Hi there, To my understanding, when a user process runs away and asks for too much memory, paging/swapping will happen, so either system performance gets a hit ( because of memory paging and the IO it caused ), or /tmp gets so shrinked that new process can't be started which renders unusable system.
So this is both a performance issue and a security issue to me. I understand with ulimit, plimit and prctl, virtual memory size, stack size and data size can be limited. But I am not aware of away to set a limit system wide. ulimit kind of depend on the user's own setting and plimit/prctl is run after the fact. So how do we set a cap at system level? Also with shared memory segment in the picture, how does it interact with the limits set above? So what would be the proper way of dealing with this kind of issue? It would be great if it also works for Solaris 9. Thanks, Sean This message posted from opensolaris.org _______________________________________________ perf-discuss mailing list perf-discuss@opensolaris.org
