Guido, Alas, this appears to be the problem. /cgroups/cgroup.clone_children does not exist.
Thanks for the assistance, ill likely begin testing with a new kernel. A workaround for this problem in RHEL/Centos that sidesteps the kernel recompile issue is to permit containers to have access to all CPUs. this will result in oversubscription however. On 2013-10-10 12:40, Guido Jäkel wrote: > On 2013-10-10 18:34, John wrote: >> Guido, Serge >> Thank you for your insight. the parent, /cgroup/cpuset/lxc, has >> cpuset.cpus however it hasnt been initialized to any number. my >> hypothesis is /cgroup/cpuset/lxc/GE/cpuset.cpus therefore cannot be >> set as its parent hasnt been initialized. >> >> im uncertain if clone_children is present, however i would suspect it >> is or else the qemu_kvm package for centos would run into serious >> problems. >> is there a means i can use to check the presence of this flag without >> fetching the source? > > Dear Jon, > > the pseudo-file cgroup.clone_children have to appear in the same way > as cpuset.cpus or others. You may read it with 'cat > /cgroups/cgroup.clone_childern' and set it via 'echo 1 > >/cgroups/cgroup.clone_childern' > > greetings > > Guido ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk _______________________________________________ Lxc-users mailing list Lxc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-users
