3 ms·
> if the kernel discovers a crash, the `fork()` syscall of the parent process is blocked for 30 seconds for each failed attempt That seems like something good
by JoshTriplett 8y ago
> if the kernel discovers a crash, the `fork()` syscall of the parent process is blocked for 30 seconds for each failed attempt
That seems like something good to be able to turn on in a stock kernel, but not with that high a timeout. Imagine your shell failing to start another process for 30 seconds after you're debugging a segfault, or your browser failing to open another tab for 30 seconds after one crash.
500ms would drastically slow brute-force attacks without noticeably inconveniencing the user (and then they can always turn it off manually when doing something like fuzz testing).
- pas 8y agojournald is a systemd service, right? cannot restartsec simply increased to 1s?
- floatingatoll 8y agoI've waited almost 15 minutes for an ssh(1) to a server to complete, with a latency of 1-2 minutes per command. 30 seconds, predictably? I'd take it any day over that.