File system of container corrupts randomly

Hello All,

We are creating image for our product and then deploy it in different AWS EC2 instances. We deploy the same image in different Instances ( Environments) and proceed with our work.

Recently we have seen that in some of the Environments, once image is deployed, its file system gets corrupted(File system inside the container). However host machine file system is still good. Destroying and recreating the container did not help.

Instance is of centos, and application is based on java.
Same image when deployed in different environment, works fine.

We have faced this multiple times now, and only thing helps is recreating EC2 instance.

I understand it can be issue with AWS instance as well, but as I have got not clue yet, posting it here.

Has any once faced this issue.


Were you able to solve it? It looks like your files are still there, they just cannot be accessed. Have you checked /var/lib/docker as root from the host (EC2 instance)?

Smells like selinux issues. It might be related to what is discussed in this RHEL bug post: 1382997 – Unable to start container with SELinux enabled

At least it should give you an idea how to pin-point if the selinux issue applies to you and might provide ideas how to solve it.