Tag Info

New answers tagged

0

I think your problem could be two-fold. You appear to only be increasing the soft limit but not the hard limit. If the hard limit default is lower than the defined soft limit, I think you will continue to be constrained by the lower value. Also, these are the per-user limits but not necessarily the kernel's system-wide configured limit. You may need to set ...


1

Run the processes as a dedicated user and set it up via cgroups: /etc/cgconfig.conf: group limitedram { memory { memory.limit_in_bytes = 6442450944; } } and /etc/cgrules.conf: serviceuser memory limitedram/ That limits the memory usage of serviceuser to 6 GB.


3

You aren't finding anything because it doesn't exist. It is up to the application hitting the limit to log that a syscall failed to allocate a resource. Some quick googlefu led me to the Logging-limits.conf on github. It attempts to log when the syscalls fail with auditd. I have no experience with this package, it may destroy your system. Caveat emptor. ...


1

As @Michael Hampton told you, your syntax is wrong, but anyway i don't think is good idea to set the file limit to unlimited You can read this post for more information http://stackoverflow.com/questions/1212925/on-linux-set-maximum-open-files-to-unlimited-possible I downloaded the kernel version linux-2.6.32.61 and i saw this from kernel/sys.c: ...


1

You've reversed the second and third fields. It should read instead: root soft nofile -1 root hard nofile -1



Top 50 recent answers are included