What problem can cause kill -9 in a production application (more precisely, on Linux)?
I have an application that does some periodic work, it takes a lot of time to stop, and I don't care if some tasks are canceled - the work can be completed by new processes. Can I use kill -9 to stop it immediately, or can it cause serious OS problems?
For example, Unicorn uses it as a normal working procedure:
When your application goes awry, BOFH can just โkill -9โ the fluent workflow, without worrying about breaking all clients, only one.
But this article states:
The -9 (or KILL) argument to kill (1) should never be used on Unix systems
PS: I understand that the kill -9 application cannot be processed by the application, but I know that it does not cause any problems for the application-application, I'm just wondering if this can cause some problems at the OS level? shared memory segments active, lingering sockets sounds dangerous to me.
linux operating-system kill unicorn
valodzka
source share