Description of problem: Site is experiencing a known issue with Docker that was resolved in Docker v1.13, however site is on OCP v3.6, which has no plan on utilizing Docker v1.13 and site is requesting a Red Hat verified solution to the Zombie Reaping issue reported here: https://blog.phusion.nl/2015/01/20/docker-and-the-pid-1-zombie-reaping-problem/ and resolved by Docker using Tini, reported here: https://github.com/krallin/tini#using-tini Version-Release number of selected component (if applicable): OCP 3.6 How reproducible: Verified by Customer and Docker Steps to Reproduce: See https://blog.phusion.nl/2015/01/20/docker-and-the-pid-1-zombie-reaping-problem/ for more information Actual results: Zombie processes are utilizing a large portion of allocated CPU and Memory for site but no known supported solution Expected results: Site would like to be able to kill the affected processes and better utilize their CPU and Memory Additional info: Issue: https://blog.phusion.nl/2015/01/20/docker-and-the-pid-1-zombie-reaping-problem/ Fix: https://github.com/krallin/tini#using-tini
Red Hat's supported solution in this case is to run systemd in the container Is running systemd inside of a docker container supported? https://access.redhat.com/solutions/1118173 You can configure the registry.access.redhat.com/rhel7 image as described in the Kbase solution, or you can use the registry.access.redhat.com/rhel7-init image which is already configured to run systemd.
Please re-open if running systemd fails to address the issue.
*** Bug 1585289 has been marked as a duplicate of this bug. ***