Hacker News Reader

I've struggled with, and am still struggling with, units hanging on shutdown. It's my biggest complaint with systemD aside from journald eating its own logs.

I've been using magic SysRq to force hanging units to quit. SysRq + e sends SIGINT to all processes, and SysRq + i sends SIGKILL.

View on HN

Sending SIGINT to all processes isn't what is needed here. It needs to go to the "processes" that are currently going on. With systemd's parallel execution that can still be narrowed down in a useful way, especially is some tail of some hanging command is running alone by now.

Obviously what the BSDs are doing, sending SIGQUIT to just the currently blocking-the-shutdown processes, is far superior. Now the coredumps are floating around the filesystem and they'll be there after reboot, and you can see who did what and was hanging in what backtrace.

Reply on news.ycombinator.com