NewDisplayName Posted December 20, 2017 Share Posted December 20, 2017 Hi, or how ever u wanna call it. I have unraid since 3 weeks, since the newest beta, i got 3 freezes (u can click everything, but nothing actually happens). I cant stop array, i cant stop docker, i cant restart. (reboot in sh also not working) I need to go to the server and restart it via click on the front paneel. But, the server is not near me, so i cant always go there. Could you please add a killswitch after clicking on reboot. Like, if user clicked reboot and system is not rebooted after X(20?) minutes, display a extra button like "Hard reboot, caution, data could be lost" and just kill -9 everything (or whatever, i guess u know what i mean). I know there could be data lost, but if i click on the button on the front, same would happen. Link to comment
pwm Posted December 20, 2017 Share Posted December 20, 2017 If reboot in the shell doesn't work then kill -9 isn't likely to work either, because you normally have some thread dead-locked in a critical section inside the kernel. And then you need to figure out if there is some magic possible to unlock that situation. But there isn't any magical interface for unlocking programs that are stuck in a kernel call. On Windows the corresponding work-around implemented is that the kernel tries to signal the PSU to turn off if the kernel hasn't manage to kill all running programs within a reasonable time frame. Link to comment
NewDisplayName Posted December 20, 2017 Author Share Posted December 20, 2017 Im no linux guru. But if i can reach ssh, and manually kill the processes, then i guess, unraid could do the same htop f9 enter. Link to comment
pwm Posted December 20, 2017 Share Posted December 20, 2017 Did you have a situation where you could manually kill processes but shutdown could not? Link to comment
NewDisplayName Posted December 20, 2017 Author Share Posted December 20, 2017 I tried to stop docker. I tried to stop array. I tried reboot. Then i logged into ssh and could kill the processes which used high cpu, but they got restartet. then i thought, lets try "reboot" in ssh. Which killed the sshd So, i cant be complete sure, but i guess, since the processes were able to get killed, if you just kill all, it could have worked, or maybe theres a better option then kill. idc. Maybe like at windows, send a cmd to psu?! Link to comment
Recommended Posts
Archived
This topic is now archived and is closed to further replies.