Hi all,
I hope this is the correct forum to ask this question. I have a question regarding process resource usage limitation via setrlimit() as it relates to RLIMIT_CPU. In the setrlimit manpage, it says the following about RLIMIT_CPU:
My question is, is this to be interpreted as total aggregate CPU usage time of the process or one-shot usage (i.e. if the process uses CPU for more than x secs uninterrupted, then send SIGXCPU)?
More concretely, the getty process sets a 60 second limit (GETTY_TIMEOUT=60) via below code, and I am trying to understand if getty will call timeoverrun signal handler when it has consumed 60 seconds of aggregate CPU time or 60 seconds at a single time.
Thank you in advance for you help in clarifying this point!
Raku
I hope this is the correct forum to ask this question. I have a question regarding process resource usage limitation via setrlimit() as it relates to RLIMIT_CPU. In the setrlimit manpage, it says the following about RLIMIT_CPU:
My question is, is this to be interpreted as total aggregate CPU usage time of the process or one-shot usage (i.e. if the process uses CPU for more than x secs uninterrupted, then send SIGXCPU)?
More concretely, the getty process sets a 60 second limit (GETTY_TIMEOUT=60) via below code, and I am trying to understand if getty will call timeoverrun signal handler when it has consumed 60 seconds of aggregate CPU time or 60 seconds at a single time.
Code:
/*
* Limit running time to deal with broken or dead lines.
*/
(void)signal(SIGXCPU, timeoverrun);
limit.rlim_max = RLIM_INFINITY;
limit.rlim_cur = GETTY_TIMEOUT;
(void)setrlimit(RLIMIT_CPU, &limit);Thank you in advance for you help in clarifying this point!
Raku
