vanhauser-thc/thc-hydra

[BUG] When resuming a session, finished tries count towards new session tries - messing up ETA

Open

#824 创建于 2023年2月19日

在 GitHub 查看
 (0 评论) (0 反应) (0 负责人)C (8,689 star) (1,913 fork)batch import
best practicegood first issuehelp wanted

描述

I left a session running for about 10 minutes during which ~9000 tries were done. Then I remebered to add -f option so I stopped the execution and resumed it using -R with the added option. Now after 1 minute I get: [STATUS] 10166.00 tries/min, 10166 tries in 00:01h, 989834 to do in 01:38h, 32 active even though the actual speed is somewhere around(from the previous session): [STATUS] 634.00 tries/min, 634 tries in 00:01h, 999366 to do in 26:17h, 32 active

With time due to averaging it does converge towards the actual speed but I believe this is undesired.

贡献者指南

技术栈
c
领域
backendsecurity
议题类型
bug
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
4
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
1-2 days
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
fresh
清晰度议题是否清楚说明期望改动、验收标准和下一步。
mostly clear
前置要求
C programminghydra session system
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
35
研究方向
Investigate the session resume logic in thc hydra, likely in the main source files such as hydra session.c or hydra core.c. The bug seems to be that the finished tries from a previous session are not reset or excluded when resuming with new options, causing the ETA calculation to use inflated try counts. Look for how the 'tries' counter is initialized on resume and how the statistics module accumulates data over time. The fix might involve resetting the finished tries counter or properly adjusting the ETA calculation to consider only new tries.