Hi,
I have an issue with an apache based image hosted on AWS Elastic Beanstalk (64bit Amazon Linux 2016.03 v2.1.1 running Multi-container Docker 1.9.1), I try to explain as best I can the problem:
sometime apache2 process reach the memory limit and oom killer send SIGKILL to that process, Docker container exit with 137 error code:
it seems that the Apache process does not exit and continues to run keeping the container frozed.
In this conditions Elastic Beanstalk try to restart the ECS task but Docker Container do not respond, this cause a freeze of entire environment without being able to do anything except “rebuild environment”, is a really annoying situation.
How can I prevent this? Is a Docker issue or maybe it is the fault of apache?
Unfortunately no, my workaround was increase instance type. In this way apache not reach the memory limit but is not a real solution, I hope in a Docker or Apache update that fix this problem.