On Fri, 26 Aug 2011, Konstantin Khlebnikov wrote: > Maybe just fix this "panic" logic? OOM killer should panic only on global > memory shortage. > NO, it shouldn't, we actually rely quite extensively on cpusets filled with OOM_DISABLE threads to panic because the job scheduler would be unresponsive in such a condition and it'd much better to panic and reboot than to brick the machine. I'm not sure where you're getting all your information from, but please don't pass it off as principles. You can set the panic logic to be whatever you want with /proc/sys/vm/panic_on_oom. See Documentation/filesystems/proc.txt for more information. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>