, logging stdout to buggy_program.1.stdout.log, buggy_program.2.stdout.log, and so on:

$ autoclave -l buggy_program

Same, but log to /tmp/buggy.ID.stdout.log instead:

$ autoclave -l -o /tmp/buggy buggy_program

Log, but only keep the 5 most recent log files:

$ autoclave -l -c 5 buggy_program

Same, but log stderr as well as stdout (keeping 10 files):

$ autoclave -l -e -c 5 buggy_program

A shortcut for autoclave -l -e -v ('s' for 'supervise'):

$ autoclave -s buggy_program

Repeatedly run buggy_program until it has failed 10 times:

$ autoclave -f 10 buggy_program

Run a program that occasionally deadlocks, halting it and counting it as a failure if it takes more than 10 seconds to complete:

$ autoclave -t 10  examples/deadlock_example

Same, but run the script examples/gdb_it when it times out, so gdb can attach to the stopped process and investigate what is deadlocking. (Note: the process is not halted after the -x command returns.)

$ autoclave -t 10 -x examples/gdb_it examples/deadlock_example

Run examples/crash_example, calling examples/gdb_it if it fails:

$ autoclave -x examples/gdb_it examples/crash_example

Issues· 4 开放

查看全部 Issues在 GitHub 打开

暂无开放 Issues,或尚未同步最近议题。

> 标签

C

暂无评论,来聊聊你的看法吧

> 工具信息

发布日期2026年8月1日
最后更新2026年9月18日
分类编程语言
定价开源

> 相关工具

T
TypeScript
JavaScript 的超集,为前端与全栈提供静态类型
P
Python
通用编程语言,广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言