CIS Docker Community Edition Benchmark v1.1.0

CIS Docker Community Edition Benchmark
v1.1.0 - 07-06-2017
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike
4.0 International Public License. The link to the license terms can be found at
https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode
To further clarify the Creative Commons license related to CIS Benchmark content, you are
authorized to copy and redistribute the content for use by you, within your organization
and outside your organization for non-commercial purposes only, provided that (i)
appropriate credit is given to CIS, (ii) a link to the license is provided. Additionally, if you
remix, transform or build upon the CIS Benchmark(s), you may only distribute the modified
materials if they are subject to the same license terms as the original Benchmark license
and your derivative will no longer be a CIS Benchmark. Commercial use of CIS Benchmarks
is subject to the prior approval of the Center for Internet Security.
1|Page
Table of Contents
Overview .................................................................................................................................................................. 8
Intended Audience ........................................................................................................................................... 8
Consensus Guidance........................................................................................................................................ 8
Typographical Conventions ......................................................................................................................... 9
Scoring Information ........................................................................................................................................ 9
Profile Definitions ......................................................................................................................................... 10
Acknowledgements ...................................................................................................................................... 11
Recommendations ............................................................................................................................................. 12
1 Host Configuration .................................................................................................................................... 12
1.1 Ensure a separate partition for containers has been created (Scored) .................. 12
1.2 Ensure the container host has been Hardened (Not Scored) ...................................... 14
1.3 Ensure Docker is up to date (Not Scored) .......................................................................... 16
1.4 Ensure only trusted users are allowed to control Docker daemon (Scored) ........ 18
1.5 Ensure auditing is configured for the docker daemon (Scored) ................................ 20
1.6 Ensure auditing is configured for Docker files and directories - /var/lib/docker
(Scored) ................................................................................................................................................... 22
1.7 Ensure auditing is configured for Docker files and directories - /etc/docker
(Scored) ................................................................................................................................................... 24
1.8 Ensure auditing is configured for Docker files and directories - docker.service
(Scored) ................................................................................................................................................... 26
1.9 Ensure auditing is configured for Docker files and directories - docker.socket
(Scored) ................................................................................................................................................... 28
1.10 Ensure auditing is configured for Docker files and directories -
/etc/default/docker (Scored) ......................................................................................................... 30
/etc/docker/daemon.json (Scored) ............................................................................................. 32
/usr/bin/docker-containerd (Scored) ........................................................................................ 34
2|Page
/usr/bin/docker-runc (Scored) ..................................................................................................... 36
2 Docker daemon configuration .............................................................................................................. 38
2.1 Ensure network traffic is restricted between containers on the default bridge
(Scored) ................................................................................................................................................... 38
2.2 Ensure the logging level is set to 'info' (Scored) .............................................................. 40
2.3 Ensure Docker is allowed to make changes to iptables (Scored) .............................. 42
2.4 Ensure insecure registries are not used (Scored) ........................................................... 44
2.5 Ensure aufs storage driver is not used (Scored) .............................................................. 46
2.6 Ensure TLS authentication for Docker daemon is configured (Scored) ................. 48
2.7 Ensure the default ulimit is configured appropriately (Not Scored) ....................... 50
2.8 Enable user namespace support (Scored) .......................................................................... 52
2.9 Ensure the default cgroup usage has been confirmed (Scored) ................................ 54
2.10 Ensure base device size is not changed until needed (Scored) ............................... 56
2.11 Ensure that authorization for Docker client commands is enabled (Scored) .... 58
2.12 Ensure centralized and remote logging is configured (Scored) .............................. 60
2.13 Ensure operations on legacy registry (v1) are Disabled (Scored) ......................... 62
2.14 Ensure live restore is Enabled (Scored) ........................................................................... 64
2.15 Ensure Userland Proxy is Disabled (Scored) .................................................................. 66
2.16 Ensure daemon-wide custom seccomp profile is applied, if needed (Not Scored)
..................................................................................................................................................................... 68
2.17 Ensure experimental features are avoided in production (Scored) ...................... 70
2.18 Ensure containers are restricted from acquiring new privileges (Scored) ........ 71
3 Docker daemon configuration files .................................................................................................... 73
3.1 Ensure that docker.service file ownership is set to root:root (Scored) .................. 73
3.2 Ensure that docker.service file permissions are set to 644 or more restrictive
(Scored) ................................................................................................................................................... 75
3.3 Ensure that docker.socket file ownership is set to root:root (Scored).................... 77
3.4 Ensure that docker.socket file permissions are set to 644 or more restrictive
(Scored) ................................................................................................................................................... 79
3.5 Ensure that /etc/docker directory ownership is set to root:root (Scored) .......... 81
3|Page
3.6 Ensure that /etc/docker directory permissions are set to 755 or more restrictive
(Scored) ................................................................................................................................................... 83
3.7 Ensure that registry certificate file ownership is set to root:root (Scored) .......... 85
3.8 Ensure that registry certificate file permissions are set to 444 or more restrictive
(Scored) ................................................................................................................................................... 87
3.9 Ensure that TLS CA certificate file ownership is set to root:root (Scored) ............ 89
3.10 Ensure that TLS CA certificate file permissions are set to 444 or more
restrictive (Scored) ............................................................................................................................. 91
3.11 Ensure that Docker server certificate file ownership is set to root:root (Scored)
..................................................................................................................................................................... 93
3.12 Ensure that Docker server certificate file permissions are set to 444 or more
restrictive (Scored) ............................................................................................................................. 95
3.13 Ensure that Docker server certificate key file ownership is set to root:root
(Scored) ................................................................................................................................................... 97
3.14 Ensure that Docker server certificate key file permissions are set to 400
(Scored) ................................................................................................................................................... 99
3.15 Ensure that Docker socket file ownership is set to root:docker (Scored) ......... 101
3.16 Ensure that Docker socket file permissions are set to 660 or more restrictive
(Scored) ................................................................................................................................................. 103
3.17 Ensure that daemon.json file ownership is set to root:root (Scored) ................. 105
3.18 Ensure that daemon.json file permissions are set to 644 or more restrictive
(Scored) ................................................................................................................................................. 107
3.19 Ensure that /etc/default/docker file ownership is set to root:root (Scored) .. 109
3.20 Ensure that /etc/default/docker file permissions are set to 644 or more
restrictive (Scored) ........................................................................................................................... 111
4 Container Images and Build File ........................................................................................................ 113
4.1 Ensure a user for the container has been created (Scored) ...................................... 113
4.2 Ensure that containers use trusted base images (Not Scored) ................................ 115
4.3 Ensure unnecessary packages are not installed in the container (Not Scored). 117
4.4 Ensure images are scanned and rebuilt to include security patches (Not Scored)
................................................................................................................................................................... 119
4.5 Ensure Content trust for Docker is Enabled (Scored) .................................................. 121
4|Page
4.6 Ensure HEALTHCHECK instructions have been added to the container image
(Scored) ................................................................................................................................................. 123
4.7 Ensure update instructions are not use alone in the Dockerfile (Not Scored) ... 125
4.8 Ensure setuid and setgid permissions are removed in the images (Not Scored)
................................................................................................................................................................... 127
4.9 Ensure COPY is used instead of ADD in Dockerfile (Not Scored) ............................ 129
4.10 Ensure secrets are not stored in Dockerfiles (Not Scored) ..................................... 131
4.11 Ensure verified packages are only Installed (Not Scored)....................................... 133
5 Container Runtime .................................................................................................................................. 135
5.1 Ensure AppArmor Profile is Enabled (Scored) ............................................................... 135
5.2 Ensure SELinux security options are set, if applicable (Scored).............................. 137
5.3 Ensure Linux Kernel Capabilities are restricted within containers (Scored) ..... 139
5.4 Ensure privileged containers are not used (Scored) .................................................... 142
5.5 Ensure sensitive host system directories are not mounted on containers
(Scored) ................................................................................................................................................. 144
5.6 Ensure ssh is not run within containers (Scored) ......................................................... 146
5.7 Ensure privileged ports are not mapped within containers (Scored) ................... 148
5.8 Ensure only needed ports are open on the container (Scored) ............................... 150
5.9 Ensure the host's network namespace is not shared (Scored) ................................ 152
5.10 Ensure memory usage for container is limited (Scored) ......................................... 154
5.11 Ensure CPU priority is set appropriately on the container (Scored)................... 156
5.12 Ensure the container's root filesystem is mounted as read only (Scored) ....... 158
5.13 Ensure incoming container traffic is binded to a specific host interface (Scored)
................................................................................................................................................................... 161
5.14 Ensure 'on-failure' container restart policy is set to '5' (Scored) ......................... 163
5.15 Ensure the host's process namespace is not shared (Scored) ............................... 165
5.16 Ensure the host's IPC namespace is not shared (Scored) ........................................ 167
5.17 Ensure host devices are not directly exposed to containers (Not Scored) ....... 169
5.18 Ensure the default ulimit is overwritten at runtime, only if needed (Not Scored)
................................................................................................................................................................... 171
5.19 Ensure mount propagation mode is not set to shared (Scored) ........................... 173
5.20 Ensure the host's UTS namespace is not shared (Scored) ....................................... 175
5|Page
5.21 Ensure the default seccomp profile is not Disabled (Scored) ................................ 177
5.22 Ensure docker exec commands are not used with privileged option (Scored)
................................................................................................................................................................... 179
5.23 Ensure docker exec commands are not used with user option (Scored) .......... 181
5.24 Ensure cgroup usage is confirmed (Scored) ................................................................. 183
5.25 Ensure the container is restricted from acquiring additional privileges (Scored)
................................................................................................................................................................... 185
5.26 Ensure container health is checked at runtime (Scored) ......................................... 187
5.27 Ensure docker commands always get the latest version of the image (Not
Scored) ................................................................................................................................................... 189
5.28 Ensure PIDs cgroup limit is used (Scored) .................................................................... 191
5.29 Ensure Docker's default bridge docker0 is not used (Not Scored) ...................... 193
5.30 Ensure the host's user namespaces is not shared (Scored) .................................... 195
5.31 Ensure the Docker socket is not mounted inside any containers (Scored) ...... 197
6 Docker Security Operations................................................................................................................. 199
6.1 Ensure image sprawl is avoided (Not Scored) ................................................................ 199
6.2 Ensure container sprawl is avoided (Not Scored) ......................................................... 202
7 Docker Swarm Configuration.............................................................................................................. 204
7.1 Ensure swarm mode is not Enabled, if not needed (Scored) .................................... 204
7.2 Ensure the minimum number of manager nodes have been created in a swarm
(Scored) ................................................................................................................................................. 206
7.3 Ensure swarm services are binded to a specific host interface (Scored) ............. 208
7.4 Ensure data exchanged between containers are encrypted on different nodes on
the overlay network (Scored) ....................................................................................................... 210
7.5 Ensure Docker's secret management commands are used for managing secrets
in a Swarm cluster (Not Scored) .................................................................................................. 212
7.6 Ensure swarm manager is run in auto-lock mode (Scored) ...................................... 213
7.7 Ensure swarm manager auto-lock key is rotated periodically (Not Scored) ...... 215
7.8 Ensure node certificates are rotated as appropriate (Not Scored) ......................... 217
7.9 Ensure CA certificates are rotated as appropriate (Not Scored) ............................. 219
7.10 Ensure management plane traffic has been separated from data plane traffic
(Not Scored) ......................................................................................................................................... 221
6|Page
Appendix: Summary Table ........................................................................................................................... 223
Appendix: Change History ............................................................................................................................ 228
7|Page
Overview
This document is intended for system and application administrators, security specialists,
auditors, help desk, and platform deployment personnel who plan to develop, deploy,
assess, or secure solutions that incorporate Docker CE 17.06 or later technology.
Intended Audience
This document, CIS Docker CE 17.06 Benchmark, provides prescriptive guidance for
establishing a secure configuration posture for Docker CE container version 17.06. This
guide was tested against Docker CE 17.06 on RHEL 7 and Debian 8. To obtain the latest
version of this guide, please visit http://benchmarks.cisecurity.org. If you have questions,
comments, or have identified ways to improve this guide, please write us at
[email protected].
Consensus Guidance
This benchmark was created using a consensus review process comprised of subject
matter experts. Consensus participants provide perspective from a diverse set of
backgrounds including consulting, software development, audit and compliance, security
research, operations, government, and legal.
Each CIS benchmark undergoes two phases of consensus review. The first phase occurs
during initial benchmark development. During this phase, subject matter experts convene
to discuss, create, and test working drafts of the benchmark. This discussion occurs until
consensus has been reached on benchmark recommendations. The second phase begins
after the benchmark has been published. During this phase, all feedback provided by the
Internet community is reviewed by the consensus team for incorporation in the
benchmark. If you are interested in participating in the consensus process, please visit
https://community.cisecurity.org.
8|Page
Typographical Conventions
The following typographical conventions are used throughout this guide:
Convention Meaning
Stylized Monospace font Used for blocks of code, command, and script examples.
Text should be interpreted exactly as presented.
Monospace font Used for inline code, commands, or examples. Text should
be interpreted exactly as presented.
<italic font in brackets> Italic texts set in angle brackets denote a variable
requiring substitution for a real value.
Italic font Used to denote the title of a book, article, or other

publication.
Note Additional information or caveats
Scoring Information
A scoring status indicates whether compliance with the given recommendation impacts the
assessed target's benchmark score. The following scoring statuses are used in this
benchmark:
Scored
Failure to comply with "Scored" recommendations will decrease the final benchmark score.
Compliance with "Scored" recommendations will increase the final benchmark score.
Not Scored
Failure to comply with "Not Scored" recommendations will not decrease the final
benchmark score. Compliance with "Not Scored" recommendations will not increase the
final benchmark score.
9|Page
Profile Definitions
The following configuration profiles are defined by this Benchmark:
 Level 1 - Docker
Items in this profile intend to:
o Be practical and prudent;

o Provide a clear security benefit; and
o Not inhibit the utility of the technology beyond acceptable means.
 Level 1 - Linux Host OS
Items in this profile pertain to the Linux Host OS and intend to:
o Be practical and prudent;

o Provide a clear security benefit; and
o Not inhibit the utility of the technology beyond acceptable means.
Items in this profile exhibit one or more of the following characteristics:
o Are intended for environments or use cases where security is paramount

o Acts as defense in depth measure
o May negatively inhibit the utility or performance of the technology
10 | P a g e
Acknowledgements
This benchmark exemplifies the great things a community of users, vendors, and subject matter
experts can accomplish through consensus collaboration. The CIS community thanks the entire
consensus team with special recognition to the following individuals who contributed greatly to
the creation of this guide:
Author
Pravin Goyal
Editor
Thomas Sjögren
Rory McCune, NCC Group PLC
Contributor
Brian Andrzejewski
Jordan Rakoske GSEC, GCWN
11 | P a g e
Recommendations
1 Host Configuration
This section covers security recommendations that you should follow to prepare the host
machine that you plan to use for executing containerized workloads. Securing the Docker
host and following your infrastructure security best practices would build a solid and
secure foundation for executing containerized workloads.
1.1 Ensure a separate partition for containers has been created (Scored)
Profile Applicability:
Description:
All Docker containers and their data and metadata is stored under /var/lib/docker
directory. By default, /var/lib/docker would be mounted under / or /var partitions based
on availability.
Rationale:
Docker depends on /var/lib/docker as the default directory where all Docker related files,
including the images, are stored. This directory might fill up fast and soon Docker and the
host could become unusable. So, it is advisable to create a separate partition (logical
volume) for storing Docker files.
Audit:
At the Docker host execute the below command:
grep /var/lib/docker /etc/fstab
This should return the partition details for /var/lib/docker mount point.
Remediation:
For new installations, create a separate partition for /var/lib/docker mount point. For
systems that were previously installed, use the Logical Volume Manager (LVM) to create
partitions.
12 | P a g e
Impact:
None.
Default Value:
By default, /var/lib/docker would be mounted under / or /var partitions based on

availability.
References:
1. https://www.projectatomic.io/docs/docker-storage-recommendation/
CIS Controls:
14 Controlled Access Based on the Need to Know

Controlled Access Based on the Need to Know
13 | P a g e
1.2 Ensure the container host has been Hardened (Not Scored)
Description:
Containers run on a Linux host. A container host can run one or more containers. It is of
utmost importance to harden the host to mitigate host security misconfiguration.
Rationale:
You should follow infrastructure security best practices and harden your host OS. Keeping
the host system hardened would ensure that the host vulnerabilities are mitigated. Not
hardening the host system could lead to security exposures and breaches.
Audit:
Ensure that the host specific security guidelines are followed. Ask the system
administrators which security benchmark does current host system comply with. Ensure
that the host systems actually comply with that host specific security benchmark.
Remediation:
You may consider various CIS Security Benchmarks for your container host. If you have
other security guidelines or regulatory requirements to adhere to, please follow them as
suitable in your environment.
Additionally, you can run a kernel with grsecurity and PaX. This would add many safety
checks, both at compile-time and run-time. It is also designed to defeat many exploits and
has powerful security features. These features do not require Docker-specific
configuration, since those security features apply system-wide, independent of containers.
Impact:
None.
Default Value:
By default, host has factory settings. It is not hardened.
References:
1. https://docs.docker.com/engine/security/security/
14 | P a g e
2. https://learn.cisecurity.org/benchmarks
3. https://docs.docker.com/engine/security/security/#other-kernel-security-features
4. https://grsecurity.net/
5. https://en.wikibooks.org/wiki/Grsecurity
6. https://pax.grsecurity.net/
7. http://en.wikipedia.org/wiki/PaX
CIS Controls:
3 Secure Configurations for Hardware and Software on Mobile Devices, Laptops,

Workstations, and Servers
Secure Configurations for Hardware and Software on Mobile Devices, Laptops,
15 | P a g e
1.3 Ensure Docker is up to date (Not Scored)
Description:
There are frequent releases for Docker software that address security vulnerabilities,
product bugs and bring in new functionality. Keep a tab on these product updates and
upgrade as frequently as when new security vulnerabilities are fixed or deemed correct for
your organization.
Rationale:
By staying up to date on Docker updates, vulnerabilities in the Docker software can be

mitigated. An educated attacker may exploit known vulnerabilities when attempting to
attain access or elevate privileges. Not installing regular Docker updates may leave you
with running vulnerable Docker software. It might lead to elevation privileges,
unauthorized access or other security breaches. Keep a track of new releases and update as
necessary.
Audit:
Execute the below command and verify that the Docker version is up to date as deemed
necessary. It is not a mandate to be on the latest one, though.
docker version
Remediation:
Keep a track of Docker releases and update as necessary.
Impact:
Perform a risk assessment for docker version updates in how they may impact your Docker
operations. Be aware that several third-party products that use Docker may require older
major version of Docker to be supported.
Default Value:
Not Applicable
16 | P a g e
References:
1. https://docs.docker.com/engine/installation/
2. https://github.com/moby/moby/releases/latest
3. https://github.com/docker/docker-ce/releases/latest
CIS Controls:
4 Continuous Vulnerability Assessment and Remediation

Continuous Vulnerability Assessment and Remediation
17 | P a g e
1.4 Ensure only trusted users are allowed to control Docker daemon
(Scored)
Description:
The Docker daemon currently requires root privileges. A user added to the docker group
gives him full root access rights.
Rationale:
Docker allows you to share a directory between the Docker host and a guest container
without limiting the access rights of the container. This means that you can start a
container and map the / directory on your host to the container. The container will then be
able to alter your host file system without any restrictions. In simple terms, it means that
you can attain elevated privileges with just being a member of the docker group and then
starting a container with mapped / directory on the host.
Audit:
Execute the below command on the docker host and ensure that only trusted users are
members of the docker group.
getent group docker
Remediation:
Remove any users from the docker group that are not trusted. Additionally, do not create a
mapping of sensitive directories on host to container volumes.
Impact:
Rights to build and execute containers as normal user would be restricted.
Default Value:
Not Applicable
18 | P a g e
References:
1. https://docs.docker.com/engine/security/security/#docker-daemon-attack-
surface
2. https://www.andreas-jung.com/contents/on-docker-security-docker-group-
considered-harmful
3. http://www.projectatomic.io/blog/2015/08/why-we-dont-let-non-root-users-run-
docker-in-centos-fedora-or-rhel/
CIS Controls:
5.1 Minimize And Sparingly Use Administrative Privileges

Minimize administrative privileges and only use administrative accounts when they are
required. Implement focused auditing on the use of administrative privileged functions and
monitor for anomalous behavior.
19 | P a g e
1.5 Ensure auditing is configured for the docker daemon (Scored)
Description:
Audit all Docker daemon activities.
Rationale:
Apart from auditing your regular Linux file system and system calls, audit Docker daemon
as well. Docker daemon runs with root privileges. It is thus necessary to audit its activities
and usage.
Audit:
Verify that there is an audit rule for Docker daemon. For example, execute below
command:
auditctl -l | grep /usr/bin/docker
This should list a rule for Docker daemon.
Remediation:
Add a rule for Docker daemon.
For example,
Add the line as below line in /etc/audit/audit.rules file:
-w /usr/bin/docker -k docker
Then, restart the audit daemon. For example,
service auditd restart
Impact:
Auditing generates quite big log files. Ensure to rotate and archive them periodically. Also,
create a separate partition of audit to avoid filling root file system.
Default Value:
By default, Docker daemon is not audited.

20 | P a g e
References:
1. https://access.redhat.com/documentation/en-
US/Red_Hat_Enterprise_Linux/6/html/Security_Guide/chap-system_auditing.html
CIS Controls:
6.2 Ensure Audit Log Settings Support Appropriate Log Entry Formatting
Validate audit log settings for each hardware device and the software installed on it,
ensuring that logs include a date, timestamp, source addresses, destination addresses, and
various other useful elements of each packet and/or transaction. Systems should record
logs in a standardized format such as syslog entries or those outlined by the Common Event
Expression initiative. If systems cannot generate logs in a standardized format, log
normalization tools can be deployed to convert logs into such a format.
21 | P a g e
/var/lib/docker (Scored)
Description:
Audit /var/lib/docker.
Rationale:
Apart from auditing your regular Linux file system and system calls, audit all Docker
related files and directories. Docker daemon runs with root privileges. Its behavior
depends on some key files and directories. /var/lib/docker is one such directory. It holds
all the information about containers. It must be audited.
Audit:
Verify that there is an audit rule corresponding to /var/lib/docker directory.
For example, execute below command:
auditctl -l | grep /var/lib/docker
This should list a rule for /var/lib/docker directory.
Remediation:
Add a rule for /var/lib/docker directory.
For example,
Add the line as below in /etc/audit/audit.rules file:
-w /var/lib/docker -k docker
Impact:
22 | P a g e
Default Value:
By default, Docker related files and directories are not audited.
References:
CIS Controls:
14.6 Enforce Detailed Audit Logging For Sensitive Information

Enforce detailed audit logging for access to nonpublic data and special authentication for
sensitive data.
23 | P a g e
/etc/docker (Scored)
Description:
Audit /etc/docker.
Rationale:
depends on some key files and directories. /etc/docker is one such directory. It holds
various certificates and keys used for TLS communication between Docker daemon and
Docker client. It must be audited.
Audit:
Verify that there is an audit rule corresponding to /etc/docker directory.
auditctl -l | grep /etc/docker
This should list a rule for /etc/docker directory.
Remediation:
Add a rule for /etc/docker directory.
For example,
-w /etc/docker -k docker
24 | P a g e
Impact:
Default Value:
By default, Docker related files and directories are not audited.
References:
CIS Controls:

sensitive data.
25 | P a g e
docker.service (Scored)
Description:
Audit docker.service, if applicable.
Rationale:
depends on some key files and directories. docker.service is one such file. The
docker.service file might be present if the daemon parameters have been changed by an
administrator. It holds various parameters for Docker daemon. It must be audited, if
applicable.
Audit:
Step 1: Find out the file location:
systemctl show -p FragmentPath docker.service
Step 2: If the file does not exist, this recommendation is not applicable. If the file exists,
verify that there is an audit rule corresponding to the file:
For example, execute the below command:
auditctl -l | grep docker.service
This should list a rule for docker.service as per its location.
Remediation:
If the file exists, add a rule for it.
For example,
-w /usr/lib/systemd/system/docker.service -k docker
26 | P a g e
Impact:
Default Value:
By default, Docker related files and directories are not audited. The file docker.service
may not be available on the system.
References:
CIS Controls:

sensitive data.
27 | P a g e
docker.socket (Scored)
Description:
Audit docker.socket, if applicable.
Rationale:
depends on some key files and directories. docker.socket is one such file. It holds various
parameters for Docker daemon socket. It must be audited, if applicable.
Audit:
systemctl show -p FragmentPath docker.socket
verify that there is an audit rule corresponding to the file:
For example, execute the below command:
auditctl -l | grep docker.socket
This should list a rule for docker.socket as per its location.
Remediation:
If the file exists, add a rule for it.
For example,
-w /usr/lib/systemd/system/docker.socket -k docker
28 | P a g e
Impact:
Default Value:
By default, Docker related files and directories are not audited. The file docker.socket may
not be available on the system.
References:
CIS Controls:

sensitive data.
29 | P a g e
/etc/default/docker (Scored)
Description:
Audit /etc/default/docker, if applicable.
Rationale:
depends on some key files and directories. /etc/default/docker is one such file. It holds
various parameters for Docker daemon. It must be audited, if applicable.
Audit:
Verify that there is an audit rule corresponding to /etc/default/docker file.
auditctl -l | grep /etc/default/docker
This should list a rule for /etc/default/docker file.
Remediation:
Add a rule for /etc/default/docker file.
For example,
-w /etc/default/docker -k docker
Impact:
30 | P a g e
Default Value:
By default, Docker related files and directories are not audited. The file
/etc/default/docker may not be available on the system. In that case, this
recommendation is not applicable.
References:
CIS Controls:

sensitive data.
31 | P a g e
/etc/docker/daemon.json (Scored)
Description:
Audit /etc/docker/daemon.json, if applicable.
Rationale:
depends on some key files and directories. /etc/docker/daemon.json is one such file. It
holds various parameters for Docker daemon. It must be audited, if applicable.
Audit:
Verify that there is an audit rule corresponding to /etc/docker/daemon.json file.
auditctl -l | grep /etc/docker/daemon.json
This should list a rule for /etc/docker/daemon.json file.
Remediation:
Add a rule for /etc/docker/daemon.json file.
For example,
-w /etc/docker/daemon.json -k docker
Impact:
32 | P a g e
Default Value:
By default, Docker related files and directories are not audited. The file
/etc/docker/daemon.json may not be available on the system. In that case, this
recommendation is not applicable.
References:
2. https://docs.docker.com/engine/reference/commandline/dockerd/#daemon-
configuration-file
CIS Controls:

sensitive data.
33 | P a g e
/usr/bin/docker-containerd (Scored)
Description:
Audit /usr/bin/docker-containerd, if applicable.
Rationale:
depends on some key files and directories. /usr/bin/docker-containerd is one such file.
Docker now relies on containerdand runC to spawn containers. It must be audited, if
applicable.
Audit:
Verify that there is an audit rule corresponding to /usr/bin/docker-containerd file.
auditctl -l | grep /usr/bin/docker-containerd
This should list a rule for /usr/bin/docker-containerd file.
Remediation:
Add a rule for /usr/bin/docker-containerd file.
For example,
-w /usr/bin/docker-containerd -k docker
34 | P a g e
Impact:
Default Value:
By default, Docker related files and directories are not audited. The file /usr/bin/docker-
containerd may not be available on the system. In that case, this recommendation is not
applicable.
References:
2. https://github.com/docker/docker/pull/20662
3. https://containerd.tools/
CIS Controls:

sensitive data.
35 | P a g e
/usr/bin/docker-runc (Scored)
Description:
Audit /usr/bin/docker-runc, if applicable.
Rationale:
depends on some key files and directories. /usr/bin/docker-runc is one such file. Docker
now relies on containerd and runC to spawn containers. It must be audited, if applicable.
Audit:
Verify that there is an audit rule corresponding to /usr/bin/docker-runc file.
auditctl -l | grep /usr/bin/docker-runc
This should list a rule for /usr/bin/docker-runc file.
Remediation:
Add a rule for /usr/bin/docker-runc file.
For example,
-w /usr/bin/docker-runc -k docker
Impact:
36 | P a g e
Default Value:
By default, Docker related files and directories are not audited. The file/usr/bin/docker-
runc may not be available on the system. In that case, this recommendation is not
applicable.
References:
3. https://containerd.tools/
4. https://github.com/opencontainers/runc
CIS Controls:

sensitive data.
37 | P a g e
2 Docker daemon configuration
This section lists the recommendations that alter and secure the behavior of the Docker
daemon. The settings that are under this section affect ALL container instances.
Note: Docker daemon options can also be controlled using files such as
/etc/sysconfig/docker, /etc/default/docker, the systemd unit file or
/etc/docker/daemon.json. Also, note that Docker in daemon mode can be identified as
/usr/bin/dockerd, or having -d or daemon as the argument to docker service.
2.1 Ensure network traffic is restricted between containers on the

default bridge (Scored)
Description:
By default, all network traffic is allowed between containers on the same host on the
default network bridge. If not desired, restrict all the inter-container communication. Link
specific containers together that require communication. Alternatively, you can create
custom network and only join containers that need to communicate to that custom
network.
Rationale:
By default, unrestricted network traffic is enabled between all containers on the same host
on the default network bridge. Thus, each container has the potential of reading all packets
across the container network on the same host. This might lead to an unintended and
unwanted disclosure of information to other containers. Hence, restrict the inter-container
communication on the default network bridge.
Audit:
Run the below command and verify that the default network bridge has been configured to
restrict inter-container communication.
docker network ls --quiet | xargs docker network inspect --format '{{ .Name
}}: {{ .Options }}'
It should return com.docker.network.bridge.enable_icc:false for the default network

bridge.
38 | P a g e
Remediation:
Run the docker in daemon mode and pass --icc=false as an argument.
For Example,
dockerd --icc=false
Alternatively, you can follow the Docker documentation and create a custom network and
only join containers that need to communicate to that custom network. The --icc
parameter only applies to the default docker bridge, if custom networks are used then the
approach of segmenting networks should be adopted instead.
Impact:
The inter-container communication would be disabled on the default network bridge. If

any communication between containers on the same host is desired, then it needs to be
explicitly defined using container linking or alternatively custom networks have to be
defined.
Default Value:
By default, all inter-container communication is allowed on the default network bridge.
References:
1. https://docs.docker.com/engine/userguide/networking/
2. https://docs.docker.com/engine/userguide/networking/default_network/containe
r-communication/#communication-between-containers
39 | P a g e
2.2 Ensure the logging level is set to 'info' (Scored)
Description:
Set Docker daemon log level to info.
Rationale:
Setting up an appropriate log level, configures the Docker daemon to log events that you
would want to review later. A base log level of info and above would capture all logs
except debug logs. Until and unless required, you should not run Docker daemon at debug
log level.
Audit:
ps -ef | grep docker
Ensure that either the --log-level parameter is not present or if present, then it is set to
info.
Remediation:
Run the Docker daemon as below:
dockerd --log-level="info"
Impact:
None.
Default Value:
By default, Docker daemon is set to log level of info.
References:
1. https://docs.docker.com/edge/engine/reference/commandline/dockerd/
CIS Controls:
6.2 Ensure Audit Log Settings Support Appropriate Log Entry Formatting
Validate audit log settings for each hardware device and the software installed on it,
40 | P a g e
ensuring that logs include a date, timestamp, source addresses, destination addresses, and
various other useful elements of each packet and/or transaction. Systems should record
logs in a standardized format such as syslog entries or those outlined by the Common Event
Expression initiative. If systems cannot generate logs in a standardized format, log
normalization tools can be deployed to convert logs into such a format.
41 | P a g e
2.3 Ensure Docker is allowed to make changes to iptables (Scored)
Description:
Iptables are used to set up, maintain, and inspect the tables of IP packet filter rules in the
Linux kernel. Allow the Docker daemon to make changes to the iptables.
Rationale:
Docker will never make changes to your system iptables rules if you choose to do so.
Docker server would automatically make the needed changes to iptables based on how you
choose your networking options for the containers if it is allowed to do so. It is
recommended to let Docker server make changes to iptablesautomatically to avoid
networking misconfiguration that might hamper the communication between containers
and to the outside world. Additionally, it would save you hassles of updating
iptablesevery time you choose to run the containers or modify networking options.
Audit:
ps -ef | grep dockerd
Ensure that the --iptables parameter is either not present or not set to false.
Remediation:
Do not run the Docker daemon with --iptables=false parameter. For example, do not
start the Docker daemon as below:
dockerd --iptables=false
Impact:
Docker daemon service start requires iptables rules to be enabled before it starts. Any
restarts of iptables during docker daemon operation may result in losing docker-created
rules. Adding iptables-persistent to your iptables install can assist with mitigating this
impact.
Default Value:
By default, iptables is set to true.
42 | P a g e
References:
1. https://docs.docker.com/engine/userguide/networking/default_network/containe
r-communication/
2. https://fralef.me/docker-and-iptables.html
CIS Controls:
5 Controlled Use of Administration Privileges

Controlled Use of Administration Privileges
43 | P a g e
2.4 Ensure insecure registries are not used (Scored)
Description:
Docker considers a private registry either secure or insecure. By default, registries are
considered secure.
Rationale:
A secure registry uses TLS. A copy of registry's CA certificate is placed on the Docker host at
/etc/docker/certs.d/<registry-name>/ directory. An insecure registry is the one not
having either valid registry certificate or is not using TLS. You should not be using any
insecure registries in the production environment. Insecure registries can be tampered
with leading to possible compromise to your production system.
Additionally, If a registry is marked as insecure then docker pull, docker push, and
docker search commands will not result in an error message and the user might be
indefinitely working with insecure registries without ever being notified of potential
danger.
Audit:
Run docker info or execute the below command to find out if any insecure registries are
used:
Ensure that the --insecure-registry parameter is not present.
Remediation:
Do not use any insecure registries.
For example, do not start the Docker daemon as below:
dockerd --insecure-registry 10.1.0.0/16
Impact:
None.
44 | P a g e
Default Value:
By default, Docker assumes all, but local, registries are secure.
References:
1. https://docs.docker.com/registry/insecure/
CIS Controls:
14.2 Encrypt All Sensitive Information Over Less-trusted Networks

All communication of sensitive information over less-trusted networks should be
encrypted. Whenever information flows over a network with a lower trust level, the
information should be encrypted.
45 | P a g e
2.5 Ensure aufs storage driver is not used (Scored)
Description:
Do not use aufs as storage driver for your Docker instance.
Rationale:
The aufs storage driver is the oldest storage driver. It is based on a Linux kernel patch-set
that is unlikely to be merged into the main Linux kernel. aufs driver is also known to cause
some serious kernel crashes. aufs just has legacy support from Docker. Most importantly,
aufs is not a supported driver in many Linux distributions using latest Linux kernels.
Audit:
Execute the below command and verify that aufs is not used as storage driver:
docker info | grep -e "^Storage Driver:\s*aufs\s*$"
The above command should not return anything.
Remediation:
Do not explicitly use aufs as storage driver.
For example, do not start Docker daemon as below:
dockerd --storage-driver aufs
Impact:
aufs is the only storage driver that allows containers to share executable and shared
library memory. It might be useful if you are running thousands of containers with the
same program or libraries.
Default Value:
By default, Docker uses devicemapper as the storage driver on most of the platforms.
Default storage driver can vary based on your OS vendor. You should use the storage driver
that is best supported by your preferred vendor.
46 | P a g e
References:
1. https://docs.docker.com/engine/userguide/storagedriver/selectadriver/#support
ed-backing-filesystems
2. http://muehe.org/posts/switching-docker-from-aufs-to-devicemapper/
3. http://jpetazzo.github.io/assets/2015-03-05-deep-dive-into-docker-storage-
drivers.html#1
4. https://docs.docker.com/engine/userguide/storagedriver/
CIS Controls:
18 Application Software Security

Application Software Security
47 | P a g e
2.6 Ensure TLS authentication for Docker daemon is configured (Scored)
Description:
It is possible to make the Docker daemon to listen on a specific IP and port and any other
Unix socket other than default Unix socket. Configure TLS authentication to restrict access
to Docker daemon via IP and port.
Rationale:
By default, Docker daemon binds to a non-networked Unix socket and runs with root
privileges. If you change the default docker daemon binding to a TCP port or any other Unix
socket, anyone with access to that port or socket can have full access to Docker daemon
and in turn to the host system. Hence, you should not bind the Docker daemon to another
IP/port or a Unix socket.
If you must expose the Docker daemon via a network socket, configure TLS authentication
for the daemon and Docker Swarm APIs (if using). This would restrict the connections to
your Docker daemon over the network to a limited number of clients who could
successfully authenticate over TLS.
Audit:
Ensure that the below parameters are present:
--tlsverify
--tlscacert
--tlscert
--tlskey
Remediation:
Follow the steps mentioned in the Docker documentation or other references.
Impact:
You would need to manage and guard certificates and keys for Docker daemon and Docker
clients.
48 | P a g e
Default Value:
By default, TLS authentication is not configured.
References:
1. https://docs.docker.com/engine/security/https/
CIS Controls:
9.1 Limit Open Ports, Protocols, and Services

Ensure that only ports, protocols, and services with validated business needs are running
on each system.
49 | P a g e
2.7 Ensure the default ulimit is configured appropriately (Not Scored)
Description:
Set the default ulimit options as appropriate in your environment.
Rationale:
ulimit provides control over the resources available to the shell and to processes started
by it. Setting system resource limits judiciously saves you from many disasters such as a
fork bomb. Sometimes, even friendly users and legitimate processes can overuse system
resources and in-turn can make the system unusable.
Setting default ulimit for the Docker daemon would enforce the ulimit for all container
instances. You would not need to setup ulimit for each container instance. However, the
default ulimit can be overridden during container runtime, if needed. Hence, to control the
system resources, define a default ulimit as needed in your environment.
Audit:
Ensure that the --default-ulimit parameter is set as appropriate.
Remediation:
Run the docker in daemon mode and pass --default-ulimit as argument with respective
ulimits as appropriate in your environment.
For Example,
dockerd --default-ulimit nproc=1024:2048 --default-ulimit nofile=100:200
Impact:
If the ulimits are not set properly, the desired resource control might not be achieved and
might even make the system unusable.
Default Value:
By default, no ulimit is set.
50 | P a g e
References:
1. https://docs.docker.com/edge/engine/reference/commandline/dockerd/#default-
ulimits
CIS Controls:

51 | P a g e
2.8 Enable user namespace support (Scored)
Description:
Enable user namespace support in Docker daemon to utilize container user to host user re-
mapping. This recommendation is beneficial where containers you are using do not have an
explicit container user defined in the container image. If container images that you are
using have a pre-defined non-root user, this recommendation may be skipped since this
feature is still in its infancy and might give you unpredictable issues and complexities.
Rationale:
The Linux kernel user namespace support in Docker daemon provides additional security
for the Docker host system. It allows a container to have a unique range of user and group
IDs which are outside the traditional user and group range utilized by the host system.
For example, the root user will have expected administrative privilege inside the container
but can effectively be mapped to an unprivileged UID on the host system.
Audit:
ps -p $(docker inspect --format='{{ .State.Pid }}' <CONTAINER ID>) -o

pid,user
The above command would find the PID of the container and then would list the host user
associated with the container process. If the container process is running as root, then this
recommendation is non-compliant.
Alternatively, you can run docker info to ensure that the userns is listed under Security
Options:
docker info --format '{{ .SecurityOptions }}'
Remediation:
Please consult Docker documentation for various ways in which this can be configured
depending upon your requirements. Your steps might also vary based on platform - For
example, on Red Hat, sub-UIDs and sub-GIDs mapping creation does not work
automatically. You might have to create your own mapping.
However, the high-level steps are as below:
52 | P a g e
Step 1: Ensure that the files /etc/subuid and /etc/subgid exist.
touch /etc/subuid /etc/subgid
Step 2: Start the docker daemon with --userns-remap flag
dockerd --userns-remap=default
Impact:
User namespace remapping makes quite a few Docker features incompatible and also
currently breaks a few functionalities. Check out the Docker documentation and referenced
links for details.
Default Value:
By default, user namespace is not remapped.
References:
1. http://man7.org/linux/man-pages/man7/user_namespaces.7.html
2. https://docs.docker.com/engine/reference/commandline/dockerd/#daemon-user-
namespace-options
3. http://events.linuxfoundation.org/sites/events/files/slides/User%20Namespaces
%20-%20ContainerCon%202015%20-%2016-9-final_0.pdf
CIS Controls:

53 | P a g e
2.9 Ensure the default cgroup usage has been confirmed (Scored)
Description:
The --cgroup-parent option allows you to set the default cgroup parent to use for all the
containers. If there is no specific use case, this setting should be left at its default.
Rationale:
System administrators typically define cgroups under which containers are supposed to
run. Even if cgroups are not explicitly defined by the system administrators, containers run
under docker cgroup by default.
It is possible to attach to a different cgroup other than that is the default. This usage should
be monitored and confirmed. By attaching to a different cgroup than the one that is a
default, it is possible to share resources unevenly and thus might starve the host for
resources.
Audit:
Ensure that the --cgroup-parent parameter is either not set or is set as appropriate non-
default cgroup.
Remediation:
The default setting is good enough and can be left as-is. If you want to specifically set a non-
default cgroup, pass --cgroup-parent parameter to the docker daemon when starting it.
For Example,
dockerd --cgroup-parent=/foobar
Impact:
None.
54 | P a g e
Default Value:
By default, docker daemon uses /docker for fs cgroup driver and system.slice for
systemd cgroup driver.
References:
1. https://docs.docker.com/engine/reference/commandline/dockerd/#default-
cgroup-parent
CIS Controls:

55 | P a g e
2.10 Ensure base device size is not changed until needed (Scored)
Description:
In certain circumstances, you might need containers bigger than 10G in size. In these cases,
carefully choose the base device size.
Rationale:
The base device size can be increased at daemon restart. Increasing the base device size
allows all future images and containers to be of the new base device size. A user can use
this option to expand the base device size however shrinking is not permitted. This value
affects the system-wide “base” empty filesystem that may already be initialized and
inherited by pulled images.
Though the file system does not allot the increased size if it is empty, it will use more space
for the empty case depending upon the device size. This may cause a denial of service by
ending up in file system being over-allocated or full.
Audit:
Execute the above command and it should not show any --storage-opt dm.basesize
parameters.
Remediation:
Do not set --storage-opt dm.basesize until needed.
Impact:
None.
Default Value:
The default base device size is 10G.
56 | P a g e
References:
1. https://docs.docker.com/engine/reference/commandline/dockerd/#storage-
driver-options
CIS Controls:

57 | P a g e
2.11 Ensure that authorization for Docker client commands is enabled
(Scored)
Description:
Use native Docker authorization plugins or a third party authorization mechanism with
Docker daemon to manage access to Docker client commands.
Rationale:
Docker’s out-of-the-box authorization model is all or nothing. Any user with permission to
access the Docker daemon can run any Docker client command. The same is true for callers
using Docker’s remote API to contact the daemon. If you require greater access control, you
can create authorization plugins and add them to your Docker daemon configuration. Using
an authorization plugin, a Docker administrator can configure granular access policies for
managing access to Docker daemon.
Third party integrations of Docker may implement their own authorization models to
require authorization with the Docker daemon outside of docker's native authorization
plugin (i.e. Kubernetes, Cloud Foundry, Openshift).
Audit:
Ensure that the --authorization-plugin parameter is set as appropriate if using docker

native authorization.
docker search hello-world
Ensure that docker daemon requires authorization to perform the above command.
Remediation:
Step 1: Install/Create an authorization plugin.
Step 2: Configure the authorization policy as desired.
Step 3: Start the docker daemon as below:
dockerd --authorization-plugin=<PLUGIN_ID>
58 | P a g e
Impact:
Each docker command specifically passes through authorization plugin mechanism. This
might introduce a slight performance drop.
Third party use of alternative container engines that utilize the docker daemon may
provide alternative mechanisms to provide this security control.
Default Value:
By default, authorization plugins are not set up.
References:
1. https://docs.docker.com/engine/reference/commandline/dockerd/#access-
authorization
2. https://docs.docker.com/engine/extend/plugins_authorization/
3. https://github.com/twistlock/authz
Notes:
As a scored control, focus should be on a PASS/FAIL if the authentication occurs when a

docker client command is executed against docker daemon to enforce authentication. The
native docker authentication plugin is just one method to enforce this control.
CIS Controls:
16 Account Monitoring and Control

Account Monitoring and Control
59 | P a g e
2.12 Ensure centralized and remote logging is configured (Scored)
Description:
Docker now supports various log drivers. A preferable way to store logs is the one that
supports centralized and remote logging.
Rationale:
Centralized and remote logging ensures that all important log records are safe despite
catastrophic events. Docker now supports various such logging drivers. Use the one that
suits your environment the best.
Audit:
Run docker info and ensure that the Logging Driverproperty set as appropriate.
docker info --format '{{ .LoggingDriver }}'
Alternatively, the below command would give you the --log-driver setting, if configured.
Ensure that it is set as appropriate.
Remediation:
Step 1: Setup the desired log driver by following its documentation.
Step 2: Start the docker daemon with that logging driver.
For example,
dockerd --log-driver=syslog --log-opt syslog-address=tcp://192.xxx.xxx.xxx
Impact:
None.
Default Value:
By default, container logs are maintained as json files
60 | P a g e
References:
1. https://docs.docker.com/engine/admin/logging/overview/
CIS Controls:
6.6 Deploy A SIEM OR Log Analysis Tools For Aggregation And Correlation/Analysis
Deploy a SIEM (Security Information and Event Management) or log analytic tools for log
aggregation and consolidation from multiple machines and for log correlation and analysis.
Using the SIEM tool, system administrators and security personnel should devise profiles of
common events from given systems so that they can tune detection to focus on unusual
activity, avoid false positives, more rapidly identify anomalies, and prevent overwhelming
analysts with insignificant alerts.
61 | P a g e
2.13 Ensure operations on legacy registry (v1) are Disabled (Scored)
Description:
The latest Docker registry is v2. All operations on the legacy registry version (v1) should be
restricted.
Rationale:
Docker registry v2 brings in many performance and security improvements over v1. It
supports container image provenance and other security features such as image signing
and verification. Hence, operations on Docker legacy registry should be restricted.
Audit:
The above command should list --disable-legacy-registry as an option passed to the

docker daemon.
Remediation:
Start the docker daemon as below:
dockerd --disable-legacy-registry
Impact:
Legacy registry operations would be restricted.
Default Value:
By default, legacy registry operations are allowed.
References:
1. https://docs.docker.com/edge/engine/reference/commandline/dockerd/#legacy-
registries
2. https://docs.docker.com/registry/spec/api/
3. https://the.binbashtheory.com/creating-private-docker-registry-2-0-with-token-
authentication-service/
62 | P a g e
4. https://blog.docker.com/2015/07/new-tool-v1-registry-docker-trusted-registry-
v2-open-source/
5. http://www.slideshare.net/Docker/docker-registry-v2
CIS Controls:

63 | P a g e
2.14 Ensure live restore is Enabled (Scored)
Description:
The --live-restore enables full support of daemon-less containers in docker. It ensures

that docker does not stop containers on shutdown or restore and properly reconnects to
the container when restarted.
Rationale:
One of the important security triads is availability. Setting --live-restore flag in the
docker daemon ensures that container execution is not interrupted when the docker
daemon is not available. This also means that it is now easier to update and patch the
docker daemon without execution downtime.
Audit:
Run docker info and ensure that the Live Restore Enabled property is set to true.
docker info --format '{{ .LiveRestoreEnabled }}'
Alternatively run the below command and ensure that --live-restore is used.
Remediation:
Run the docker in daemon mode and pass --live-restore as an argument.
For Example,
dockerd --live-restore
Impact:
None.
Default Value:
By default, --live-restore is not enabled.
64 | P a g e
References:
1. https://docs.docker.com/engine/admin/live-restore/
CIS Controls:

65 | P a g e
2.15 Ensure Userland Proxy is Disabled (Scored)
Description:
The docker daemon starts a userland proxy service for port forwarding whenever a port is
exposed. Where hairpin NAT is available, this service is generally superfluous to
requirements and can be disabled.
Rationale:
Docker engine provides two mechanisms for forwarding ports from the host to containers,
hairpin NAT, and a userland proxy. In most circumstances, the hairpin NAT mode is
preferred as it improves performance and makes use of native Linux iptables functionality
instead of an additional component.
Where hairpin NAT is available, the userland proxy should be disabled on startup to reduce
the attack surface of the installation.
Audit:
Ensure that the --userland-proxy parameter is set to false.
Remediation:
dockerd --userland-proxy=false
Impact:
Some systems with older Linux kernels may not be able to support hairpin NAT and
therefore require the userland proxy service. Also, some networking setups can be
impacted by the removal of the userland proxy.
Default Value:
By default, the userland proxy is enabled.
66 | P a g e
References:
1. http://windsock.io/the-docker-proxy/
2. https://github.com/docker/docker/issues/14856
4. https://docs.docker.com/engine/userguide/networking/default_network/binding/
CIS Controls:

on each system.
67 | P a g e
2.16 Ensure daemon-wide custom seccomp profile is applied, if needed
(Not Scored)
Description:
You can choose to apply your custom seccomp profile at the daemon-wide level if needed
and override Docker's default seccomp profile.
Rationale:
A large number of system calls are exposed to every userland process with many of them
going unused for the entire lifetime of the process. Most of the applications do not need all
the system calls and thus benefit by having a reduced set of available system calls. The
reduced set of system calls reduces the total kernel surface exposed to the application and
thus improvises application security.
You could apply your own custom seccomp profile instead of Docker's default seccomp
profile. Alternatively, if Docker's default profile is good for your environment, you can
choose to ignore this recommendation.
Audit:
Run the below command and review the seccomp profile listed in the Security Options
section. If it is default, that means, Docker's default seccomp profile is applied.
docker info --format '{{ .SecurityOptions }}'
Remediation:
By default, Docker's default seccomp profile is applied. If this is good for your environment,
no action is necessary. Alternatively, if you choose to apply your own seccomp profile, use
the --seccomp-profile flag at daemon start or put it in the daemon runtime parameters
file.
dockerd --seccomp-profile </path/to/seccomp/profile>
Impact:
A misconfigured seccomp profile could possibly interrupt your container environment.

Docker-default blocked calls have been carefully scrutinized. These address some critical
68 | P a g e
vulnerabilities/issues within container environments (for example, kernel key ring calls).
So, you should be very careful while overriding the defaults.
Default Value:
By default, Docker applies a seccomp profile.
References:
1. https://docs.docker.com/engine/security/seccomp/
CIS Controls:

69 | P a g e
2.17 Ensure experimental features are avoided in production (Scored)
Description:
Avoid experimental features in production.
Rationale:
Experimental is now a runtime docker daemon flag instead of a separate build. Passing --
experimental as a runtime flag to the docker daemon, activates experimental features.
Experimental is now considered a stable release, but with a couple of features which might
not have tested and guaranteed API stability.
Audit:
Run the below command and ensure that the Experimental property is set to false in the
Server section.
docker version --format '{{ .Server.Experimental }}'
Remediation:
Do not pass --experimental as a runtime parameter to the docker daemon.
Impact:
None
Default Value:
By default, experimental features are not activated on the docker daemon.
References:
1. https://docs.docker.com/edge/engine/reference/commandline/dockerd/#options
CIS Controls:

70 | P a g e
2.18 Ensure containers are restricted from acquiring new privileges
(Scored)
Description:
Restrict containers from acquiring additional privileges via suid or sgid bits, by default.
Rationale:
A process can set the no_new_priv bit in the kernel. It persists across fork, clone and
execve. The no_new_priv bit ensures that the process or its children processes do not gain
any additional privileges via suid or sgid bits. This way a lot of dangerous operations
become a lot less dangerous because there is no possibility of subverting privileged
binaries.
Setting this at the daemon level ensures that by default all new containers are restricted
from acquiring new privileges.
Audit:
Ensure that the --no-new-privileges parameter is present and is not set to false.
Remediation:
dockerd --no-new-privileges
Impact:
no_new_priv prevents LSMs like SELinux from transitioning to process labels that have
access not allowed to the current process.
Default Value:
By default, containers are not restricted from acquiring new privileges.
71 | P a g e
References:
1. https://github.com/moby/moby/pull/29984
CIS Controls:

72 | P a g e
3 Docker daemon configuration files
This section covers Docker related files and directory permissions and ownership. Keeping
the files and directories, that may contain sensitive parameters, secure is important for
correct and secure functioning of Docker daemon.
3.1 Ensure that docker.service file ownership is set to root:root (Scored)

Description:
Verify that the docker.service file ownership and group-ownership are correctly set to
root.
Rationale:
docker.service file contains sensitive parameters that may alter the behavior of Docker
daemon. Hence, it should be owned and group-owned by root to maintain the integrity of
the file.
Audit:
execute the below command with the correct file path to verify that the file is owned and
group-owned by root.
For example,
stat -c %U:%G /usr/lib/systemd/system/docker.service | grep -v root:root
Remediation:
73 | P a g e
execute the below command with the correct file path to set the ownership and group
ownership for the file to root.
For example,
chown root:root /usr/lib/systemd/system/docker.service
Impact:
None.
Default Value:
This file may not be present on the system. In that case, this recommendation is not
applicable. By default, if the file is present, the ownership and group-ownership for this file
is correctly set to root.
References:
1. https://docs.docker.com/engine/admin/systemd/
CIS Controls:

74 | P a g e
3.2 Ensure that docker.service file permissions are set to 644 or more
restrictive (Scored)
Description:
Verify that the docker.service file permissions are correctly set to 644 or more restrictive.
Rationale:
docker.service file contains sensitive parameters that may alter the behavior of Docker
daemon. Hence, it should not be writable by any other user other than root to maintain the
integrity of the file.
Audit:
execute the below command with the correct file path to verify that the file permissions are
set to 644 or more restrictive.
For example,
stat -c %a /usr/lib/systemd/system/docker.service
Remediation:
execute the below command with the correct file path to set the file permissions to 644.
For example,
chmod 644 /usr/lib/systemd/system/docker.service
75 | P a g e
Impact:
None.
Default Value:
applicable. By default, if the file is present, the file permissions are correctly set to 644.
References:
1. https://docs.docker.com/articles/systemd/
CIS Controls:
14.4 Protect Information With Access Control Lists

All information stored on systems shall be protected with file system, network share,
claims, application, or database specific access control lists. These controls will enforce the
principle that only authorized individuals should have access to the information based on
their need to access the information as a part of their responsibilities.
76 | P a g e
3.3 Ensure that docker.socket file ownership is set to root:root (Scored)
Description:
Verify that the docker.socket file ownership and group ownership is correctly set to root.
Rationale:
docker.socket file contains sensitive parameters that may alter the behavior of Docker
remote API. Hence, it should be owned and group-owned by root to maintain the integrity
of the file.
Audit:
execute the below command with the correct file path to verify that the file is owned and
group-owned by root.
For example,
stat -c %U:%G /usr/lib/systemd/system/docker.socket | grep -v root:root
Remediation:
execute the below command with the correct file path to set the ownership and group
ownership for the file to root.
For example,
chown root:root /usr/lib/systemd/system/docker.socket
77 | P a g e
Impact:
None.
Default Value:
applicable. By default, if the file is present, the ownership and group-ownership for this file
is correctly set to root.
References:
socket-option
2. https://github.com/docker/docker-
ce/blob/master/components/packaging/deb/systemd/docker.socket
CIS Controls:

78 | P a g e
3.4 Ensure that docker.socket file permissions are set to 644 or more
Description:
Verify that the docker.socket file permissions are correctly set to 644 or more restrictive.
Rationale:
docker.socket file contains sensitive parameters that may alter the behavior of Docker
remote API. Hence, it should be writable only by root to maintain the integrity of the file.
Audit:
execute the below command with the correct file path to verify that the file permissions are
set to 644 or more restrictive.
For example,
stat -c %a /usr/lib/systemd/system/docker.socket
Remediation:
execute the below command with the correct file path to set the file permissions to 644.
For example,
chmod 644 /usr/lib/systemd/system/docker.socket
Impact:
None.
79 | P a g e
Default Value:
applicable. By default, if the file is present, the file permissions for this file are correctly set
to 644.
References:
1. https://docs.docker.com/engine/reference/commandline/dockerd/#bind-docker-
to-another-hostport-or-a-unix-socket
2. https://github.com/YungSang/fedora-atomic-
packer/blob/master/oem/docker.socket
3. http://daviddaeschler.com/2014/12/14/centos-7rhel-7-and-docker-containers-
on-boot/
CIS Controls:

80 | P a g e
3.5 Ensure that /etc/docker directory ownership is set to root:root
(Scored)
Description:
Verify that the /etc/docker directory ownership and group-ownership is correctly set to
root.
Rationale:
/etc/docker directory contains certificates and keys in addition to various sensitive files.
Hence, it should be owned and group-owned by root to maintain the integrity of the
directory.
Audit:
Execute the below command to verify that the directory is owned and group-owned by
root:
stat -c %U:%G /etc/docker | grep -v root:root
Remediation:
chown root:root /etc/docker

This would set the ownership and group-ownership for the directory to root.
Impact:
None.
Default Value:
By default, the ownership and group-ownership for this directory is correctly set to root.
References:
81 | P a g e
CIS Controls:

82 | P a g e
3.6 Ensure that /etc/docker directory permissions are set to 755 or more
Description:
Verify that the /etc/docker directory permissions are correctly set to 755 or more
restrictive.
Rationale:
/etc/docker directory contains certificates and keys in addition to various sensitive files.
Hence, it should only be writable by root to maintain the integrity of the directory.
Audit:
Execute the below command to verify that the directory has permissions of 755 or more
restrictive:
stat -c %a /etc/docker
Remediation:
chmod 755 /etc/docker
This would set the permissions for the directory to 755.
Impact:
None.
Default Value:
By default, the permissions for this directory are correctly set to 755.
References:
83 | P a g e
CIS Controls:

84 | P a g e
3.7 Ensure that registry certificate file ownership is set to root:root
(Scored)
Description:
Verify that all the registry certificate files (usually found under
/etc/docker/certs.d/<registry-name> directory) are owned and group-owned by root.
Rationale:
/etc/docker/certs.d/<registry-name> directory contains Docker registry certificates.

These certificate files must be owned and group-owned by root to maintain the integrity of
the certificates.
Audit:
Execute the below command to verify that the registry certificate files are owned and
group-owned by root:
stat -c %U:%G /etc/docker/certs.d/* | grep -v root:root
Remediation:
chown root:root /etc/docker/certs.d/<registry-name>/*
This would set the ownership and group-ownership for the registry certificate files to root.
Impact:
None.
Default Value:
By default, the ownership and group-ownership for registry certificate files is correctly set
to root.
References:
85 | P a g e
CIS Controls:

86 | P a g e
3.8 Ensure that registry certificate file permissions are set to 444 or
more restrictive (Scored)
Description:
Verify that all the registry certificate files (usually found under
/etc/docker/certs.d/<registry-name> directory) have permissions of 444 or more
restrictive.
Rationale:
/etc/docker/certs.d/<registry-name> directory contains Docker registry certificates.

These certificate files must have permissions of 444 to maintain the integrity of the
certificates.
Audit:
Execute the below command to verify that the registry certificate files have permissions of
444 or more restrictive:
stat -c %a /etc/docker/certs.d/<registry-name>/*
Remediation:
chmod 444 /etc/docker/certs.d/<registry-name>/*
This would set the permissions for registry certificate files to 444.
Impact:
None.
Default Value:
By default, the permissions for registry certificate files might not be 444. The default file
permissions are governed by the system or user specific umaskvalues.
References:
87 | P a g e
CIS Controls:

88 | P a g e
3.9 Ensure that TLS CA certificate file ownership is set to root:root
(Scored)
Description:
Verify that the TLS CA certificate file (the file that is passed alongwith --tlscacert
parameter) is owned and group-owned by root.
Rationale:
The TLS CA certificate file should be protected from any tampering. It is used to
authenticate Docker server based on given CA certificate. Hence, it must be owned and
group-owned by root to maintain the integrity of the CA certificate.
Audit:
Execute the below command to verify that the TLS CA certificate file is owned and group-
owned by root:
stat -c %U:%G <path to TLS CA certificate file> | grep -v root:root
Remediation:
chown root:root <path to TLS CA certificate file>
This would set the ownership and group-ownership for the TLS CA certificate file to root.
Impact:
None.
Default Value:
By default, the ownership and group-ownership for TLS CA certificate file is correctly set to
root.
References:
89 | P a g e
CIS Controls:

90 | P a g e
3.10 Ensure that TLS CA certificate file permissions are set to 444 or
Description:
Verify that the TLS CA certificate file (the file that is passed alongwith --tlscacert
parameter) has permissions of 444 or more restrictive.
Rationale:
The TLS CA certificate file should be protected from any tampering. It is used to
authenticate Docker server based on given CA certificate. Hence, it must have permissions
of 444 to maintain the integrity of the CA certificate.
Audit:
Execute the below command to verify that the TLS CA certificate file has permissions of 444
or more restrictive:
stat -c %a <path to TLS CA certificate file>
Remediation:
chmod 444 <path to TLS CA certificate file>
This would set the file permissions of the TLS CA file to 444.
Impact:
None.
Default Value:
By default, the permissions for TLS CA certificate file might not be 444. The default file
permissions are governed by the system or user specific umask values.
References:
91 | P a g e
CIS Controls:

92 | P a g e
3.11 Ensure that Docker server certificate file ownership is set to
root:root (Scored)
Description:
Verify that the Docker server certificate file (the file that is passed alongwith --tlscert
Rationale:
The Docker server certificate file should be protected from any tampering. It is used to
authenticate Docker server based on the given server certificate. Hence, it must be owned
and group-owned by root to maintain the integrity of the certificate.
Audit:
Execute the below command to verify that the Docker server certificate file is owned and
group-owned by root:
stat -c %U:%G <path to Docker server certificate file> | grep -v root:root
Remediation:
chown root:root <path to Docker server certificate file>
This would set the ownership and group-ownership for the Docker server certificate file to
root.
Impact:
None.
Default Value:
By default, the ownership and group-ownership for Docker server certificate file is
correctly set to root.
93 | P a g e
References:
CIS Controls:

94 | P a g e
3.12 Ensure that Docker server certificate file permissions are set to 444
or more restrictive (Scored)
Description:
Verify that the Docker server certificate file (the file that is passed alongwith --tlscert
parameter) has permissions of 444 or more restrictive.
Rationale:
The Docker server certificate file should be protected from any tampering. It is used to
authenticate Docker server based on the given server certificate. Hence, it must have
permissions of 444 to maintain the integrity of the certificate.
Audit:
Execute the below command to verify that the Docker server certificate file has
permissions of 444 or more restrictive:
stat -c %a <path to Docker server certificate file>
Remediation:
chmod 444 <path to Docker server certificate file>
This would set the file permissions of the Docker server file to 444.
Impact:
None.
Default Value:
By default, the permissions for Docker server certificate file might not be 444. The default
file permissions are governed by the system or user specific umask values.
References:
95 | P a g e
CIS Controls:

96 | P a g e
3.13 Ensure that Docker server certificate key file ownership is set to
root:root (Scored)
Description:
Verify that the Docker server certificate key file (the file that is passed alongwith --tlskey
Rationale:
The Docker server certificate key file should be protected from any tampering or unneeded
reads. It holds the private key for the Docker server certificate. Hence, it must be owned
and group-owned by root to maintain the integrity of the Docker server certificate.
Audit:
Execute the below command to verify that the Docker server certificate key file is owned
and group-owned by root:
stat -c %U:%G <path to Docker server certificate key file> | grep -v

root:root
Remediation:
chown root:root <path to Docker server certificate key file>
This would set the ownership and group-ownership for the Docker server certificate key
file to root.
Impact:
None.
Default Value:
By default, the ownership and group-ownership for Docker server certificate key file is
correctly set to root.
97 | P a g e
References:
CIS Controls:

98 | P a g e
3.14 Ensure that Docker server certificate key file permissions are set to
400 (Scored)
Description:
Verify that the Docker server certificate key file (the file that is passed alongwith --tlskey
parameter) has permissions of 400.
Rationale:
The Docker server certificate key file should be protected from any tampering or unneeded
reads. It holds the private key for the Docker server certificate. Hence, it must have
permissions of 400 to maintain the integrity of the Docker server certificate.
Audit:
Execute the below command to verify that the Docker server certificate key file has
permissions of 400:
stat -c %a <path to Docker server certificate key file>
Remediation:
chmod 400 <path to Docker server certificate key file>
This would set the Docker server certificate key file permissions to 400.
Impact:
None.
Default Value:
By default, the permissions for Docker server certificate key file might not be 400. The
default file permissions are governed by the system or user specific umask values.
References:
99 | P a g e
CIS Controls:

100 | P a g e
3.15 Ensure that Docker socket file ownership is set to root:docker
(Scored)
Description:
Verify that the Docker socket file is owned by root and group-owned by docker.
Rationale:
Docker daemon runs as root. The default Unix socket hence must be owned by root. If any
other user or process owns this socket, then it might be possible for that non-privileged
user or process to interact with Docker daemon. Also, such a non-privileged user or
process might interact with containers. This is neither secure nor desired behavior.
Additionally, the Docker installer creates a Unix group called docker. You can add users to
this group, and then those users would be able to read and write to default Docker Unix
socket. The membership to the docker group is tightly controlled by the system
administrator. If any other group owns this socket, then it might be possible for members
of that group to interact with Docker daemon. Also, such a group might not be as tightly
controlled as the docker group. This is neither secure nor desired behavior.
Hence, the default Docker Unix socket file must be owned by root and group-owned by
docker to maintain the integrity of the socket file.
Audit:
Execute the below command to verify that the Docker socket file is owned by root and
group-owned by docker:
stat -c %U:%G /var/run/docker.sock | grep -v root:docker
Remediation:
chown root:docker /var/run/docker.sock
This would set the ownership to root and group-ownership to docker for default Docker
socket file.
101 | P a g e
Impact:
None.
Default Value:
By default, the ownership and group-ownership for Docker socket file is correctly set to
root:docker.
References:
socket-option
CIS Controls:

102 | P a g e
3.16 Ensure that Docker socket file permissions are set to 660 or more
Description:
Verify that the Docker socket file has permissions of 660 or more restrictive.
Rationale:
Only root and members of docker group should be allowed to read and write to default
Docker Unix socket. Hence, the Docket socket file must have permissions of 660 or more
restrictive.
Audit:
Execute the below command to verify that the Docker socket file has permissions of 660 or
more restrictive:
stat -c %a /var/run/docker.sock
Remediation:
chmod 660 /var/run/docker.sock
This would set the file permissions of the Docker socket file to 660.
Impact:
None.
Default Value:
By default, the permissions for Docker socket file is correctly set to 660.
References:
socket-option
103 | P a g e
CIS Controls:

104 | P a g e
3.17 Ensure that daemon.json file ownership is set to root:root (Scored)
Description:
Verify that the daemon.json file ownership and group-ownership is correctly set to root.
Rationale:
daemon.json file contains sensitive parameters that may alter the behavior of docker
daemon. Hence, it should be owned and group-owned by root to maintain the integrity of
the file.
Audit:
Execute the below command to verify that the file is owned and group-owned by root:
stat -c %U:%G /etc/docker/daemon.json | grep -v root:root
Remediation:
chown root:root /etc/docker/daemon.json
This would set the ownership and group-ownership for the file to root.
Impact:
None.
Default Value:
applicable.
References:
configuration-file
105 | P a g e
CIS Controls:

106 | P a g e
3.18 Ensure that daemon.json file permissions are set to 644 or more
Description:
Verify that the daemon.json file permissions are correctly set to 644 or more restrictive.
Rationale:
daemon.json file contains sensitive parameters that may alter the behavior of docker
daemon. Hence, it should be writable only by root to maintain the integrity of the file.
Audit:
Execute the below command to verify that the file permissions are correctly set to 644 or
more restrictive:
stat -c %a /etc/docker/daemon.json
Remediation:
chmod 644 /etc/docker/daemon.json
This would set the file permissions for this file to 644.
Impact:
None.
Default Value:
applicable.
References:
configuration-file
107 | P a g e
CIS Controls:

108 | P a g e
3.19 Ensure that /etc/default/docker file ownership is set to root:root
(Scored)
Description:
Verify that the /etc/default/docker file ownership and group-ownership is correctly set
to root.
Rationale:
/etc/default/docker file contains sensitive parameters that may alter the behavior of
docker daemon. Hence, it should be owned and group-owned by root to maintain the
integrity of the file.
Audit:
Execute the below command to verify that the file is owned and group-owned by root:
stat -c %U:%G /etc/default/docker | grep -v root:root
Remediation:
chown root:root /etc/default/docker
This would set the ownership and group-ownership for the file to root.
Impact:
None.
Default Value:
applicable.
References:
1. https://docs.docker.com/engine/admin/configuring/
109 | P a g e
CIS Controls:

110 | P a g e
3.20 Ensure that /etc/default/docker file permissions are set to 644 or
Description:
Verify that the /etc/default/docker file permissions are correctly set to 644 or more
restrictive.
Rationale:
/etc/default/docker file contains sensitive parameters that may alter the behavior of
docker daemon. Hence, it should be writable only by root to maintain the integrity of the
file.
Audit:
Execute the below command to verify that the file permissions are correctly set to 644 or
more restrictive:
stat -c %a /etc/default/docker
Remediation:
chmod 644 /etc/default/docker
This would set the file permissions for this file to 644.
Impact:
None.
Default Value:
applicable.
References:
1. https://docs.docker.com/engine/admin/configuring/
111 | P a g e
CIS Controls:

112 | P a g e
4 Container Images and Build File
Container base images and build files govern the fundamentals of how a container instance
from a particular image would behave. Ensuring that you are using proper base images and
appropriate build files can be very important for building your containerized
infrastructure. Below are some of the recommendations that you should follow for
container base images and build files to ensure that your containerized infrastructure is
secure.
4.1 Ensure a user for the container has been created (Scored)
Description:
Create a non-root user for the container in the Dockerfile for the container image.
Rationale:
It is a good practice to run the container as a non-root user, if possible. Though user
namespace mapping is now available, if a user is already defined in the container image, the
container is run as that user by default and specific user namespace remapping is not
required.
Audit:
docker ps --quiet --all | xargs docker inspect --format '{{ .Id }}: User={{
.Config.User }}'
The above command should return container username or user ID. If it is blank it means,
the container is running as root.
Remediation:
Ensure that the Dockerfile for the container image contains below instruction:
USER <username or ID>
where username or ID refers to the user that could be found in the container base image. If
there is no specific user created in the container base image, then add a useradd command
to add the specific user before USER instruction.
113 | P a g e
For example, add the below lines in the Dockerfile to create a user in the container:
RUN useradd -d /home/username -m -s /bin/bash username

USER username
Note: If there are users in the image that the containers do not need, consider deleting
them. After deleting those users, commit the image and then generate new instances of
containers for use.
Impact:
None.
Default Value:
By default, the containers are run with rootprivileges and as user rootinside the container.
References:
CIS Controls:

114 | P a g e
4.2 Ensure that containers use trusted base images (Not Scored)
Description:
Ensure that the container image is written either from scratch or is based on another
established and trusted base image downloaded over a secure channel.
Rationale:
Official repositories are Docker images curated and optimized by the Docker community or
the vendor. There could be other potentially unsafe public repositories. Caution should be
exercised when obtaining container images from Docker and third parties to how they will
be used for your organization's data.
Audit:
Step 1 - Inspect the Docker host for Docker images used by executing the below command:
docker images
This would list all the container images that are currently available for use on the Docker
host. Interview the system administrator and obtain a proof of evidence that the list of
images was obtained from trusted source over a secure channel or from a trusted, secure
private Docker registry.
Step 2 - For each Docker image found on the Docker host, inspect the image for how it was
built to verify if from trusted sources and hardened configuration:
docker history <imageName>
Remediation:
 Configure and use Docker Content trust.

 Inspect Docker image history to evaluate their risk to operate on your network.
 Scan Docker images for vulnerabilities in their dependencies and configurations
they will impose upon your network.
Impact:
None.
115 | P a g e
Default Value:
Not Applicable.
References:
1. https://titanous.com/posts/docker-insecurity
2. https://registry.hub.docker.com/
3. http://blog.docker.com/2014/10/docker-1-3-signed-images-process-injection-
security-options-mac-shared-directories/
5. https://docs.docker.com/engine/reference/commandline/pull/
7. https://blog.docker.com/2015/11/docker-trusted-registry-1-4/
CIS Controls:
3 Secure Configurations for Hardware and Software on Mobile Devices, Laptops,

Secure Configurations for Hardware and Software on Mobile Devices, Laptops,
116 | P a g e
4.3 Ensure unnecessary packages are not installed in the container (Not
Scored)
Description:
Containers tend to be minimal and slim down versions of the Operating System. Do not
install anything that does not justify the purpose of container.
Rationale:
Bloating containers with unnecessary software could possibly increase the attack surface
of the container. This also voids the concept of minimal and slim down versions of
container images. Hence, do not install anything else apart from what is truly needed for
the purpose of the container.
Audit:
Step 1: List all the running instances of containers by executing below command:
docker ps --quiet
Step 2: For each container instance, execute the below or equivalent command:
docker exec $INSTANCE_ID rpm -qa
The above command would list the packages installed on the container. Review the list and
ensure that it is legitimate.
Remediation:
At the outset, do not install anything on the container that does not justify the purpose. If
the image had some packages that your container does not use, uninstall them.
Consider using a minimal base image rather than the standard Redhat/Centos/Debian
images if you can. Some of the options include BusyBox and Alpine.
Not only does this trim your image size from >150Mb to ~20 Mb, there are also fewer tools
and paths to escalate privileges. You can even remove the package installer as a final
hardening measure for leaf/production containers.
117 | P a g e
Impact:
None.
Default Value:
Not Applicable.
References:
1. https://docs.docker.com/userguide/dockerimages/
2. http://www.livewyer.com/blog/2015/02/24/slimming-down-your-docker-
containers-alpine-linux
3. https://github.com/progrium/busybox
CIS Controls:

118 | P a g e
4.4 Ensure images are scanned and rebuilt to include security patches
(Not Scored)
Description:
Images should be scanned "frequently" for any vulnerabilities. Rebuild the images to
include patches and then instantiate new containers from it.
Rationale:
Vulnerabilities are loopholes/bugs that can be exploited and security patches are updates
to resolve these vulnerabilities. We can use image vulnerability scanning tools to find any
kind of vulnerabilities within the images and then check for available patches to mitigate
these vulnerabilities. Patches update the system to the most recent code base. Being on the
current code base is important because that's where vendors focus on fixing problems.
Evaluate the security patches before applying and follow the patching best practices.
Also, it would be better if, image vulnerability scanning tools could perform binary level
analysis or hash based verification instead of just version string matching.
Audit:
docker ps --quiet
Step 2: For each container instance, execute the below or equivalent command to find the
list of packages installed within the container. Ensure that the security updates for various
affected packages are installed.
docker exec $INSTANCE_ID rpm -qa
Alternatively, you could run image vulnerability scanning tools which can scan all the
images in your ecosystem and then apply patches for the detected vulnerabilities based on
your patch management procedures.
Remediation:
Follow the below steps to rebuild the images with security patches:
119 | P a g e
Step 1: Pull all the base images (i.e., given your set of Dockerfiles, extract all images
declared in FROM instructions, and re-pull them to check for an updated/patched versions).
Patch the packages within the images too.
docker pull
Step 2: Force a rebuild of each image:
docker build --no-cache
Step 3: Restart all containers with the updated images.
You could also use ONBUILD directive in the Dockerfile to trigger particular update
instructions for images that you know are used as base images frequently.
Impact:
None
Default Value:
By default, containers and images are not updated of their own.
References:
1. https://docs.docker.com/userguide/dockerimages/
2. https://docs.docker.com/docker-cloud/builds/image-scan/
3. https://blog.docker.com/2016/05/docker-security-scanning/
4. https://docs.docker.com/engine/reference/builder/#/onbuild
CIS Controls:
18.1 Use Only Vendor-supported Software

For all acquired application software, check that the version you are using is still supported
by the vendor. If not, update to the most current version and install all relevant patches and
vendor security recommendations.
120 | P a g e
4.5 Ensure Content trust for Docker is Enabled (Scored)
Description:
Content trust is disabled by default. You should enable it.
Rationale:
Content trust provides the ability to use digital signatures for data sent to and received
from remote Docker registries. These signatures allow client-side verification of the
integrity and publisher of specific image tags. This ensures provenance of container images.
Audit:
echo $DOCKER_CONTENT_TRUST
This should return 1.
Remediation:
To enable content trust in a bash shell, enter the following command:
export DOCKER_CONTENT_TRUST=1
Alternatively, set this environment variable in your profile file so that content trust in
enabled on every login.
Impact:
In an environment where DOCKER_CONTENT_TRUST is set, you are required to follow trust

procedures while working with images - build, create, pull, pushand run. You can use the
--disable-content-trust flag to run individual operations on tagged images without
content trust on an as-needed basis but that defeats the purpose of enabling content trust
and hence, should be avoided wherever possible.
Note: Content trust is currently only available for users of the public Docker Hub. It is
currently not available for the Docker Trusted Registry or for private registries.
Default Value:
By default, content trust is disabled.
121 | P a g e
References:
1. https://docs.docker.com/engine/security/trust/content_trust/
2. https://docs.docker.com/engine/reference/commandline/cli/#notary
3. https://docs.docker.com/engine/reference/commandline/cli/#environment-
variables
CIS Controls:

122 | P a g e
4.6 Ensure HEALTHCHECK instructions have been added to the container
image (Scored)
Description:
Add HEALTHCHECK instruction in your docker container images to perform the health check
on running containers.
Rationale:
One of the important security triads is availability. Adding HEALTHCHECK instruction to your
container image ensures that the docker engine periodically checks the running container
instances against that instruction to ensure that the instances are still working.
Based on the reported health status, the docker engine could then exit non-working
containers and instantiate new ones.
Audit:
Run the below command and ensure that the docker image has appropriate HEALTHCHECK
instruction set up.
docker inspect --format='{{ .Config.Healthcheck }}' <IMAGE>
Remediation:
Follow Docker documentation and rebuild your container image with HEALTHCHECK
instruction.
Impact:
None.
Default Value:
By default, HEALTHCHECK is not set.
References:
1. https://docs.docker.com/engine/reference/builder/#healthcheck
123 | P a g e
CIS Controls:

124 | P a g e
4.7 Ensure update instructions are not use alone in the Dockerfile (Not
Scored)
Description:
Do not use update instructions such as apt-get update alone or in a single line in the
Dockerfile.
Rationale:
Adding the update instructions in a single line on the Dockerfile will cache the update layer.
Thus, when you build any image later using the same instruction, previously cached update
layer will be used. This could potentially deny any fresh updates to go in the later builds.
Audit:
Step 1: Run the below command to get the list of images:
docker images
Step 2: Run the below command for each image in the list above, and look for any update
instructions being in a single line:
docker history <Image_ID>
Alternatively, if you have access to Dockerfile for the image, verify that there are no update
instructions as described above.
Remediation:
Use update instructions along with install instructions (or any other) and version pinning
for packages while installing them. This would bust the cache and force to extract the
required versions.
Alternatively, you could use --no-cache flag during docker build process to avoid using
cached layers.
Impact:
None
125 | P a g e
Default Value:
By default, docker does not enforce any restrictions on using update instructions.
References:
1. https://docs.docker.com/engine/userguide/eng-image/dockerfile_best-
practices/#run
CIS Controls:

126 | P a g e
4.8 Ensure setuid and setgid permissions are removed in the images
(Not Scored)
Description:
Removing setuid and setgid permissions in the images would prevent privilege escalation
attacks in the containers.
Rationale:
setuid and setgid permissions could be used for elevating privileges. While these
permissions are at times legitimately needed, these could potentially be used in privilege
escalation attacks. Thus, you should consider dropping these permissions for the packages
which do not need them within the images.
Audit:
Run the below command on the image to list the executables having setuid and setgid
permissions:
docker run <Image_ID> find / -perm +6000 -type f -exec ls -ld {} \; 2>
/dev/null
Carefully, review the list and ensure that it is legitimate.
Remediation:
Allow setuid and setgid permissions only on executables which need them. You could
remove these permissions during build time by adding the following command in your
Dockerfile, preferably towards the end of the Dockerfile:
RUN find / -perm +6000 -type f -exec chmod a-s {} \; || true
Impact:
Above command breaks all the executables that depend on setuid or setgid permissions
including the legitimate ones. Hence, be careful to modify the command to suit your
requirements so that it does not drop the permissions of legitimate programs. This
requires a careful examination of each executable and fine tuning the permissions.
127 | P a g e
Default Value:
Not Applicable
References:
1. http://www.oreilly.com/webops-perf/free/files/docker-security.pdf
2. http://container-
solutions.com/content/uploads/2015/06/15.06.15_DockerCheatSheet_A2.pdf
3. http://man7.org/linux/man-pages/man2/setuid.2.html
4. http://man7.org/linux/man-pages/man2/setgid.2.html
CIS Controls:

128 | P a g e
4.9 Ensure COPY is used instead of ADD in Dockerfile (Not Scored)
Description:
Use COPY instruction instead of ADD instruction in the Dockerfile.
Rationale:
COPY instruction just copies the files from the local host machine to the container file
system. ADD instruction potentially could retrieve files from remote URLs and perform
operations such as unpacking. Thus, ADD instruction introduces risks such as adding
malicious files from URLs without scanning and unpacking procedure vulnerabilities.
Audit:
docker images
Step 2: Run the below command for each image in the list above and look for any ADD
instructions:
Alternatively, if you have access to Dockerfile for the image, verify that there are no ADD
instructions.
Remediation:
Use COPY instructions in Dockerfiles.
Impact:
You would need to take care of the functionalities provided by ADD instructions such as
fetching files from remote URLs.
Default Value:
Not Applicable
129 | P a g e
References:
1. https://docs.docker.com/engine/userguide/eng-image/dockerfile_best-
practices/#add-or-copy
CIS Controls:

130 | P a g e
4.10 Ensure secrets are not stored in Dockerfiles (Not Scored)
Description:
Do not store any secrets in Dockerfiles.
Rationale:
Dockerfiles could be backtracked easily by using native Docker commands such as docker
history and various tools and utilities. Also, as a general practice, image publishers
provide Dockerfiles to build the credibility for their images. Hence, the secrets within these
Dockerfiles could be easily exposed and potentially be exploited.
Audit:
docker images
Step 2: Run the below command for each image in the list above, and look for any secrets:
Alternatively, if you have access to Dockerfile for the image, verify that there are no secrets
as described above.
Remediation:
Do not store any kind of secrets within Dockerfiles.
Impact:
You would need to identify a way to handle secrets for your Docker images.
Default Value:
By default, there are no restrictions on storing config secrets in the Dockerfiles.
References:
2. http://12factor.net/config
131 | P a g e
3. https://avicoder.me/2016/07/22/Twitter-Vine-Source-code-dump/
CIS Controls:

132 | P a g e
4.11 Ensure verified packages are only Installed (Not Scored)
Description:
Verify authenticity of the packages before installing them in the image.
Rationale:
Verifying authenticity of the packages is essential for building a secure container image.
Tampered packages could potentially be malicious or have some known vulnerabilities that
could be exploited.
Audit:
docker images
Step 2: Run the below command for each image in the list above, and look for how the
authenticity of the packages is determined. This could be via the use of GPG keys or other
secure package distribution mechanisms
Alternatively, if you have access to Dockerfile for the image, verify that the authenticity of
the packages is checked.
Remediation:
Use GPG keys for downloading and verifying packages or any other secure package
distribution mechanism of your choice.
Impact:
None
Default Value:
Not Applicable
133 | P a g e
References:
2. https://github.com/docker-
library/httpd/blob/12bf8c8883340c98b3988a7bade8ef2d0d6dcf8a/2.4/Dockerfil
e
3. https://github.com/docker-
library/php/blob/d8a4ccf4d620ec866d5b42335b699742df08c5f0/7.0/alpine/Doc
kerfile
4. https://access.redhat.com/security/team/key
CIS Controls:

134 | P a g e
5 Container Runtime
The ways in which a container is started governs a lot of security implications. It is possible
to provide potentially dangerous runtime parameters that might compromise the host and
other containers on the host. Verifying container runtime is thus very important. Various
recommendations to assess the container runtime are as below:
5.1 Ensure AppArmor Profile is Enabled (Scored)

Description:
AppArmor is an effective and easy-to-use Linux application security system. It is available

on quite a few Linux distributions by default such as Debian and Ubuntu.
Rationale:
AppArmor protects the Linux OS and applications from various threats by enforcing
security policy which is also known as AppArmor profile. You can create your own
AppArmor profile for containers or use the Docker's default AppArmor profile. This would
enforce security policies on the containers as defined in the profile.
Audit:
docker ps --quiet --all | xargs docker inspect --format '{{ .Id }}:
AppArmorProfile={{ .AppArmorProfile }}'
The above command should return a valid AppArmor Profile for each container instance.
Remediation:
If AppArmor is applicable for your Linux OS, use it. You may have to follow below set of
steps:
1. Verify if AppArmor is installed. If not, install it.

2. Create or import a AppArmor profile for Docker containers.
3. Put this profile in enforcing mode.
4. Start your Docker container using the customized AppArmor profile. For example,
docker run --interactive --tty --security-opt="apparmor:PROFILENAME" centos

/bin/bash
135 | P a g e
Alternatively, you can keep the docker's default apparmor profile.
Impact:
The container (process) would have set of restrictions as defined in AppArmor profile. If
your AppArmor profile is mis-configured, then the container may not entirely work as
expected.
Default Value:
By default, docker-default AppArmor profile is applied for running containers and this
profile can be found at /etc/apparmor.d/docker.
References:
1. https://docs.docker.com/engine/security/apparmor/
2. https://docs.docker.com/engine/reference/run/#security-configuration
CIS Controls:

136 | P a g e
5.2 Ensure SELinux security options are set, if applicable (Scored)
Description:
SELinux is an effective and easy-to-use Linux application security system. It is available on

quite a few Linux distributions by default such as Red Hat and Fedora.
Rationale:
SELinux provides a Mandatory Access Control (MAC) system that greatly augments the
default Discretionary Access Control (DAC) model. You can thus add an extra layer of safety
by enabling SELinux on your Linux host, if applicable.
Audit:
SecurityOpt={{ .HostConfig.SecurityOpt }}'
The above command should return all the security options currently configured for the
containers.
Remediation:
If SELinux is applicable for your Linux OS, use it. You may have to follow below set of steps:
1. Set the SELinux State.

2. Set the SELinux Policy.
3. Create or import a SELinux policy template for Docker containers.
4. Start Docker in daemon mode with SELinux enabled. For example,
docker daemon --selinux-enabled
5. Start your Docker container using the security options. For example,
docker run --interactive --tty --security-opt label=level:TopSecret centos

/bin/bash
Impact:
The container (process) would have set of restrictions as defined in SELinux policy. If your
SELinux policy is mis-configured, then the container may not entirely work as expected.
137 | P a g e
Default Value:
By default, no SELinux security options are applied on containers.
References:
3. http://docs.fedoraproject.org/en-US/Fedora/13/html/Security-Enhanced_Linux/
us/red_hat_enterprise_linux_atomic_host/7/html/container_security_guide/docker_
selinux_security_policy
CIS Controls:

138 | P a g e
5.3 Ensure Linux Kernel Capabilities are restricted within containers
(Scored)
Description:
By default, Docker starts containers with a restricted set of Linux Kernel Capabilities. It
means that any process may be granted the required capabilities instead of root access.
Using Linux Kernel Capabilities, the processes do not have to run as root for almost all the
specific areas where root privileges are usually needed.
Rationale:
Docker supports the addition and removal of capabilities, allowing the use of a non-default
profile. This may make Docker more secure through capability removal, or less secure
through the addition of capabilities. It is thus recommended to remove all capabilities
except those explicitly required for your container process.
For example, capabilities such as below are usually not needed for container process:
NET_ADMIN
SYS_ADMIN
SYS_MODULE
Audit:
docker ps --quiet --all | xargs docker inspect --format '{{ .Id }}: CapAdd={{
.HostConfig.CapAdd }} CapDrop={{ .HostConfig.CapDrop }}'
Verify that the added and dropped Linux Kernel Capabilities are in line with the ones
needed for container process for each container instance.
Remediation:
Execute the below command to add needed capabilities:
$> docker run --cap-add={"Capability 1","Capability 2"}
For example,
docker run --interactive --tty --cap-add={"NET_ADMIN","SYS_ADMIN"}

centos:latest /bin/bash
139 | P a g e
Execute the below command to drop unneeded capabilities:
$> docker run --cap-drop={"Capability 1","Capability 2"}
For example,
docker run --interactive --tty --cap-drop={"SETUID","SETGID"} centos:latest

/bin/bash
Alternatively,
You may choose to drop all capabilities and add only add the needed ones:
$> docker run --cap-drop=all --cap-add={"Capability 1","Capability 2"}
For example,
docker run --interactive --tty --cap-drop=all --cap-

add={"NET_ADMIN","SYS_ADMIN"} centos:latest /bin/bash
Impact:
Based on what Linux Kernel Capabilities were added or dropped, restrictions within the
container would apply.
Default Value:
By default, below capabilities are available for containers:
AUDIT_WRITE
CHOWN
DAC_OVERRIDE
FOWNER
FSETID
KILL
MKNOD
NET_BIND_SERVICE
NET_RAW
SETFCAP
SETGID
SETPCAP
SETUID
SYS_CHROOT
References:
1. https://docs.docker.com/engine/security/security/#linux-kernel-capabilities
2. http://man7.org/linux/man-pages/man7/capabilities.7.html
140 | P a g e
CIS Controls:

141 | P a g e
5.4 Ensure privileged containers are not used (Scored)
Description:
Using the --privileged flag gives all Linux Kernel Capabilities to the container thus
overwriting the --cap-add and --cap-drop flags. Ensure that it is not used.
Rationale:
The --privileged flag gives all capabilities to the container, and it also lifts all the
limitations enforced by the device cgroup controller. In other words, the container can then
do almost everything that the host can do. This flag exists to allow special use-cases, like
running Docker within Docker.
Audit:
Privileged={{ .HostConfig.Privileged }}'
The above command should return Privileged=false for each container instance.
Remediation:
Do not run container with the --privileged flag.
For example, do not start a container as below:
docker run --interactive --tty --privileged centos /bin/bash
Impact:
Linux Kernel Capabilities other than defaults would not be available for use within
container.
Default Value:
False.
References:
1. https://docs.docker.com/engine/reference/run/#runtime-privilege-and-linux-
capabilities
142 | P a g e
CIS Controls:

143 | P a g e
5.5 Ensure sensitive host system directories are not mounted on
containers (Scored)
Description:
Sensitive host system directories such as below should not be allowed to be mounted as
container volumes especially in read-write mode.
/
/boot
/dev
/etc
/lib
/proc
/sys
/usr
Rationale:
If sensitive directories are mounted in read-write mode, it would be possible to make

changes to files within those sensitive directories. The changes might bring down security
implications or unwarranted changes that could put the Docker host in compromised state.
Audit:
Volumes={{ .Mounts }}'
The above commands would return the list of current mapped directories and whether
they are mounted in read-write mode for each container instance.
Remediation:
Do not mount host sensitive directories on containers especially in read-write mode.
Impact:
None.
Default Value:
Docker defaults to a read-write volume but you can also mount a directory read-only. By
default, no sensitive host directories are mounted on containers.
144 | P a g e
References:
1. https://docs.docker.com/engine/tutorials/dockervolumes/
CIS Controls:

145 | P a g e
5.6 Ensure ssh is not run within containers (Scored)
Description:
SSH server should not be running within the container. You should SSH into the Docker
host, and use nsenter tool to enter a container from a remote host.
Rationale:
Running SSH within the container increases the complexity of security management by
making it
 Difficult to manage access policies and security compliance for SSH server
 Difficult to manage keys and passwords across various containers
 Difficult to manage security upgrades for SSH server
It is possible to have shell access to a container without using SSH, the needlessly
increasing the complexity of security management should be avoided.
Audit:
docker ps --quiet
Step 2: For each container instance, execute the below command:
docker exec $INSTANCE_ID ps -el
Ensure that there is no process for SSH server.
Remediation:
Uninstall SSH server from the container and use nsenteror any other commands such as
docker exec or docker attach to interact with the container instance.
docker exec --interactive --tty $INSTANCE_ID sh
OR
docker attach $INSTANCE_ID
146 | P a g e
Impact:
None.
Default Value:
By default, SSH server is not running inside the container. Only one process per container is
allowed.
References:
1. http://blog.docker.com/2014/06/why-you-dont-need-to-run-sshd-in-docker/
CIS Controls:

on each system.
147 | P a g e
5.7 Ensure privileged ports are not mapped within containers (Scored)
Description:
The TCP/IP port numbers below 1024are considered privileged ports. Normal users and
processes are not allowed to use them for various security reasons. Docker allows a
container port to be mapped to a privileged port.
Rationale:
By default, if the user does not specifically declare the container port to host port mapping,
Docker automatically and correctly maps the container port to one available in 49153-
65535 block on the host. But, Docker allows a container port to be mapped to a privileged
port on the host if the user explicitly declared it. This is so because containers are executed
with NET_BIND_SERVICE Linux kernel capability that does not restrict the privileged port
mapping. The privileged ports receive and transmit various sensitive and privileged data.
Allowing containers to use them can bring serious implications.
Audit:
List all running containers instances and their port mapping by executing the below
command:
docker ps --quiet | xargs docker inspect --format '{{ .Id }}: Ports={{
.NetworkSettings.Ports }}'
Review the list and ensure that container ports are not mapped to host port numbers
below 1024.
Remediation:
Do not map the container ports to privileged host ports when starting a container. Also,
ensure that there is no such container to host privileged port mapping declarations in the
Dockerfile.
Impact:
None.
148 | P a g e
Default Value:
By default, mapping a container port to a privileged port on the host is allowed.
Note: There might be certain cases where you want to map privileged ports, because if you
forbid it, then the corresponding application has to run outside of a container.
For example: HTTP and HTTPS load balancers have to bind 80/tcp and 443/tcp
respectively. Forbidding to map privileged ports effectively forbids from running those in a
container, and mandates using an external load balancer. In such cases, those containers
instances should be marked as exceptions for this recommendation.
References:
CIS Controls:

on each system.
149 | P a g e
5.8 Ensure only needed ports are open on the container (Scored)
Description:
Dockerfile for a container image defines the ports to be opened by default on a container
instance. The list of ports may or may not be relevant to the application you are running
within the container.
Rationale:
A container can be run just with the ports defined in the Dockerfile for its image or can be
arbitrarily passed run time parameters to open a list of ports. Additionally, Overtime,
Dockerfile may undergo various changes and the list of exposed ports may or may not be
relevant to the application you are running within the container. Opening unneeded ports
increase the attack surface of the container and the containerized application. As a
recommended practice, do not open unneeded ports.
Audit:
List all the running instances of containers and their port mapping by executing the below
command:
Review the list and ensure that the ports mapped are the ones that are really needed for
the container.
Remediation:
Fix the Dockerfile of the container image to expose only needed ports by your
containerized application. You can also completely ignore the list of ports defined in the
Dockerfile by NOT using -P (UPPERCASE) or --publish-all flag when starting the
container. Use the -p (lowercase) or --publish flag to explicitly define the ports that you
need for a particular container instance.
For example,
docker run --interactive --tty --publish 5000 --publish 5001 --publish 5002
centos /bin/bash
150 | P a g e
Impact:
None.
Default Value:
By default, all the ports that are listed in the Dockerfile under EXPOSE instruction for an
image are opened when a container is run with -P or --publish-all flag.
References:
CIS Controls:

on each system.
151 | P a g e
5.9 Ensure the host's network namespace is not shared (Scored)
Description:
The networking mode on a container when set to --net=host, skips placing the container
inside separate network stack. In essence, this choice tells Docker to not containerize the
container's networking. This would network-wise mean that the container lives "outside"
in the main Docker host and has full access to its network interfaces.
Rationale:
This is potentially dangerous. It allows the container process to open low-numbered ports
like any other root process. It also allows the container to access network services like D-
bus on the Docker host. Thus, a container process can potentially do unexpected things
such as shutting down the Docker host. You should not use this option.
Audit:
NetworkMode={{ .HostConfig.NetworkMode }}'
If the above command returns NetworkMode=host, it means that --net=host option was
passed when container was started. This would be non-compliant.
Remediation:
Do not pass --net=host option when starting the container.
Impact:
None.
Default Value:
By default, container connects to Docker bridge.
References:
2. https://docs.docker.com/engine/reference/run/#network-settings
152 | P a g e
CIS Controls:
12 Boundary Defense
Boundary Defense
153 | P a g e
5.10 Ensure memory usage for container is limited (Scored)
Description:
By default, all containers on a Docker host share the resources equally. By using the
resource management capabilities of Docker host, such as memory limit, you can control
the amount of memory that a container may consume.
Rationale:
By default, container can use all of the memory on the host. You can use memory limit
mechanism to prevent a denial of service arising from one container consuming all of the
host’s resources such that other containers on the same host cannot perform their intended
functions. Having no limit on memory can lead to issues where one container can easily
make the whole system unstable and as a result unusable.
Audit:
docker ps --quiet --all | xargs docker inspect --format '{{ .Id }}: Memory={{
.HostConfig.Memory }}'
If the above command returns 0, it means the memory limits are not in place. If the above
command returns a non-zero value, it means memory limits are in place.
Remediation:
Run the container with only as much memory as required. Always run the container using
the --memory argument.
For example, you could run a container as below:
docker run --interactive --tty --memory 256m centos /bin/bash
In the above example, the container is started with a memory limit of 256 MB.
Note: Please note that the output of the below command would return values in scientific
notation if memory limits are in place.
docker inspect --format='{{.Config.Memory}}' 7c5a2d4c7fe0
154 | P a g e
For example, if the memory limit is set to 256 MB for the above container instance, the
output of the above command would be 2.68435456e+08 and NOT 256m. You should
convert this value using a scientific calculator or programmatic methods.
Impact:
If you do not set proper limits, the container process may have to starve.
Default Value:
By default, all containers on a Docker host share the resources equally. No memory limits
are enforced.
References:
1. https://goldmann.pl/blog/2014/09/11/resource-management-in-docker/
2. https://docs.docker.com/engine/reference/commandline/run/#options
3. https://docs.docker.com/engine/admin/runmetrics/
CIS Controls:

155 | P a g e
5.11 Ensure CPU priority is set appropriately on the container (Scored)
Description:
By default, all containers on a Docker host share the resources equally. By using the
resource management capabilities of Docker host, such as CPU shares, you can control the
host CPU resources that a container may consume.
Rationale:
By default, CPU time is divided between containers equally. If it is desired, to control the
CPU time amongst the container instances, you can use CPU sharing feature. CPU sharing
allows to prioritize one container over the other and forbids the lower priority container to
claim CPU resources more often. This ensures that the high priority containers are served
better.
Audit:
CpuShares={{ .HostConfig.CpuShares }}'
If the above command returns 0 or 1024, it means the CPU shares are not in place. If the
above command returns a non-zero value other than 1024, it means CPU shares are in
place.
Remediation:
Manage the CPU shares between your containers. To do so start the container using the --
cpu-shares argument.
For example, you could run a container as below:
docker run --interactive --tty --cpu-shares 512 centos /bin/bash
In the above example, the container is started with CPU shares of 50% of what the other
containers use. So, if the other container has CPU shares of 80%, this container will have
CPU shares of 40%.
Note: Every new container will have 1024 shares of CPU by default. However, this value is
shown as 0 if you run the command mentioned in the audit section.
156 | P a g e
Alternatively,
1. Navigate to /sys/fs/cgroup/cpu/system.slice/ directory.

2. Check your container instance ID using docker ps.
3. Now, inside the above directory (in step 1), you would have a directory by name
docker-<Instance ID>.scope. For example, docker-
4acae729e8659c6be696ee35b2237cc1fe4edd2672e9186434c5116e1a6fbed6.scope.
Navigate to this directory.
4. You will find a file named cpu.shares. Execute cat cpu.shares. This will always
give you the CPU share value based on the system. So, even if there is no CPU shares
configured using -c or --cpu-shares argument in the docker run command, this
file will have a value of 1024.
If we set one container’s CPU shares to 512 it will receive half of the CPU time compared to
the other container. So, take 1024 as 100% and then do quick math to derive the number
that you should set for respective CPU shares. For example, use 512 if you want to set 50%
and 256 if you want to set 25%.
Impact:
If you do not set proper CPU shares, the container process may have to starve if the
resources on the host are not available. If the CPU resources on the host are free, CPU
shares do not place any restrictions on the CPU that the container may use.
Default Value:
By default, all containers on a Docker host share the resources equally. No CPU shares are
enforced.
References:
1. https://goldmann.pl/blog/2014/09/11/resource-management-in-docker/
3. https://docs.docker.com/engine/admin/runmetrics/
CIS Controls:

157 | P a g e
5.12 Ensure the container's root filesystem is mounted as read only
(Scored)
Description:
The container's root filesystem should be treated as a 'golden image' by using Docker run's
--read-only option. This prevents any writes to the container's root filesystem at
container runtime and enforces the principle of immutable infrastructure.
Rationale:
Enabling this option forces containers at runtime to explicitly define their data writing
strategy to persist or not persist their data.
This also reduces security attack vectors since the container instance's filesystem cannot
be tampered with or written to unless it has explicit read-write permissions on its
filesystem folder and directories.
Audit:
Run the following command on the docker host:
ReadonlyRootfs={{ .HostConfig.ReadonlyRootfs }}'
If the above command returns true, it means the container's root filesystem is mounted
read-only.
If the above command returns false, it means the container's root filesystem is writable.
Remediation:
Add a --read-only flag at a container's runtime to enforce the container's root filesystem
to be mounted as read only.
docker run <Run arguments> --read-only <Container Image Name or ID> <Command>
Enabling the --read-only option at a container's runtime should be used by administrators

to force a container's executable processes to only write container data to explicit storage
locations during the container's runtime.
158 | P a g e
Examples of explicit storage locations during a container's runtime include, but not limited
to:
1. Use the --tmpfs option to mount a temporary file system for non-persistent data
writes.
docker run --interactive --tty --read-only --tmpfs "/run" --tmpfs "/tmp"

centos /bin/bash
2. Enabling Docker rw mounts at a container's runtime to persist container data

directly on the Docker host filesystem.
docker run --interactive --tty --read-only -v /opt/app/data:/run/app/data:rw

centos /bin/bash
3. Utilizing Docker shared-storage volume plugins for Docker data volume to persist
container data.
docker volume create -d convoy --opt o=size=20GB my-named-volume

docker run --interactive --tty --read-only -v my-named-volume:/run/app/data
centos /bin/bash
3. Transmitting container data outside of the docker during the container's runtime
for container data to persist container data. Examples include hosted databases,
network file shares, and APIs.
Impact:
Enabling --read-only at container runtime may break some container OS packages if a

data writing strategy is not defined.
Define what the container's data should and should not persist at runtime to determine
which recommendation procedure to utilize.
Example:
 Enable use --tmpfs for temporary file writes to /tmp

 Use Docker shared data volumes for persistent data writes
Default Value:
By default, a container will have its root filesystem writable allowing all container
processes to write files owned by the container's runtime user.
159 | P a g e
References:
1. http://docs.docker.com/reference/commandline/cli/#run
2. https://docs.docker.com/engine/tutorials/dockervolumes/
3. http://www.projectatomic.io/blog/2015/12/making-docker-images-write-only-in-
production/
4. https://docs.docker.com/engine/reference/commandline/run/#mount-tmpfs-
tmpfs
5. https://docs.docker.com/engine/tutorials/dockervolumes/#creating-and-
mounting-a-data-volume-container
CIS Controls:

160 | P a g e
5.13 Ensure incoming container traffic is binded to a specific host
interface (Scored)
Description:
By default, Docker containers can make connections to the outside world, but the outside
world cannot connect to containers. Each outgoing connection will appear to originate
from one of the host machine's own IP addresses. Only allow container services to be
contacted through a specific external interface on the host machine.
Rationale:
If you have multiple network interfaces on your host machine, the container can accept
connections on the exposed ports on any network interface. This might not be desired and
may not be secured. Many a times a particular interface is exposed externally and services
such as intrusion detection, intrusion prevention, firewall, load balancing, etc. are run on
those interfaces to screen incoming public traffic. Hence, you should not accept incoming
connections on any interface. You should only allow incoming connections from a
particular external interface.
Audit:
List all the running instances of containers and their port mapping by executing the below
command:
Review the list and ensure that the exposed container ports are tied to a particular
interface and not to the wildcard IP address - 0.0.0.0.
For example, if the above command returns as below, then this is non-compliant and the
container can accept connections on any host interface on the specified port 49153.
Ports=map[443/tcp:<nil> 80/tcp:[map[HostPort:49153 HostIp:0.0.0.0]]]
However, if the exposed port is tied to a particular interface on the host as below, then this
recommendation is configured as desired and is compliant.
Ports=map[443/tcp:<nil> 80/tcp:[map[HostIp:10.2.3.4 HostPort:49153]]]
161 | P a g e
Remediation:
Bind the container port to a specific host interface on the desired host port.
For example,
docker run --detach --publish 10.2.3.4:49153:80 nginx
In the example above, the container port 80 is bound to the host port on 49153 and would
accept incoming connection only from 10.2.3.4 external interface.
Impact:
None.
Default Value:
By default, Docker exposes the container ports on 0.0.0.0, the wildcard IP address that
will match any possible incoming network interface on the host machine.
References:
CIS Controls:
9 Limitation and Control of Network Ports, Protocols, and Services

Limitation and Control of Network Ports, Protocols, and Services
162 | P a g e
5.14 Ensure 'on-failure' container restart policy is set to '5' (Scored)
Description:
Using the --restart flag in docker run command you can specify a restart policy for how a
container should or should not be restarted on exit. You should choose the on-failure
restart policy and limit the restart attempts to 5.
Rationale:
If you indefinitely keep trying to start the container, it could possibly lead to a denial of
service on the host. It could be an easy way to do a distributed denial of service attack
especially if you have many containers on the same host. Additionally, ignoring the exit
status of the container and always attempting to restart the container leads to non-
investigation of the root cause behind containers getting terminated. If a container gets
terminated, you should investigate on the reason behind it instead of just attempting to
restart it indefinitely. Thus, it is recommended to use on-failure restart policy and limit it
to maximum of 5 restart attempts.
Audit:
RestartPolicyName={{ .HostConfig.RestartPolicy.Name }} MaximumRetryCount={{
.HostConfig.RestartPolicy.MaximumRetryCount }}'
 If the above command returns RestartPolicyName=always, then the system is not

configured as desired and hence this recommendation is non-compliant.
 If the above command returns RestartPolicyName=no or just RestartPolicyName=,
then the restart policies are not being used and the container would never be
restarted of its own. This recommendation is then Not Applicable and can be
assumed to be compliant.
 If the above command returns RestartPolicyName=on-failure, then verify that the
number of restart attempts is set to 5 or less by looking at MaximumRetryCount.
Remediation:
If a container is desired to be restarted of its own, then, for example, you could start the
container as below:
docker run --detach --restart=on-failure:5 nginx
163 | P a g e
Impact:
The container would attempt to restart only for 5 times.
Default Value:
By default, containers are not configured with restart policies. Hence, containers do not
attempt to restart of their own.
References:
1. https://docs.docker.com/engine/reference/commandline/run/#restart-policies-
restart
CIS Controls:

164 | P a g e
5.15 Ensure the host's process namespace is not shared (Scored)
Description:
Process ID (PID) namespaces isolate the process ID number space, meaning that processes
in different PID namespaces can have the same PID. This is process level isolation between
containers and the host.
Rationale:
PID namespace provides separation of processes. The PID Namespace removes the view of
the system processes, and allows process ids to be reused including PID 1. If the host's PID
namespace is shared with the container, it would basically allow processes within the
container to see all of the processes on the host system. This breaks the benefit of process
level isolation between the host and the containers. Someone having access to the
container can eventually know all the processes running on the host system and can even
kill the host system processes from within the container. This can be catastrophic. Hence,
do not share the host's process namespace with the containers.
Audit:
PidMode={{ .HostConfig.PidMode }}'
If the above command returns host, it means the host PID namespace is shared with the
container else this recommendation is compliant.
Remediation:
Do not start a container with --pid=host argument.
docker run --interactive --tty --pid=host centos /bin/bash
Impact:
Container processes cannot see the processes on the host system. In certain cases, you
want your container to share the host's process namespace. For example, you could build a
container with debugging tools like strace or gdb, but want to use these tools when
165 | P a g e
debugging processes within the container. If this is desired, then share only one (or
needed) host process by using the -p switch.
For example,
docker run --pid=host rhel7 strace -p 1234
Default Value:
By default, all containers have the PID namespace enabled and the host's process
namespace is not shared with the containers.
References:
1. https://docs.docker.com/engine/reference/run/#pid-settings-pid
2. http://man7.org/linux/man-pages/man7/pid_namespaces.7.html
CIS Controls:

166 | P a g e
5.16 Ensure the host's IPC namespace is not shared (Scored)
Description:
IPC (POSIX/SysV IPC) namespace provides separation of named shared memory segments,
semaphores and message queues. IPC namespace on the host thus should not be shared
with the containers and should remain isolated.
Rationale:
IPC namespace provides separation of IPC between the host and containers. If the host's
IPC namespace is shared with the container, it would basically allow processes within the
container to see all of the IPC on the host system. This breaks the benefit of IPC level
isolation between the host and the containers. Someone having access to the container can
eventually manipulate the host IPC. This can be catastrophic. Hence, do not share the host's
IPC namespace with the containers.
Audit:
IpcMode={{ .HostConfig.IpcMode }}'
If the above command returns host, it means the host IPC namespace is shared with the
container. If the above command returns nothing, then the host's IPC namespace is not
shared. This recommendation is then compliant.
Remediation:
Do not start a container with --ipc=host argument. For example, do not start a container
as below:
docker run --interactive --tty --ipc=host centos /bin/bash
Impact:
Shared memory segments are used to accelerate inter-process communication. It is

commonly used by high-performance applications. If such applications are containerized
into multiple containers, you might need to share the IPC namespace of the containers to
achieve high performance. In such cases, you should still be sharing container specific IPC
167 | P a g e
namespaces only and not the host IPC namespace. You may share the container's IPC
namespace with another container as below:
For example,
docker run --interactive --tty --ipc=container:e3a7a1a97c58 centos /bin/bash
Default Value:
By default, all containers have the IPC namespace enabled and host IPC namespace is not
shared with any container.
References:
1. https://docs.docker.com/engine/reference/run/#ipc-settings-ipc
2. http://man7.org/linux/man-pages/man7/namespaces.7.html
CIS Controls:

168 | P a g e
5.17 Ensure host devices are not directly exposed to containers (Not
Scored)
Description:
Host devices can be directly exposed to containers at runtime. Do not directly expose host
devices to containers especially for containers that are not trusted.
Rationale:
The --device option exposes the host devices to the containers and consequently, the
containers can directly access such host devices. You would not require the container to
run in privileged mode to access and manipulate the host devices. By default, the
container will be able to read, write and mknod these devices. Additionally, it is possible for
containers to remove block devices from the host. Hence, do not expose host devices to
containers directly.
If at all, you would want to expose the host device to a container, use the sharing
permissions appropriately:
 r - read only
 w - writable
 m - mknod allowed
Audit:
Devices={{ .HostConfig.Devices }}'
The above command would list out each device with below information:
 CgroupPermissions - For example, rwm

 PathInContainer - Device path within the container
 PathOnHost - Device path on the host
Verify that the host device is needed to be accessed from within the container and the
permissions required are correctly set. If the above command returns [], then the container
does not have access to host devices. This recommendation can be assumed to be
compliant.
169 | P a g e
Remediation:
Do not directly expose the host devices to containers. If at all, you need to expose the host
devices to containers, use the correct set of permissions:
docker run --interactive --tty --device=/dev/tty0:/dev/tty0:rwm --

device=/dev/temp_sda:/dev/temp_sda:rwm centos bash
For example, share the host device with correct permissions:
docker run --interactive --tty --device=/dev/tty0:/dev/tty0:rw --

device=/dev/temp_sda:/dev/temp_sda:r centos bash
Impact:
You would not be able to use the host devices directly within the containers.
Default Value:
By default, no host devices are exposed to containers. If you do not provide sharing
permissions and choose to expose a host device to a container, the host device would be
exposed with read, write and mknod permissions.
References:
CIS Controls:

170 | P a g e
5.18 Ensure the default ulimit is overwritten at runtime, only if needed
(Not Scored)
Description:
The default ulimit is set at the Docker daemon level. However, you may override the default
ulimit setting, if needed, during container runtime.
Rationale:
ulimit provides control over the resources available to the shell and to processes started
by it. Setting system resource limits judiciously saves you from many disasters such as a
fork bomb. Sometimes, even friendly users and legitimate processes can overuse system
resources and in-turn can make the system unusable.
The default ulimit set at the Docker daemon level should be honored. If the default ulimit
settings are not appropriate for a particular container instance, you may override them as
an exception. But, do not make this a practice. If most of the container instances are
overriding default ulimit settings, consider changing the default ulimit settings to
something that is appropriate for your needs.
Audit:
Ulimits={{ .HostConfig.Ulimits }}'
The above command should return Ulimits=<no value> for each container instance until
and unless there is an exception and a need to override the default ulimit settings.
Remediation:
Only override the default ulimit settings if needed.
For example, to override default ulimit settings start a container as below:
docker run --ulimit nofile=1024:1024 --interactive --tty centos /bin/bash
Impact:
If the ulimits are not set properly, the desired resource control might not be achieved and
might even make the system unusable.
171 | P a g e
Default Value:
Container instances inherit the default ulimit settings set at the Docker daemon level.
References:
1. https://docs.docker.com/engine/reference/commandline/run/#set-ulimits-in-
container-ulimit
CIS Controls:

172 | P a g e
5.19 Ensure mount propagation mode is not set to shared (Scored)
Description:
Mount propagation mode allows mounting volumes in shared, slave or private mode on a
container. Do not use shared mount propagation mode until needed.
Rationale:
A shared mount is replicated at all mounts and the changes made at any mount point are
propagated to all mounts. Mounting a volume in shared mode does not restrict any other
container to mount and make changes to that volume. This might be catastrophic if the
mounted volume is sensitive to changes. Do not set mount propagation mode to shared
until needed.
Audit:
Propagation={{range $mnt := .Mounts}} {{json $mnt.Propagation}} {{end}}'
The above command would return the propagation mode for mounted volumes.
Propagation mode should not be set to shared unless needed. The above command might
throw errors if there are no mounts. In that case, this recommendation is not applicable.
Remediation:
Do not mount volumes in shared mode propagation.
For example, do not start container as below:
docker run <Run arguments> --volume=/hostPath:/containerPath:shared

<Container Image Name or ID> <Command>
Impact:
None.
Default Value:
By default, the container mounts are private.
173 | P a g e
References:
2. https://docs.docker.com/engine/reference/run/#volume-shared-filesystems
3. https://www.kernel.org/doc/Documentation/filesystems/sharedsubtree.txt
CIS Controls:

174 | P a g e
5.20 Ensure the host's UTS namespace is not shared (Scored)
Description:
UTS namespaces provide isolation of two system identifiers: the hostname and the NIS
domain name. It is used for setting the hostname and the domain that is visible to running
processes in that namespace. Processes running within containers do not typically require
to know hostname and domain name. Hence, the namespace should not be shared with the
host.
Rationale:
Sharing the UTS namespace with the host provides full permission to the container to
change the hostname of the host. This is insecure and should not be allowed.
Audit:
UTSMode={{ .HostConfig.UTSMode }}'
If the above command returns host, it means the host UTS namespace is shared with the
container and this recommendation is non-compliant. If the above command returns
nothing, then the host's UTS namespace is not shared. This recommendation is then
compliant.
Remediation:
Do not start a container with --uts=host argument.
docker run --rm --interactive --tty --uts=host rhel7.2
Impact:
None.
Default Value:
By default, all containers have the UTS namespace enabled and host UTS namespace is not
shared with any container.
175 | P a g e
References:
1. https://docs.docker.com/engine/reference/run/#uts-settings-uts
2. http://man7.org/linux/man-pages/man7/namespaces.7.html
CIS Controls:

176 | P a g e
5.21 Ensure the default seccomp profile is not Disabled (Scored)
Description:
Seccomp filtering provides a means for a process to specify a filter for incoming system
calls. The default Docker seccomp profile works on whitelist basis and allows 311 system
calls blocking all others. It should not be disabled unless it hinders your container
application usage.
Rationale:
A large number of system calls are exposed to every userland process with many of them
going unused for the entire lifetime of the process. Most of the applications do not need all
the system calls and thus benefit by having a reduced set of available system calls. The
reduced set of system calls reduces the total kernel surface exposed to the application and
thus improvises application security.
Audit:
The above command should return <no value> or your modified seccomp profile. If it
returns [seccomp:unconfined], that means this recommendation is non-compliant and the
container is running without any seccomp profiles.
Remediation:
By default, seccomp profiles are enabled. You do not need to do anything unless you want
to modify and use the modified seccomp profile.
Impact:
With Docker 1.10 and greater, the default seccomp profile blocks syscalls, regardless of --
cap-add passed to the container. You should create your own custom seccomp profile in
such cases. You may also disable the default seccomp profile by passing --security-
opt=seccomp:unconfined on docker run.
177 | P a g e
Default Value:
When you run a container, it uses the default profile unless you override it with the --
security-opt option.
References:
1. http://blog.scalock.com/new-docker-security-features-and-what-they-mean-
seccomp-profiles
3. https://github.com/docker/docker/blob/master/profiles/seccomp/default.json
4. https://docs.docker.com/engine/security/seccomp/
5. https://www.kernel.org/doc/Documentation/prctl/seccomp_filter.txt
CIS Controls:

178 | P a g e
5.22 Ensure docker exec commands are not used with privileged option
(Scored)
Description:
Do not docker exec with --privileged option.
Rationale:
Using --privileged option in docker exec gives extended Linux capabilities to the
command. This could potentially be insecure and unsafe to do especially when you are
running containers with dropped capabilities or with enhanced restrictions.
Audit:
If you have auditing enabled as prescribed in Section 1, you can use the below command to
filter out docker exec commands that used --privileged option.
ausearch -k docker | grep exec | grep privileged
Remediation:
Do not use --privileged option in docker exec command.
Impact:
None. If you need enhanced capabilities within the container, then run the container with
the needed capabilities.
Default Value:
By default, docker exec command runs without --privileged option.
References:
1. https://docs.docker.com/engine/reference/commandline/exec/
CIS Controls:

179 | P a g e
180 | P a g e
5.23 Ensure docker exec commands are not used with user option
(Scored)
Description:
Do not docker exec with --user option.
Rationale:
Using --user option in docker exec executes the command within the container as that
user. This could potentially be insecure and unsafe to do especially when you are running
containers with dropped capabilities or with enhanced restrictions.
For example, suppose your container is running as tomcat user (or any other non-root
user), it would be possible to run a command through docker exec as rootwith --
user=root option. This could potentially be dangerous.
Audit:
If you have auditing enabled as prescribed in Section 1, you can use the below command to
filter out docker exec commands that used --user option.
ausearch -k docker | grep exec | grep user
Remediation:
Do not use --user option in docker exec command.
Impact:
None.
Default Value:
By default, docker exec command runs without --user option.
References:
1. https://docs.docker.com/engine/reference/commandline/exec/
181 | P a g e
CIS Controls:

182 | P a g e
5.24 Ensure cgroup usage is confirmed (Scored)
Description:
It is possible to attach to a particular cgroup on container run. Confirming cgroup usage

would ensure that containers are running under defined cgroups.
Rationale:
System administrators typically define cgroups under which containers are supposed to
run. Even if cgroups are not explicitly defined by the system administrators, containers run
under docker cgroup by default.
At run-time, it is possible to attach to a different cgroup other than the one that was
expected to be used. This usage should be monitored and confirmed. By attaching to a
different cgroup than the one that is expected, excess permissions and resources might be
granted to the container and thus, can prove to be unsafe.
Audit:
CgroupParent={{ .HostConfig.CgroupParent }}'
The above command would return the cgroup under which the containers are running. If it
is blank, it means containers are running under default docker cgroup. In that case, this
recommendation is compliant. If the containers are found to be running under cgroup
other than the one that was expected, this recommendation is non-compliant.
Remediation:
Do not use --cgroup-parent option in docker run command unless needed.
Impact:
None.
Default Value:
By default, containers run under docker cgroup.
183 | P a g e
References:
1. https://docs.docker.com/engine/reference/run/#specify-custom-cgroups
US/Red_Hat_Enterprise_Linux/6/html/Resource_Management_Guide/ch01.html
CIS Controls:

184 | P a g e
5.25 Ensure the container is restricted from acquiring additional
privileges (Scored)
Description:
Restrict the container from acquiring additional privileges via suid or sgid bits.
Rationale:
A process can set the no_new_priv bit in the kernel. It persists across fork, clone and
execve. The no_new_priv bit ensures that the process or its children processes do not gain
any additional privileges via suid or sgid bits. This way a lot of dangerous operations
become a lot less dangerous because there is no possibility of subverting privileged
binaries.
Audit:
The above command should return all the security options currently configured for the
containers. no-new-privileges should also be one of them.
Remediation:
For example, you should start your container as below:
docker run --rm -it --security-opt=no-new-privileges ubuntu bash
Impact:
no_new_priv prevents LSMs like SELinux from transitioning to process labels that have
access not allowed to the current process.
Default Value:
By default, new privileges are not restricted.
References:
1. https://github.com/projectatomic/atomic-site/issues/269
185 | P a g e
3. https://www.kernel.org/doc/Documentation/prctl/no_new_privs.txt
4. https://lwn.net/Articles/475678/
5. https://lwn.net/Articles/475362/
CIS Controls:

186 | P a g e
5.26 Ensure container health is checked at runtime (Scored)
Description:
If the container image does not have an HEALTHCHECK instruction defined, use --health-cmd
parameter at container runtime for checking container health.
Rationale:
One of the important security triads is availability. If the container image you are using
does not have a pre-defined HEALTHCHECK instruction, use the --health-cmd parameter to
check container health at runtime.
Based on the reported health status, you could take necessary actions.
Audit:
Run the below command and ensure that all the containers are reporting health status:
docker ps --quiet | xargs docker inspect --format '{{ .Id }}: Health={{
.State.Health.Status }}'
Remediation:
Run the container using --health-cmd and the other parameters.
For example,
docker run -d --health-cmd='stat /etc/passwd || exit 1' nginx
Impact:
None.
Default Value:
By default, health checks are not done at container runtime.
References:
1. https://docs.docker.com/engine/reference/run/#healthcheck
187 | P a g e
CIS Controls:

188 | P a g e
5.27 Ensure docker commands always get the latest version of the
image (Not Scored)
Description:
Always ensure that you are using the latest version of the image within your repository and
not the cached older versions.
Rationale:
Multiple docker commands such as docker pull, docker run, etc. are known to have an
issue that by default, they extract the local copy of the image, if present, even though there
is an updated version of the image with the "same tag" in the upstream repository. This
could lead to using older and vulnerable images.
Audit:
Step 1: Open your image repository and list the image version history for the image you
are inspecting.
Step 2: Observe the status when the docker pull command is triggered.
If the status is shown as Image is up to date, it means that you are getting the cached
version of the image.
Step 3: Match the version of the image you are running with the latest version reported in
your repository which tells if you are running the cached version or the latest copy.
Remediation:
Use proper version pinning mechanisms (the latest tag which is assigned by default is still
vulnerable to caching attacks) to avoid extracting the cached older versions. Version
pinning mechanisms should be used for base images, packages, and entire images too. You
can customize version pinning rules as per your requirements.
Impact:
None
189 | P a g e
Default Value:
By default, docker commands extract the local copy unless version pinning mechanisms are
used or the local cache is cleared.
References:
CIS Controls:

190 | P a g e
5.28 Ensure PIDs cgroup limit is used (Scored)
Description:
Use --pids-limit flag at container runtime.
Rationale:
Attackers could launch a fork bomb with a single command inside the container. This fork
bomb can crash the entire system and requires a restart of the host to make the system
functional again. PIDs cgroup --pids-limit will prevent this kind of attacks by restricting
the number of forks that can happen inside a container at a given time.
Audit:
Run the below command and ensure that PidsLimit is not set to 0 or -1. A PidsLimit of 0
or -1 means that any number of processes can be forked inside the container concurrently.
PidsLimit={{ .HostConfig.PidsLimit }}'
Remediation:
Use --pids-limit flag while launching the container with an appropriate value.
For example,
docker run -it --pids-limit 100 <Image_ID>
In the above example, the number of processes allowed to run at any given time is set to
100. After a limit of 100 concurrently running processes is reached, docker would restrict
any new process creation.
Impact:
Set the PIDs limit value as appropriate. Incorrect values might leave the containers
unusable.
Default Value:
The Default value for --pids-limit is 0 which means there is no restriction on the number
of forks. Also, note that PIDs cgroup limit works only for the kernel versions 4.3+.
191 | P a g e
References:
CIS Controls:

192 | P a g e
5.29 Ensure Docker's default bridge docker0 is not used (Not Scored)
Description:
Do not use Docker's default bridge docker0. Use docker's user-defined networks for
container networking.
Rationale:
Docker connects virtual interfaces created in the bridge mode to a common bridge called
docker0. This default networking model is vulnerable to ARP spoofing and MAC flooding
attacks since there is no filtering applied.
Audit:
Run the below command, and verify that containers are on a user-defined network and not
the default docker0 bridge.
docker network ls --quiet | xargs xargs docker network inspect --format '{{
.Name }}: {{ .Options }}'
Remediation:
Follow Docker documentation and setup a user-defined network. Run all the containers in
the defined network.
Impact:
You have to manage the user-defined networks.
Default Value:
By default, docker runs containers on its docker0 bridge.
References:
1. https://github.com/nyantec/narwhal
2. https://arxiv.org/pdf/1501.02967
193 | P a g e
CIS Controls:

194 | P a g e
5.30 Ensure the host's user namespaces is not shared (Scored)
Description:
Do not share the host's user namespaces with the containers.
Rationale:
User namespaces ensure that a root process inside the container will be mapped to a non-
root process outside the container. Sharing the user namespaces of the host with the
container thus does not isolate users on the host with users on the containers.
Audit:
Run the below command and ensure that it does not return any value for UsernsMode. If it
returns a value of host, it means the host user namespace is shared with the containers.
UsernsMode={{ .HostConfig.UsernsMode }}'
Remediation:
Do not share user namespaces between host and containers.
For example, do not run a container as below:
docker run --rm -it --userns=host ubuntu bash
Impact:
None
Default Value:
By default, the host user namespace is shared with the containers until user namespace
support is enabled.
References:
1. https://docs.docker.com/engine/security/userns-remap/
195 | P a g e
4. https://events.linuxfoundation.org/sites/events/files/slides/User%20Namespaces
%20-%20ContainerCon%202015%20-%2016-9-final_0.pdf
CIS Controls:
12 Boundary Defense
Boundary Defense
196 | P a g e
5.31 Ensure the Docker socket is not mounted inside any containers
(Scored)
Description:
The docker socket docker.sock should not be mounted inside a container.
Rationale:
If the docker socket is mounted inside a container it would allow processes running within
the container to execute docker commands which effectively allows for full control of the
host.
Audit:
Volumes={{ .Mounts }}' | grep docker.sock
The above command would return any instances where docker.sock had been mapped to
a container as a volume.
Remediation:
Ensure that no containers mount docker.sock as a volume.
Impact:
None
Default Value:
By default, docker.sock is not mounted inside containers.
References:
1. https://raesene.github.io/blog/2016/03/06/The-Dangers-Of-Docker.sock/
2. https://forums.docker.com/t/docker-in-docker-vs-mounting-var-run-docker-
sock/9450/2
197 | P a g e
CIS Controls:

198 | P a g e
6 Docker Security Operations
This sections covers some of the operational security aspects for Docker deployments.
These are best practices that should be followed. Most of the recommendations here are
just reminders that organizations should extend their current security best practices and
policies to include containers.
6.1 Ensure image sprawl is avoided (Not Scored)

Description:
Do not keep a large number of container images on the same host. Use only tagged images
as appropriate.
Rationale:
Tagged images are useful to fall back from "latest" to a specific version of an image in
production. Images with unused or old tags may contain vulnerabilities that might be
exploited, if instantiated. Additionally, if you fail to remove unused images from the system
and there are various such redundant and unused images, the host filesystem may become
full and could lead to denial of service.
Audit:
Step 1 Make a list of all image IDs that are currently instantiated by executing below
command:
docker images --quiet | xargs docker inspect --format '{{ .Id }}: Image={{
.Config.Image }}'
Step 2: List all the images present on the system by executing below command:
docker images
Step 3: Compare the list of image IDs populated from Step 1 and Step 2 and find out images
that are currently not being instantiated. If any such unused or old images are found,
discuss with the system administrator the need to keep such images on the system. If such
a need is not justified enough, then this recommendation is non-compliant.
199 | P a g e
Remediation:
Keep the set of the images that you actually need and establish a workflow to remove old or
stale images from the host. Additionally, use features such as pull-by-digest to get specific
images from the registry.
Additionally, you can follow below set of steps to find out unused images on the system and
delete them.
Step 1 Make a list of all image IDs that are currently instantiated by executing below
command:
docker images --quiet | xargs docker inspect --format '{{ .Id }}: Image={{
.Config.Image }}'
Step 2: List all the images present on the system by executing below command:
docker images
Step 3: Compare the list of image IDs populated from Step 1 and Step 2 and find out images
that are currently not being instantiated.
Step 4: Decide if you want to keep the images that are not currently in use. If not delete
them by executing below command:
docker rmi $IMAGE_ID
Impact:
None
Default Value:
Images and layered filesystems remain accessible on the host until the administrator
removes all tags that refer to those images or layers.
References:
1. http://craiccomputing.blogspot.in/2014/09/clean-up-unused-docker-containers-
and.html
2. https://forums.docker.com/t/command-to-remove-all-unused-images/20/8
4. https://docs.docker.com/engine/reference/commandline/rmi/
5. https://docs.docker.com/engine/reference/commandline/pull/
200 | P a g e
CIS Controls:

201 | P a g e
6.2 Ensure container sprawl is avoided (Not Scored)
Description:
Do not keep a large number of containers on the same host.
Rationale:
The flexibility of containers makes it easy to run multiple instances of applications and
indirectly leads to Docker images that exist at varying security patch levels. It also means
that you are consuming host resources that otherwise could have been used for running
'useful' containers. Having more than just the manageable number of containers on a
particular host makes the situation vulnerable to mishandling, misconfiguration and
fragmentation. Thus, avoid container sprawl and keep the number of containers on a host
to a manageable total.
Audit:
Step 1 - Find the total number of containers you have on the host:
docker info --format '{{ .Containers }}'
Step 2 - Execute the below commands to find the total number of containers that are
actually running or in the stopped state on the host.
docker info --format '{{ .ContainersStopped }}'

docker info --format '{{ .ContainersRunning }}'
If the difference between the number of containers that are stopped on the host and the
number of containers that are actually running on the host is large (say 25 or more), then
perhaps, the containers are sprawled on the host.
Remediation:
Periodically check your container inventory per host and clean up the stopped containers
using the below command:
docker container prune
202 | P a g e
Impact:
If you keep way too few number of containers per host, then perhaps you are not utilizing
your host resources very adequately.
Default Value:
By default, Docker does not restrict the number of containers you may have on a host.
References:
1. https://zeltser.com/security-risks-and-benefits-of-docker-application/
2. http://searchsdn.techtarget.com/feature/Docker-networking-How-Linux-
containers-will-change-your-network
CIS Controls:

203 | P a g e
7 Docker Swarm Configuration
This section lists the recommendations that alter and secure the behavior of the Docker
Swarm. If you are not using Docker Swarm then the recommendations in this section do
not apply.
7.1 Ensure swarm mode is not Enabled, if not needed (Scored)

Description:
Do not enable swarm mode on a docker engine instance unless needed.
Rationale:
By default, a Docker engine instance will not listen on any network ports, with all
communications with the client coming over the Unix socket. When Docker swarm mode is
enabled on a docker engine instance, multiple network ports are opened on the system and
made available to other systems on the network for the purposes of cluster management
and node communications.
Opening network ports on a system increase its attack surface and this should be avoided
unless required.
Audit:
Review the output of the docker info command. If the output includes Swarm: active it
indicates that swarm mode has been activated on the Docker engine. Confirm if swarm
mode on the docker engine instance is actually needed.
Remediation:
If swarm mode has been enabled on a system in error, run
docker swarm leave
Impact:
None.
204 | P a g e
Default Value:
By default, docker swarm mode is not enabled.
References:
1. https://docs.docker.com/engine/reference/commandline/swarm_init/
CIS Controls:

on each system.
205 | P a g e
7.2 Ensure the minimum number of manager nodes have been created
in a swarm (Scored)
Description:
Ensure that the minimum number of required manager nodes is created in a swarm.
Rationale:
Manager nodes within a swarm have control over the swarm and change its configuration
modifying security parameters. Having excessive manager nodes could render the swarm
more susceptible to compromise.
If fault tolerance is not required in the manager nodes, a single node should be elected as a
manger. If fault tolerance is required then the smallest practical odd number to achieve the
appropriate level of tolerance should be configured.
Audit:
Run docker info and verify the number of managers.
docker info --format '{{ .Swarm.Managers }}'
Alternatively run the below command.
docker node ls | grep 'Leader'
Remediation:
If an excessive number of managers is configured, the excess can be demoted as worker

using the following command:
docker node demote <ID>
Where is the node ID value of the manager to be demoted.
Impact:
None
206 | P a g e
Default Value:
A single manager is all that is required to start a given cluster.
References:
1. https://docs.docker.com/engine/swarm/manage-nodes/
2. https://docs.docker.com/engine/swarm/admin_guide/#/add-manager-nodes-for-
fault-tolerance
CIS Controls:

207 | P a g e
7.3 Ensure swarm services are binded to a specific host interface
(Scored)
Description:
By default, the docker swarm services will listen to all interfaces on the host, which may
not be necessary for the operation of the swarm where the host has multiple network
interfaces.
Rationale:
When a swarm is initialized the default value for the --listen-addr flag is 0.0.0.0:2377
which means that the swarm services will listen on all interfaces on the host. If a host has
multiple network interfaces this may be undesirable as it may expose the docker swarm
services to networks which are not involved in the operation of the swarm.
By passing a specific IP address to the --listen-addr, a specific network interface can be

specified limiting this exposure.
Audit:
List the network listener on port 2377/TCP (the default for docker swarm) and confirm
that it is only listening on specific interfaces. For example, using ubuntu this could be done
with the following command:
netstat -lt | grep -i 2377
Remediation:
Remediation of this requires re-initialization of the swarm specifying a specific interface

for the --listen-addr parameter.
Impact:
None
Default Value:
By default, docker swarm services listen on all available host interfaces.
208 | P a g e
References:
1. https://docs.docker.com/engine/reference/commandline/swarm_init/#--listen-
addr
2. https://docs.docker.com/engine/swarm/admin_guide/#recover-from-disaster
Notes:
A couple of points I noted looking at this one. there doesn't seem to be a parameter for
docker swarm update to change the listen-addr. For the remediation I did wonder if --
force-new-swarm could be used to change this, but I'm not sure what other effects that
would have on the swarm so just left with a general requirement to re-initialize the swarm.
Also interestingly the node communication service running on 7946/TCP doesn't respect
the --listen-addr parameter. this seems like a bug to me, I'll likely file an issue on github for
it after a bit more exploration.
CIS Controls:

209 | P a g e
7.4 Ensure data exchanged between containers are encrypted on
different nodes on the overlay network (Scored)
Description:
Encrypt data exchanged between containers on different nodes on the overlay network.
Rationale:
By default, data exchanged between containers on different nodes on the overlay network
is not encrypted. This could potentially expose traffic between the container nodes.
Audit:
Run the below command and ensure that each overlay network has been encrypted.
docker network ls --filter driver=overlay --quiet | xargs docker network

inspect --format '{{.Name}} {{ .Options }}'
Remediation:
Create overlay network with --opt encrypted flag.
Impact:
None
Default Value:
By default, data exchanged between containers on different nodes on the overlay network
are not encrypted in the Docker swarm mode.
References:
1. https://docs.docker.com/engine/userguide/networking/overlay-security-model/
CIS Controls:

210 | P a g e
211 | P a g e
7.5 Ensure Docker's secret management commands are used for
managing secrets in a Swarm cluster (Not Scored)
Description:
Use Docker's in-built secret management command.
Rationale:
Docker has various commands for managing secrets in a Swarm cluster. This is the
foundation for future secret support in Docker with potential improvements such as
Windows support, different backing stores, etc.
Audit:
On a swarm manager node, run the below command and ensure docker secret
management is used in your environment, if applicable.
docker secret ls
Remediation:
Follow docker secret documentation and use it to manage secrets effectively.
Impact:
None
Default Value:
Not Applicable
References:
1. https://docs.docker.com/engine/reference/commandline/secret/
CIS Controls:

212 | P a g e
7.6 Ensure swarm manager is run in auto-lock mode (Scored)
Description:
Run Docker swarm manager in auto-lock mode.
Rationale:
When Docker restarts, both the TLS key used to encrypt communication among swarm
nodes, and the key used to encrypt and decrypt Raft logs on disk, are loaded into each
manager node's memory. You should protect the mutual TLS encryption key and the key
used to encrypt and decrypt Raft logs at rest. This protection could be enabled by
initializing swarm with --autolock flag.
With --autolockenabled, when Docker restarts, you must unlock the swarm first, using a
key encryption key generated by Docker when the swarm was initialized.
Audit:
Run the below command. If it outputs the key, it means swarm was initialized with the --
autolock flag. If the output is no unlock key is set, it means that swarm was NOT
initialized with the --autolock flag and is non-compliant with respect to this
recommendation.
docker swarm unlock-key
Remediation:
If you are initializing swarm, use the below command.
docker swarm init --autolock
If you want to set --autolock on an existing swarm manager node, use the below
command.
docker swarm update --autolock
213 | P a g e
Impact:
A swarm in auto-lock mode won't recover from a re-start without manual intervention
from a user to enter the unlock key. In some deployments, this might not be good for
availability.
Default Value:
By default, swarm manager does not run in auto-lock mode.
References:
1. https://docs.docker.com/engine/swarm/swarm_manager_locking/
CIS Controls:

214 | P a g e
7.7 Ensure swarm manager auto-lock key is rotated periodically (Not
Scored)
Description:
Rotate swarm manager auto-lock key periodically.
Rationale:
Swarm manager auto-lock key is not automatically rotated. You should rotate them
periodically as a best practice.
Audit:
Currently, there is no mechanism to find out when the key was last rotated on a swarm
manager node. You should check with the system administrator if there is a key rotation
record and the keys were rotated at a pre-defined frequency.
Remediation:
Run the below command to rotate the keys.
docker swarm unlock-key --rotate
Additionally, to facilitate audit for this recommendation, maintain key rotation records and
ensure that you establish a pre-defined frequency for key rotation.
Impact:
None
Default Value:
By default, keys are not rotated automatically.
References:
1. https://docs.docker.com/engine/reference/commandline/swarm_unlock-key/
215 | P a g e
CIS Controls:

216 | P a g e
7.8 Ensure node certificates are rotated as appropriate (Not Scored)
Description:
Rotate swarm node certificates as appropriate.
Rationale:
Docker Swarm uses mutual TLS for clustering operations amongst its nodes. Certificate
rotation ensures that in an event such as compromised node or key, it is difficult to
impersonate a node. By default, node certificates are rotated every 90 days. You should
rotate it more often or as appropriate in your environment.
Audit:
Run the below command and ensure that the node certificate Expiry Duration is set as
appropriate.
docker info | grep "Expiry Duration"
Remediation:
Run the below command to set the desired expiry time.
For example,
docker swarm update --cert-expiry 48h
Impact:
None
Default Value:
By default, node certificates are rotated automatically every 90 days.
References:
1. https://docs.docker.com/engine/reference/commandline/swarm_update/#exampl
es
217 | P a g e
CIS Controls:

218 | P a g e
7.9 Ensure CA certificates are rotated as appropriate (Not Scored)
Description:
Rotate root CA certificates as appropriate.
Rationale:
Docker Swarm uses mutual TLS for clustering operations amongst its nodes. Certificate
rotation ensures that in an event such as compromised node or key, it is difficult to
impersonate a node. Node certificates depend upon root CA certificates. For operational
security, it is important to rotate these frequently. Currently, root CA certificates are not
rotated automatically. You should thus establish a process to rotate it at the desired
frequency.
Audit:
Based on your installation path, check the time stamp on the root CA certificate file.
For example,
ls -l /var/lib/docker/swarm/certificates/swarm-root-ca.crt
The certificate should have been rotated at the established frequency.
Remediation:
Run the below command to rotate the certificate.
docker swarm ca --rotate
Impact:
None
Default Value:
By default, root CA certificates are not rotated.
219 | P a g e
References:
1. https://docs.docker.com/engine/swarm/how-swarm-mode-works/pki/#rotating-
the-ca-certificate
CIS Controls:

220 | P a g e
7.10 Ensure management plane traffic has been separated from data
plane traffic (Not Scored)
Description:
Separate management plane traffic from data plane traffic.
Rationale:
Separating the management plane traffic from data plane traffic ensures that these traffics
are on their respective paths. These paths could then be individually monitored and could
be tied to different traffic control policies and monitoring. It also ensures that management
plane is always reachable despite the huge volume of data flow.
Audit:
Run the below command on each swarm node and ensure that the management plane
address is different from data plane address.
docker node inspect --format '{{ .Status.Addr }}' self
Note: At the time of writing of this benchmark, there is no way to inspect data plane
address. An issue has been raised and is in the reference link.
Remediation:
Initialize Swarm with dedicated interfaces for management and data planes respectively.
For example,
docker swarm init --advertise-addr=192.168.0.1 --data-path-addr=17.1.0.3
Impact:
You would require 2 network interface cards per node.
Default Value:
By default, the data plane traffic is not separated from management plane traffic.
221 | P a g e
References:
1. https://docs.docker.com/engine/reference/commandline/swarm_init/#--data-
path-addr
2. https://github.com/moby/moby/issues/33938
CIS Controls:

222 | P a g e
Appendix: Summary Table
Control Set
Correctly
Yes No
1 Host Configuration
1.1 Ensure a separate partition for containers has been created
 
(Scored)
1.2 Ensure the container host has been Hardened (Not Scored)  
1.3 Ensure Docker is up to date (Not Scored)  
1.4 Ensure only trusted users are allowed to control Docker
 
daemon (Scored)
1.5 Ensure auditing is configured for the docker daemon
 
(Scored)
1.6 Ensure auditing is configured for Docker files and
 
directories - /var/lib/docker (Scored)
 
directories - /etc/docker (Scored)
 
directories - docker.service (Scored)
 
directories - docker.socket (Scored)
 
directories - /etc/default/docker (Scored)
 
directories - /etc/docker/daemon.json (Scored)
 
directories - /usr/bin/docker-containerd (Scored)
 
directories - /usr/bin/docker-runc (Scored)
2 Docker daemon configuration
2.1 Ensure network traffic is restricted between containers on
 
the default bridge (Scored)
2.2 Ensure the logging level is set to 'info' (Scored)  
2.3 Ensure Docker is allowed to make changes to iptables
 
(Scored)
2.4 Ensure insecure registries are not used (Scored)  
2.5 Ensure aufs storage driver is not used (Scored)  
2.6 Ensure TLS authentication for Docker daemon is configured
 
(Scored)
2.7 Ensure the default ulimit is configured appropriately (Not
 
Scored)
2.8 Enable user namespace support (Scored)  
223 | P a g e
2.9 Ensure the default cgroup usage has been confirmed
 
(Scored)
2.10 Ensure base device size is not changed until needed
 
(Scored)
2.11 Ensure that authorization for Docker client commands is
 
enabled (Scored)
2.12 Ensure centralized and remote logging is configured
 
(Scored)
2.13 Ensure operations on legacy registry (v1) are Disabled
 
(Scored)
2.14 Ensure live restore is Enabled (Scored)  
2.15 Ensure Userland Proxy is Disabled (Scored)  
2.16 Ensure daemon-wide custom seccomp profile is applied, if
 
needed (Not Scored)
2.17 Ensure experimental features are avoided in production
 
(Scored)
2.18 Ensure containers are restricted from acquiring new
 
privileges (Scored)
3 Docker daemon configuration files
3.1 Ensure that docker.service file ownership is set to root:root
 
(Scored)
3.2 Ensure that docker.service file permissions are set to 644 or
 
3.3 Ensure that docker.socket file ownership is set to root:root
 
(Scored)
3.4 Ensure that docker.socket file permissions are set to 644 or
 
3.5 Ensure that /etc/docker directory ownership is set to
 
root:root (Scored)
3.6 Ensure that /etc/docker directory permissions are set to
 
755 or more restrictive (Scored)
3.7 Ensure that registry certificate file ownership is set to
 
root:root (Scored)
3.8 Ensure that registry certificate file permissions are set to
 
3.9 Ensure that TLS CA certificate file ownership is set to
 
root:root (Scored)
3.10 Ensure that TLS CA certificate file permissions are set to 444
 
or more restrictive (Scored)
3.11 Ensure that Docker server certificate file ownership is set to
 
root:root (Scored)
3.12 Ensure that Docker server certificate file permissions are set
 
to 444 or more restrictive (Scored)
224 | P a g e
3.13 Ensure that Docker server certificate key file ownership is
 
set to root:root (Scored)
3.14 Ensure that Docker server certificate key file permissions
 
are set to 400 (Scored)
3.15 Ensure that Docker socket file ownership is set to
 
root:docker (Scored)
3.16 Ensure that Docker socket file permissions are set to 660 or
 
3.17 Ensure that daemon.json file ownership is set to root:root
 
(Scored)
3.18 Ensure that daemon.json file permissions are set to 644 or
 
3.19 Ensure that /etc/default/docker file ownership is set to
 
root:root (Scored)
3.20 Ensure that /etc/default/docker file permissions are set to
 
4 Container Images and Build File
4.1 Ensure a user for the container has been created (Scored)  
4.2 Ensure that containers use trusted base images (Not Scored)  
4.3 Ensure unnecessary packages are not installed in the
 
container (Not Scored)
4.4 Ensure images are scanned and rebuilt to include security
 
patches (Not Scored)
4.5 Ensure Content trust for Docker is Enabled (Scored)  
4.6 Ensure HEALTHCHECK instructions have been added to the
 
container image (Scored)
4.7 Ensure update instructions are not use alone in the
 
Dockerfile (Not Scored)
4.8 Ensure setuid and setgid permissions are removed in the
 
images (Not Scored)
4.9 Ensure COPY is used instead of ADD in Dockerfile (Not
 
Scored)
4.10 Ensure secrets are not stored in Dockerfiles (Not Scored)  
4.11 Ensure verified packages are only Installed (Not Scored)  
5 Container Runtime
5.1 Ensure AppArmor Profile is Enabled (Scored)  
5.2 Ensure SELinux security options are set, if applicable
 
(Scored)
5.3 Ensure Linux Kernel Capabilities are restricted within
 
containers (Scored)
5.4 Ensure privileged containers are not used (Scored)  
5.5 Ensure sensitive host system directories are not mounted
 
on containers (Scored)
5.6 Ensure ssh is not run within containers (Scored)  
225 | P a g e
5.7 Ensure privileged ports are not mapped within containers
 
(Scored)
5.8 Ensure only needed ports are open on the container
 
(Scored)
5.9 Ensure the host's network namespace is not shared
 
(Scored)
5.10 Ensure memory usage for container is limited (Scored)  
5.11 Ensure CPU priority is set appropriately on the container
 
(Scored)
5.12 Ensure the container's root filesystem is mounted as read
 
only (Scored)
5.13 Ensure incoming container traffic is binded to a specific host
 
interface (Scored)
5.14 Ensure 'on-failure' container restart policy is set to '5'
 
(Scored)
5.15 Ensure the host's process namespace is not shared (Scored)  
5.16 Ensure the host's IPC namespace is not shared (Scored)  
5.17 Ensure host devices are not directly exposed to containers
 
(Not Scored)
5.18 Ensure the default ulimit is overwritten at runtime, only if
 
needed (Not Scored)
5.19 Ensure mount propagation mode is not set to shared
 
(Scored)
5.20 Ensure the host's UTS namespace is not shared (Scored)  
5.21 Ensure the default seccomp profile is not Disabled (Scored)  
5.22 Ensure docker exec commands are not used with privileged
 
option (Scored)
5.23 Ensure docker exec commands are not used with user
 
option (Scored)
5.24 Ensure cgroup usage is confirmed (Scored)  
5.25 Ensure the container is restricted from acquiring additional
 
privileges (Scored)
5.26 Ensure container health is checked at runtime (Scored)  
5.27 Ensure docker commands always get the latest version of
 
the image (Not Scored)
5.28 Ensure PIDs cgroup limit is used (Scored)  
5.29 Ensure Docker's default bridge docker0 is not used (Not
 
Scored)
5.30 Ensure the host's user namespaces is not shared (Scored)  
5.31 Ensure the Docker socket is not mounted inside any
 
containers (Scored)
6 Docker Security Operations
6.1 Ensure image sprawl is avoided (Not Scored)  
226 | P a g e
6.2 Ensure container sprawl is avoided (Not Scored)  
7 Docker Swarm Configuration
7.1 Ensure swarm mode is not Enabled, if not needed (Scored)  
7.2 Ensure the minimum number of manager nodes have been
 
created in a swarm (Scored)
7.3 Ensure swarm services are binded to a specific host
 
interface (Scored)
7.4 Ensure data exchanged between containers are encrypted
 
on different nodes on the overlay network (Scored)
7.5 Ensure Docker's secret management commands are used for
 
managing secrets in a Swarm cluster (Not Scored)
7.6 Ensure swarm manager is run in auto-lock mode (Scored)  
7.7 Ensure swarm manager auto-lock key is rotated periodically
 
(Not Scored)
7.8 Ensure node certificates are rotated as appropriate (Not
 
Scored)
7.9 Ensure CA certificates are rotated as appropriate (Not
 
Scored)
7.10 Ensure management plane traffic has been separated from
 
data plane traffic (Not Scored)
227 | P a g e
Appendix: Change History
Date Version Changes for this version
01-19-17 1.0.0 Initial Release
07-06-17 1.1.0 NEW - Ensure containers are

restricted from acquiring new
privileges. Ticket #5222
07-06-17 1.1.0 NEW - Ensure node certificates are

rotated as appropriate. Ticket #5227
07-06-17 1.1.0 NEW - Ensure CA certificates are

rotated as appropriate. Ticket #5228
07-06-17 1.1.0 NEW - Ensure management plane

traffic has been separated from data
plane traffic. Ticket # 5229
07-06-17 1.1.0 MODIFY - clarified the intent of the

recommendation - Ensure network
traffic is restricted between
containers on the default bridge.
Ticket # 5210
07-06-17 1.1.0 DELETED – 6.1 - Perform regular

security audits of your host system
and containers. Ticket # 5223
07-06-17 1.1.0 DELETED – 6.2 - Monitor Docker

containers usage, performance and
metering. Ticket # 5224
07-06-17 1.1.0 DELETED – 6.3 -Backup container

data. Ticket # 5225
07-06-17 1.1.0 UPDATES - Updated several reference

URLs
228 | P a g e
07-06-17 1.1.0 NEW - New Section - "7 Docker
Swarm Configuration"
07-06-17 1.1.0 NEW – CIS CONTROLS Mappings
07-06-17 1.1.0 UPDATES – Updated all

recommendation titles to conform to
CIS standard.
07-06-17 1.1.0 UPDATES – Cleaned up formatting of

benchmark
229 | P a g e

CIS Docker Community Edition Benchmark v1.1.0

Uploaded by

Copyright:

Available Formats

CIS Docker Community Edition Benchmark v1.1.0

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CIS Docker Community Edition Benchmark v1.1.0

Uploaded by

Copyright:

Available Formats

CIS Docker Community Edition Benchmark

Italic font Used to denote the title of a book, article, or other

Note Additional information or caveats

Items in this profile intend to:

o Be practical and prudent;

 Level 1 - Linux Host OS

o Be practical and prudent;

Items in this profile exhibit one or more of the following characteristics:

o Are intended for environments or use cases where security is paramount

 Level 1 - Linux Host OS

At the Docker host execute the below command:

grep /var/lib/docker /etc/fstab

By default, /var/lib/docker would be mounted under / or /var partitions based on

14 Controlled Access Based on the Need to Know

 Level 1 - Linux Host OS

By default, host has factory settings. It is not hardened.

3 Secure Configurations for Hardware and Software on Mobile Devices, Laptops,

 Level 1 - Linux Host OS

By staying up to date on Docker updates, vulnerabilities in the Docker software can be

Keep a track of Docker releases and update as necessary.

4 Continuous Vulnerability Assessment and Remediation

 Level 1 - Linux Host OS

getent group docker

Rights to build and execute containers as normal user would be restricted.

5.1 Minimize And Sparingly Use Administrative Privileges

 Level 1 - Linux Host OS

Audit all Docker daemon activities.

auditctl -l | grep /usr/bin/docker

This should list a rule for Docker daemon.

Add a rule for Docker daemon.

Add the line as below line in /etc/audit/audit.rules file:

Then, restart the audit daemon. For example,

service auditd restart

By default, Docker daemon is not audited.

 Level 1 - Linux Host OS

Verify that there is an audit rule corresponding to /var/lib/docker directory.

For example, execute below command:

auditctl -l | grep /var/lib/docker

This should list a rule for /var/lib/docker directory.

Add a rule for /var/lib/docker directory.

Add the line as below in /etc/audit/audit.rules file:

Then, restart the audit daemon. For example,

service auditd restart

By default, Docker related files and directories are not audited.

14.6 Enforce Detailed Audit Logging For Sensitive Information

 Level 1 - Linux Host OS

Verify that there is an audit rule corresponding to /etc/docker directory.

For example, execute below command:

auditctl -l | grep /etc/docker

This should list a rule for /etc/docker directory.

Add a rule for /etc/docker directory.

Add the line as below in /etc/audit/audit.rules file:

Then, restart the audit daemon. For example,

service auditd restart

By default, Docker related files and directories are not audited.

14.6 Enforce Detailed Audit Logging For Sensitive Information

 Level 1 - Linux Host OS

Audit docker.service, if applicable.

Step 1: Find out the file location: