Skip to content
This repository was archived by the owner on Oct 16, 2020. It is now read-only.
This repository was archived by the owner on Oct 16, 2020. It is now read-only.

Wiping root file system is not the same as a "factory reset" #2276

@ajeddeloh

Description

@ajeddeloh

Issue Report

Bug

Container Linux Version

$ cat /etc/os-release
NAME="Container Linux by CoreOS"
ID=coreos
VERSION=1590.0.0+2017-11-27-1556
VERSION_ID=1590.0.0
BUILD_ID=2017-11-27-1556
PRETTY_NAME="Container Linux by CoreOS 1590.0.0+2017-11-27-1556 (Ladybug)"
ANSI_COLOR="38;5;75"
HOME_URL="https://coreos.com/"
BUG_REPORT_URL="https://issues.coreos.com"
COREOS_BOARD="amd64-usr"

Environment

Any

Expected Behavior

The root filesystem we ship in the image should be the same as the one that is populated by tmpfiles. From our documentation on our disk layout:

Due to the unique disk layout of Container Linux, an rm -rf / is an un-supported but valid operation to do a "factory reset". The machine should boot and operate normally afterwards.

Actual Behavior

The root filesystem contains extra files and some differences as well. The differences are all in the /etc/g?{shadow,passwd}-? files, except /etc/audit/audit.rules which is generated by other files.

Reproduction Steps

  1. run the coreos.filesystem test from this PR

Other Information

Here is a list of all the files that are not created (at least on my current build, current alpha/beta/stable may differ slightly).

We should audit this list and determine if it is safe to ship an empty root.

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions