summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--.dockerignore4
-rw-r--r--BUILD.md29
-rw-r--r--README_CONTAINER_IMAGE.md6
-rw-r--r--images/installer/Dockerfile62
-rw-r--r--images/installer/Dockerfile.rhel767
-rw-r--r--images/installer/README_CONTAINER_IMAGE.md (renamed from images/installer/system-container/README.md)27
-rw-r--r--images/installer/root/exports/config.json.template (renamed from images/installer/system-container/root/exports/config.json.template)0
-rw-r--r--images/installer/root/exports/manifest.json (renamed from images/installer/system-container/root/exports/manifest.json)0
-rw-r--r--images/installer/root/exports/service.template (renamed from images/installer/system-container/root/exports/service.template)0
-rw-r--r--images/installer/root/exports/tmpfiles.template (renamed from images/installer/system-container/root/exports/tmpfiles.template)0
-rwxr-xr-ximages/installer/root/usr/local/bin/entrypoint17
-rwxr-xr-ximages/installer/root/usr/local/bin/run46
-rwxr-xr-ximages/installer/root/usr/local/bin/run-system-container.sh (renamed from images/installer/system-container/root/usr/local/bin/run-system-container.sh)0
-rwxr-xr-ximages/installer/root/usr/local/bin/usage33
-rwxr-xr-ximages/installer/root/usr/local/bin/usage.ocp33
-rwxr-xr-ximages/installer/root/usr/local/bin/user_setup17
-rw-r--r--inventory/byo/hosts.origin.example5
-rw-r--r--inventory/byo/hosts.ose.example5
-rw-r--r--playbooks/byo/openshift-checks/README.md11
-rw-r--r--playbooks/common/openshift-cluster/openshift_hosted.yml4
-rw-r--r--roles/ansible_service_broker/defaults/main.yml1
-rw-r--r--roles/ansible_service_broker/tasks/install.yml9
-rw-r--r--roles/docker/tasks/package_docker.yml2
-rw-r--r--roles/openshift_default_storage_class/defaults/main.yml1
-rw-r--r--roles/openshift_default_storage_class/tasks/main.yml3
-rw-r--r--roles/openshift_health_checker/library/search_journalctl.py150
-rw-r--r--roles/openshift_health_checker/openshift_checks/docker_storage.py177
-rw-r--r--roles/openshift_health_checker/openshift_checks/etcd_traffic.py47
-rw-r--r--roles/openshift_health_checker/test/docker_storage_test.py107
-rw-r--r--roles/openshift_health_checker/test/etcd_traffic_test.py80
-rw-r--r--roles/openshift_health_checker/test/search_journalctl_test.py157
-rw-r--r--roles/openshift_service_catalog/files/kubeservicecatalog_roles_bindings.yml6
32 files changed, 942 insertions, 164 deletions
diff --git a/.dockerignore b/.dockerignore
index 968811df5..0a70c5bfa 100644
--- a/.dockerignore
+++ b/.dockerignore
@@ -1,8 +1,12 @@
.*
bin
docs
+hack
+inventory
test
utils
**/*.md
*.spec
+*.ini
+*.txt
setup*
diff --git a/BUILD.md b/BUILD.md
index d9de3f5aa..1c270db23 100644
--- a/BUILD.md
+++ b/BUILD.md
@@ -33,35 +33,6 @@ To build a container image of `openshift-ansible` using standalone **Docker**:
cd openshift-ansible
docker build -f images/installer/Dockerfile -t openshift-ansible .
-### Building on OpenShift
-
-To build an openshift-ansible image using an **OpenShift** [build and image stream](https://docs.openshift.org/latest/architecture/core_concepts/builds_and_image_streams.html) the straightforward command would be:
-
- oc new-build registry.centos.org/openshift/playbook2image~https://github.com/openshift/openshift-ansible
-
-However: because the `Dockerfile` for this repository is not in the top level directory, and because we can't change the build context to the `images/installer` path as it would cause the build to fail, the `oc new-app` command above will create a build configuration using the *source to image* strategy, which is the default approach of the [playbook2image](https://github.com/openshift/playbook2image) base image. This does build an image successfully, but unfortunately the resulting image will be missing some customizations that are handled by the [Dockerfile](images/installer/Dockerfile) in this repo.
-
-At the time of this writing there is no straightforward option to [set the dockerfilePath](https://docs.openshift.org/latest/dev_guide/builds/build_strategies.html#dockerfile-path) of a `docker` build strategy with `oc new-build`. The alternatives to achieve this are:
-
-- Use the simple `oc new-build` command above to generate the BuildConfig and ImageStream objects, and then manually edit the generated build configuration to change its strategy to `dockerStrategy` and set `dockerfilePath` to `images/installer/Dockerfile`.
-
-- Download and pass the `Dockerfile` to `oc new-build` with the `-D` option:
-
-```
-curl -s https://raw.githubusercontent.com/openshift/openshift-ansible/master/images/installer/Dockerfile |
- oc new-build -D - \
- --docker-image=registry.centos.org/openshift/playbook2image \
- https://github.com/openshift/openshift-ansible
-```
-
-Once a build is started, the progress of the build can be monitored with:
-
- oc logs -f bc/openshift-ansible
-
-Once built, the image will be visible in the Image Stream created by `oc new-app`:
-
- oc describe imagestream openshift-ansible
-
## Build the Atomic System Container
A system container runs using runC instead of Docker and it is managed
diff --git a/README_CONTAINER_IMAGE.md b/README_CONTAINER_IMAGE.md
index cf3b432df..a2151352d 100644
--- a/README_CONTAINER_IMAGE.md
+++ b/README_CONTAINER_IMAGE.md
@@ -1,6 +1,6 @@
# Containerized openshift-ansible to run playbooks
-The [Dockerfile](images/installer/Dockerfile) in this repository uses the [playbook2image](https://github.com/openshift/playbook2image) source-to-image base image to containerize `openshift-ansible`. The resulting image can run any of the provided playbooks. See [BUILD.md](BUILD.md) for image build instructions.
+The [Dockerfile](images/installer/Dockerfile) in this repository can be used to build a containerized `openshift-ansible`. The resulting image can run any of the provided playbooks. See [BUILD.md](BUILD.md) for image build instructions.
The image is designed to **run as a non-root user**. The container's UID is mapped to the username `default` at runtime. Therefore, the container's environment reflects that user's settings, and the configuration should match that. For example `$HOME` is `/opt/app-root/src`, so ssh keys are expected to be under `/opt/app-root/src/.ssh`. If you ran a container as `root` you would have to adjust the container's configuration accordingly, e.g. by placing ssh keys under `/root/.ssh` instead. Nevertheless, the expectation is that containers will be run as non-root; for example, this container image can be run inside OpenShift under the default `restricted` [security context constraint](https://docs.openshift.org/latest/architecture/additional_concepts/authorization.html#security-context-constraints).
@@ -14,8 +14,6 @@ This provides consistency with other images used by the platform and it's also a
## Usage
-The `playbook2image` base image provides several options to control the behaviour of the containers. For more details on these options see the [playbook2image](https://github.com/openshift/playbook2image) documentation.
-
At the very least, when running a container you must specify:
1. An **inventory**. This can be a location inside the container (possibly mounted as a volume) with a path referenced via the `INVENTORY_FILE` environment variable. Alternatively you can serve the inventory file from a web server and use the `INVENTORY_URL` environment variable to fetch it, or `DYNAMIC_SCRIPT_URL` to download a script that provides a dynamic inventory.
@@ -52,8 +50,6 @@ Here is a detailed explanation of the options used in the command above:
Further usage examples are available in the [examples directory](examples/) with samples of how to use the image from within OpenShift.
-Additional usage information for images built from `playbook2image` like this one can be found in the [playbook2image examples](https://github.com/openshift/playbook2image/tree/master/examples).
-
## Running openshift-ansible as a System Container
Building the System Container: See the [BUILD.md](BUILD.md).
diff --git a/images/installer/Dockerfile b/images/installer/Dockerfile
index 915dfe377..d03f33a1d 100644
--- a/images/installer/Dockerfile
+++ b/images/installer/Dockerfile
@@ -1,10 +1,18 @@
-# Using playbook2image as a base
-# See https://github.com/openshift/playbook2image for details on the image
-# including documentation for the settings/env vars referenced below
-FROM registry.centos.org/openshift/playbook2image:latest
+FROM centos:7
MAINTAINER OpenShift Team <dev@lists.openshift.redhat.com>
+USER root
+
+# install ansible and deps
+RUN INSTALL_PKGS="python-lxml pyOpenSSL python2-cryptography openssl java-1.8.0-openjdk-headless httpd-tools openssh-clients" \
+ && yum install -y --setopt=tsflags=nodocs $INSTALL_PKGS \
+ && EPEL_PKGS="ansible python-passlib python2-boto" \
+ && yum install -y epel-release \
+ && yum install -y --setopt=tsflags=nodocs $EPEL_PKGS \
+ && rpm -q $INSTALL_PKGS $EPEL_PKGS \
+ && yum clean all
+
LABEL name="openshift/origin-ansible" \
summary="OpenShift's installation and configuration tool" \
description="A containerized openshift-ansible image to let you run playbooks to install, upgrade, maintain and check an OpenShift cluster" \
@@ -12,40 +20,24 @@ LABEL name="openshift/origin-ansible" \
io.k8s.display-name="openshift-ansible" \
io.k8s.description="A containerized openshift-ansible image to let you run playbooks to install, upgrade, maintain and check an OpenShift cluster" \
io.openshift.expose-services="" \
- io.openshift.tags="openshift,install,upgrade,ansible"
+ io.openshift.tags="openshift,install,upgrade,ansible" \
+ atomic.run="once"
-USER root
+ENV USER_UID=1001 \
+ HOME=/opt/app-root/src \
+ WORK_DIR=/usr/share/ansible/openshift-ansible \
+ OPTS="-v"
-# Create a symlink to /opt/app-root/src so that files under /usr/share/ansible are accessible.
-# This is required since the system-container uses by default the playbook under
-# /usr/share/ansible/openshift-ansible. With this change we won't need to keep two different
-# configurations for the two images.
-RUN mkdir -p /usr/share/ansible/ && ln -s /opt/app-root/src /usr/share/ansible/openshift-ansible
+# Add image scripts and files for running as a system container
+COPY images/installer/root /
+# Include playbooks, roles, plugins, etc. from this repo
+COPY . ${WORK_DIR}
-RUN INSTALL_PKGS="skopeo openssl java-1.8.0-openjdk-headless httpd-tools" && \
- yum install -y --setopt=tsflags=nodocs $INSTALL_PKGS && \
- rpm -V $INSTALL_PKGS && \
- yum clean all
+RUN /usr/local/bin/user_setup \
+ && rm /usr/local/bin/usage.ocp
USER ${USER_UID}
-# The playbook to be run is specified via the PLAYBOOK_FILE env var.
-# This sets a default of openshift_facts.yml as it's an informative playbook
-# that can help test that everything is set properly (inventory, sshkeys)
-ENV PLAYBOOK_FILE=playbooks/byo/openshift_facts.yml \
- OPTS="-v" \
- INSTALL_OC=true
-
-# playbook2image's assemble script expects the source to be available in
-# /tmp/src (as per the source-to-image specs) so we import it there
-ADD . /tmp/src
-
-# Running the 'assemble' script provided by playbook2image will install
-# dependencies specified in requirements.txt and install the 'oc' client
-# as per the INSTALL_OC environment setting above
-RUN /usr/libexec/s2i/assemble
-
-# Add files for running as a system container
-COPY images/installer/system-container/root /
-
-CMD [ "/usr/libexec/s2i/run" ]
+WORKDIR ${WORK_DIR}
+ENTRYPOINT [ "/usr/local/bin/entrypoint" ]
+CMD [ "/usr/local/bin/run" ]
diff --git a/images/installer/Dockerfile.rhel7 b/images/installer/Dockerfile.rhel7
index f861d4bcf..3110f409c 100644
--- a/images/installer/Dockerfile.rhel7
+++ b/images/installer/Dockerfile.rhel7
@@ -1,55 +1,46 @@
-FROM openshift3/playbook2image
+FROM rhel7.3:7.3-released
MAINTAINER OpenShift Team <dev@lists.openshift.redhat.com>
-# override env vars from base image
-ENV SUMMARY="OpenShift's installation and configuration tool" \
- DESCRIPTION="A containerized openshift-ansible image to let you run playbooks to install, upgrade, maintain and check an OpenShift cluster"
+USER root
+
+# Playbooks, roles, and their dependencies are installed from packages.
+RUN INSTALL_PKGS="atomic-openshift-utils atomic-openshift-clients python-boto openssl java-1.8.0-openjdk-headless httpd-tools" \
+ && yum repolist > /dev/null \
+ && yum-config-manager --enable rhel-7-server-ose-3.6-rpms \
+ && yum-config-manager --enable rhel-7-server-rh-common-rpms \
+ && yum install -y --setopt=tsflags=nodocs $INSTALL_PKGS \
+ && rpm -q $INSTALL_PKGS \
+ && yum clean all
LABEL name="openshift3/ose-ansible" \
- summary="$SUMMARY" \
- description="$DESCRIPTION" \
+ summary="OpenShift's installation and configuration tool" \
+ description="A containerized openshift-ansible image to let you run playbooks to install, upgrade, maintain and check an OpenShift cluster" \
url="https://github.com/openshift/openshift-ansible" \
io.k8s.display-name="openshift-ansible" \
- io.k8s.description="$DESCRIPTION" \
+ io.k8s.description="A containerized openshift-ansible image to let you run playbooks to install, upgrade, maintain and check an OpenShift cluster" \
io.openshift.expose-services="" \
io.openshift.tags="openshift,install,upgrade,ansible" \
com.redhat.component="aos3-installation-docker" \
version="v3.6.0" \
release="1" \
- architecture="x86_64"
-
-# Playbooks, roles and their dependencies are installed from packages.
-# Unlike in Dockerfile, we don't invoke the 'assemble' script here
-# because all content and dependencies (like 'oc') is already
-# installed via yum.
-USER root
-RUN INSTALL_PKGS="atomic-openshift-utils atomic-openshift-clients python-boto skopeo openssl java-1.8.0-openjdk-headless httpd-tools" && \
- yum repolist > /dev/null && \
- yum-config-manager --enable rhel-7-server-ose-3.6-rpms && \
- yum-config-manager --enable rhel-7-server-rh-common-rpms && \
- yum install -y $INSTALL_PKGS && \
- yum clean all
-
-# The symlinks below are a (hopefully temporary) hack to work around the fact that this
-# image is based on python s2i which uses the python27 SCL instead of system python,
-# and so the system python modules we need would otherwise not be in the path.
-RUN ln -s /usr/lib/python2.7/site-packages/{boto,passlib} /opt/app-root/lib64/python2.7/
-
-USER ${USER_UID}
+ architecture="x86_64" \
+ atomic.run="once"
-# The playbook to be run is specified via the PLAYBOOK_FILE env var.
-# This sets a default of openshift_facts.yml as it's an informative playbook
-# that can help test that everything is set properly (inventory, sshkeys).
-# As the playbooks are installed via packages instead of being copied to
-# $APP_HOME by the 'assemble' script, we set the WORK_DIR env var to the
-# location of openshift-ansible.
-ENV PLAYBOOK_FILE=playbooks/byo/openshift_facts.yml \
- ANSIBLE_CONFIG=/usr/share/atomic-openshift-utils/ansible.cfg \
+ENV USER_UID=1001 \
+ HOME=/opt/app-root/src \
WORK_DIR=/usr/share/ansible/openshift-ansible \
+ ANSIBLE_CONFIG=/usr/share/atomic-openshift-utils/ansible.cfg \
OPTS="-v"
-# Add files for running as a system container
-COPY system-container/root /
+# Add image scripts and files for running as a system container
+COPY root /
+
+RUN /usr/local/bin/user_setup \
+ && mv /usr/local/bin/usage{.ocp,}
+
+USER ${USER_UID}
-CMD [ "/usr/libexec/s2i/run" ]
+WORKDIR ${WORK_DIR}
+ENTRYPOINT [ "/usr/local/bin/entrypoint" ]
+CMD [ "/usr/local/bin/run" ]
diff --git a/images/installer/system-container/README.md b/images/installer/README_CONTAINER_IMAGE.md
index fbcd47c4a..bc1ebb4a8 100644
--- a/images/installer/system-container/README.md
+++ b/images/installer/README_CONTAINER_IMAGE.md
@@ -1,17 +1,34 @@
-# System container installer
+ORIGIN-ANSIBLE IMAGE INSTALLER
+===============================
+
+Contains Dockerfile information for building an openshift/origin-ansible image
+based on `centos:7` or `rhel7.3:7.3-released`.
+
+Read additional setup information for this image at: https://hub.docker.com/r/openshift/origin-ansible/
+
+Read additional information about the `openshift/origin-ansible` at: https://github.com/openshift/openshift-ansible/blob/master/README_CONTAINER_IMAGE.md
+
+Also contains necessary components for running the installer using an Atomic System Container.
+
+
+System container installer
+==========================
These files are needed to run the installer using an [Atomic System container](http://www.projectatomic.io/blog/2016/09/intro-to-system-containers/).
+These files can be found under `root/exports`:
* config.json.template - Template of the configuration file used for running containers.
-* manifest.json - Used to define various settings for the system container, such as the default values to use for the installation.
-
-* run-system-container.sh - Entrypoint to the container.
+* manifest.json - Used to define various settings for the system container, such as the default values to use for the installation.
* service.template - Template file for the systemd service.
* tmpfiles.template - Template file for systemd-tmpfiles.
+These files can be found under `root/usr/local/bin`:
+
+* run-system-container.sh - Entrypoint to the container.
+
## Options
These options may be set via the ``atomic`` ``--set`` flag. For defaults see ``root/exports/manifest.json``
@@ -28,4 +45,4 @@ These options may be set via the ``atomic`` ``--set`` flag. For defaults see ``r
* ANSIBLE_CONFIG - Full path for the ansible configuration file to use inside the container
-* INVENTORY_FILE - Full path for the inventory to use from the host
+* INVENTORY_FILE - Full path for the inventory to use from the host \ No newline at end of file
diff --git a/images/installer/system-container/root/exports/config.json.template b/images/installer/root/exports/config.json.template
index 739c0080f..739c0080f 100644
--- a/images/installer/system-container/root/exports/config.json.template
+++ b/images/installer/root/exports/config.json.template
diff --git a/images/installer/system-container/root/exports/manifest.json b/images/installer/root/exports/manifest.json
index 8b984d7a3..8b984d7a3 100644
--- a/images/installer/system-container/root/exports/manifest.json
+++ b/images/installer/root/exports/manifest.json
diff --git a/images/installer/system-container/root/exports/service.template b/images/installer/root/exports/service.template
index bf5316af6..bf5316af6 100644
--- a/images/installer/system-container/root/exports/service.template
+++ b/images/installer/root/exports/service.template
diff --git a/images/installer/system-container/root/exports/tmpfiles.template b/images/installer/root/exports/tmpfiles.template
index b1f6caf47..b1f6caf47 100644
--- a/images/installer/system-container/root/exports/tmpfiles.template
+++ b/images/installer/root/exports/tmpfiles.template
diff --git a/images/installer/root/usr/local/bin/entrypoint b/images/installer/root/usr/local/bin/entrypoint
new file mode 100755
index 000000000..777bf3f11
--- /dev/null
+++ b/images/installer/root/usr/local/bin/entrypoint
@@ -0,0 +1,17 @@
+#!/bin/bash -e
+#
+# This file serves as the main entrypoint to the openshift-ansible image.
+#
+# For more information see the documentation:
+# https://github.com/openshift/openshift-ansible/blob/master/README_CONTAINER_IMAGE.md
+
+
+# Patch /etc/passwd file with the current user info.
+# The current user's entry must be correctly defined in this file in order for
+# the `ssh` command to work within the created container.
+
+if ! whoami &>/dev/null; then
+ echo "${USER:-default}:x:$(id -u):$(id -g):Default User:$HOME:/sbin/nologin" >> /etc/passwd
+fi
+
+exec "$@"
diff --git a/images/installer/root/usr/local/bin/run b/images/installer/root/usr/local/bin/run
new file mode 100755
index 000000000..9401ea118
--- /dev/null
+++ b/images/installer/root/usr/local/bin/run
@@ -0,0 +1,46 @@
+#!/bin/bash -e
+#
+# This file serves as the default command to the openshift-ansible image.
+# Runs a playbook with inventory as specified by environment variables.
+#
+# For more information see the documentation:
+# https://github.com/openshift/openshift-ansible/blob/master/README_CONTAINER_IMAGE.md
+
+# SOURCE and HOME DIRECTORY: /opt/app-root/src
+
+if [[ -z "${PLAYBOOK_FILE}" ]]; then
+ echo
+ echo "PLAYBOOK_FILE must be provided."
+ exec /usr/local/bin/usage
+fi
+
+INVENTORY="$(mktemp)"
+if [[ -v INVENTORY_FILE ]]; then
+ # Make a copy so that ALLOW_ANSIBLE_CONNECTION_LOCAL below
+ # does not attempt to modify the original
+ cp -a ${INVENTORY_FILE} ${INVENTORY}
+elif [[ -v INVENTORY_URL ]]; then
+ curl -o ${INVENTORY} ${INVENTORY_URL}
+elif [[ -v DYNAMIC_SCRIPT_URL ]]; then
+ curl -o ${INVENTORY} ${DYNAMIC_SCRIPT_URL}
+ chmod 755 ${INVENTORY}
+else
+ echo
+ echo "One of INVENTORY_FILE, INVENTORY_URL or DYNAMIC_SCRIPT_URL must be provided."
+ exec /usr/local/bin/usage
+fi
+INVENTORY_ARG="-i ${INVENTORY}"
+
+if [[ "$ALLOW_ANSIBLE_CONNECTION_LOCAL" = false ]]; then
+ sed -i s/ansible_connection=local// ${INVENTORY}
+fi
+
+if [[ -v VAULT_PASS ]]; then
+ VAULT_PASS_FILE=.vaultpass
+ echo ${VAULT_PASS} > ${VAULT_PASS_FILE}
+ VAULT_PASS_ARG="--vault-password-file ${VAULT_PASS_FILE}"
+fi
+
+cd ${WORK_DIR}
+
+exec ansible-playbook ${INVENTORY_ARG} ${VAULT_PASS_ARG} ${OPTS} ${PLAYBOOK_FILE}
diff --git a/images/installer/system-container/root/usr/local/bin/run-system-container.sh b/images/installer/root/usr/local/bin/run-system-container.sh
index 9ce7c7328..9ce7c7328 100755
--- a/images/installer/system-container/root/usr/local/bin/run-system-container.sh
+++ b/images/installer/root/usr/local/bin/run-system-container.sh
diff --git a/images/installer/root/usr/local/bin/usage b/images/installer/root/usr/local/bin/usage
new file mode 100755
index 000000000..3518d7f19
--- /dev/null
+++ b/images/installer/root/usr/local/bin/usage
@@ -0,0 +1,33 @@
+#!/bin/bash -e
+cat <<"EOF"
+
+The origin-ansible image provides several options to control the behaviour of the containers.
+For more details on these options see the documentation:
+
+ https://github.com/openshift/openshift-ansible/blob/master/README_CONTAINER_IMAGE.md
+
+At a minimum, when running a container using this image you must provide:
+
+* ssh keys so that Ansible can reach your hosts. These should be mounted as a volume under
+ /opt/app-root/src/.ssh
+* An inventory file. This can be mounted inside the container as a volume and specified with the
+ INVENTORY_FILE environment variable. Alternatively you can serve the inventory file from a web
+ server and use the INVENTORY_URL environment variable to fetch it.
+* The playbook to run. This is set using the PLAYBOOK_FILE environment variable.
+
+Here is an example of how to run a containerized origin-ansible with
+the openshift_facts playbook, which collects and displays facts about your
+OpenShift environment. The inventory and ssh keys are mounted as volumes
+(the latter requires setting the uid in the container and SELinux label
+in the key file via :Z so they can be accessed) and the PLAYBOOK_FILE
+environment variable is set to point to the playbook within the image:
+
+docker run -tu `id -u` \
+ -v $HOME/.ssh/id_rsa:/opt/app-root/src/.ssh/id_rsa:Z,ro \
+ -v /etc/ansible/hosts:/tmp/inventory:Z,ro \
+ -e INVENTORY_FILE=/tmp/inventory \
+ -e OPTS="-v" \
+ -e PLAYBOOK_FILE=playbooks/byo/openshift_facts.yml \
+ openshift/origin-ansible
+
+EOF
diff --git a/images/installer/root/usr/local/bin/usage.ocp b/images/installer/root/usr/local/bin/usage.ocp
new file mode 100755
index 000000000..50593af6e
--- /dev/null
+++ b/images/installer/root/usr/local/bin/usage.ocp
@@ -0,0 +1,33 @@
+#!/bin/bash -e
+cat <<"EOF"
+
+The ose-ansible image provides several options to control the behaviour of the containers.
+For more details on these options see the documentation:
+
+ https://github.com/openshift/openshift-ansible/blob/master/README_CONTAINER_IMAGE.md
+
+At a minimum, when running a container using this image you must provide:
+
+* ssh keys so that Ansible can reach your hosts. These should be mounted as a volume under
+ /opt/app-root/src/.ssh
+* An inventory file. This can be mounted inside the container as a volume and specified with the
+ INVENTORY_FILE environment variable. Alternatively you can serve the inventory file from a web
+ server and use the INVENTORY_URL environment variable to fetch it.
+* The playbook to run. This is set using the PLAYBOOK_FILE environment variable.
+
+Here is an example of how to run a containerized ose-ansible with
+the openshift_facts playbook, which collects and displays facts about your
+OpenShift environment. The inventory and ssh keys are mounted as volumes
+(the latter requires setting the uid in the container and SELinux label
+in the key file via :Z so they can be accessed) and the PLAYBOOK_FILE
+environment variable is set to point to the playbook within the image:
+
+docker run -tu `id -u` \
+ -v $HOME/.ssh/id_rsa:/opt/app-root/src/.ssh/id_rsa:Z,ro \
+ -v /etc/ansible/hosts:/tmp/inventory:Z,ro \
+ -e INVENTORY_FILE=/tmp/inventory \
+ -e OPTS="-v" \
+ -e PLAYBOOK_FILE=playbooks/byo/openshift_facts.yml \
+ openshift3/ose-ansible
+
+EOF
diff --git a/images/installer/root/usr/local/bin/user_setup b/images/installer/root/usr/local/bin/user_setup
new file mode 100755
index 000000000..b76e60a4d
--- /dev/null
+++ b/images/installer/root/usr/local/bin/user_setup
@@ -0,0 +1,17 @@
+#!/bin/sh
+set -x
+
+# ensure $HOME exists and is accessible by group 0 (we don't know what the runtime UID will be)
+mkdir -p ${HOME}
+chown ${USER_UID}:0 ${HOME}
+chmod ug+rwx ${HOME}
+
+# runtime user will need to be able to self-insert in /etc/passwd
+chmod g+rw /etc/passwd
+
+# ensure that the ansible content is accessible
+chmod -R g+r ${WORK_DIR}
+find ${WORK_DIR} -type d -exec chmod g+x {} +
+
+# no need for this script to remain in the image after running
+rm $0
diff --git a/inventory/byo/hosts.origin.example b/inventory/byo/hosts.origin.example
index 93e821ba7..6e53b4fd9 100644
--- a/inventory/byo/hosts.origin.example
+++ b/inventory/byo/hosts.origin.example
@@ -113,6 +113,11 @@ openshift_release=v3.6
# Downgrades are not supported and will error out. Be careful when upgrading docker from < 1.10 to > 1.10.
# docker_version="1.12.1"
+# Specify whether to run Docker daemon with SELinux enabled in containers. Default is True.
+# Uncomment below to disable; for example if your kernel does not support the
+# Docker overlay/overlay2 storage drivers with SELinux enabled.
+#openshift_docker_selinux_enabled=False
+
# Skip upgrading Docker during an OpenShift upgrade, leaves the current Docker version alone.
# docker_upgrade=False
diff --git a/inventory/byo/hosts.ose.example b/inventory/byo/hosts.ose.example
index 9a0dce01b..e3e9220fc 100644
--- a/inventory/byo/hosts.ose.example
+++ b/inventory/byo/hosts.ose.example
@@ -109,6 +109,11 @@ openshift_release=v3.6
# Default value: "--log-driver=journald"
#openshift_docker_options="-l warn --ipv6=false"
+# Specify whether to run Docker daemon with SELinux enabled in containers. Default is True.
+# Uncomment below to disable; for example if your kernel does not support the
+# Docker overlay/overlay2 storage drivers with SELinux enabled.
+#openshift_docker_selinux_enabled=False
+
# Specify exact version of Docker to configure or upgrade to.
# Downgrades are not supported and will error out. Be careful when upgrading docker from < 1.10 to > 1.10.
# docker_version="1.12.1"
diff --git a/playbooks/byo/openshift-checks/README.md b/playbooks/byo/openshift-checks/README.md
index 4b2ff1f94..f0f14b268 100644
--- a/playbooks/byo/openshift-checks/README.md
+++ b/playbooks/byo/openshift-checks/README.md
@@ -39,7 +39,9 @@ against your inventory file. Here is the step-by-step:
$ cd openshift-ansible
```
-2. Run the appropriate playbook:
+2. Install the [dependencies](../../../README.md#setup)
+
+3. Run the appropriate playbook:
```console
$ ansible-playbook -i <inventory file> playbooks/byo/openshift-checks/pre-install.yml
@@ -57,9 +59,8 @@ against your inventory file. Here is the step-by-step:
$ ansible-playbook -i <inventory file> playbooks/byo/openshift-checks/certificate_expiry/default.yaml -v
```
-## Running via Docker image
+## Running in a container
This repository is built into a Docker image including Ansible so that it can
-be run anywhere Docker is available. Instructions for doing so may be found
-[in the README](../../README_CONTAINER_IMAGE.md).
-
+be run anywhere Docker is available, without the need to manually install dependencies.
+Instructions for doing so may be found [in the README](../../../README_CONTAINER_IMAGE.md).
diff --git a/playbooks/common/openshift-cluster/openshift_hosted.yml b/playbooks/common/openshift-cluster/openshift_hosted.yml
index 8d94b6509..ce7f981ab 100644
--- a/playbooks/common/openshift-cluster/openshift_hosted.yml
+++ b/playbooks/common/openshift-cluster/openshift_hosted.yml
@@ -26,6 +26,8 @@
logging_elasticsearch_cluster_size: "{{ openshift_hosted_logging_elasticsearch_cluster_size | default(1) }}"
logging_elasticsearch_ops_cluster_size: "{{ openshift_hosted_logging_elasticsearch_ops_cluster_size | default(1) }}"
roles:
+ - role: openshift_default_storage_class
+ when: openshift_cloudprovider_kind is defined and (openshift_cloudprovider_kind == 'aws' or openshift_cloudprovider_kind == 'gce')
- role: openshift_hosted
- role: openshift_metrics
when: openshift_hosted_metrics_deploy | default(false) | bool
@@ -45,8 +47,6 @@
- role: cockpit-ui
when: ( openshift.common.version_gte_3_3_or_1_3 | bool ) and ( openshift_hosted_manage_registry | default(true) | bool ) and not (openshift.docker.hosted_registry_insecure | default(false) | bool)
- - role: openshift_default_storage_class
- when: openshift_cloudprovider_kind is defined and (openshift_cloudprovider_kind == 'aws' or openshift_cloudprovider_kind == 'gce')
- name: Update master-config for publicLoggingURL
hosts: oo_masters_to_config:!oo_first_master
diff --git a/roles/ansible_service_broker/defaults/main.yml b/roles/ansible_service_broker/defaults/main.yml
index aa1c5022b..12929b354 100644
--- a/roles/ansible_service_broker/defaults/main.yml
+++ b/roles/ansible_service_broker/defaults/main.yml
@@ -4,6 +4,7 @@ ansible_service_broker_remove: false
ansible_service_broker_log_level: info
ansible_service_broker_output_request: false
ansible_service_broker_recovery: true
+ansible_service_broker_bootstrap_on_startup: true
# Recommended you do not enable this for now
ansible_service_broker_dev_broker: false
ansible_service_broker_launch_apb_on_bind: false
diff --git a/roles/ansible_service_broker/tasks/install.yml b/roles/ansible_service_broker/tasks/install.yml
index 65dffc89b..b3797ef96 100644
--- a/roles/ansible_service_broker/tasks/install.yml
+++ b/roles/ansible_service_broker/tasks/install.yml
@@ -42,6 +42,14 @@
namespace: openshift-ansible-service-broker
state: present
+- name: Set SA cluster-role
+ oc_adm_policy_user:
+ state: present
+ namespace: "openshift-ansible-service-broker"
+ resource_kind: cluster-role
+ resource_name: admin
+ user: "system:serviceaccount:openshift-ansible-service-broker:asb"
+
- name: create ansible-service-broker service
oc_service:
name: asb
@@ -254,6 +262,7 @@
launch_apb_on_bind: {{ ansible_service_broker_launch_apb_on_bind | bool | lower }}
recovery: {{ ansible_service_broker_recovery | bool | lower }}
output_request: {{ ansible_service_broker_output_request | bool | lower }}
+ bootstrap_on_startup: {{ ansible_service_broker_bootstrap_on_startup | bool | lower }}
- name: Create the Broker resource in the catalog
oc_obj:
diff --git a/roles/docker/tasks/package_docker.yml b/roles/docker/tasks/package_docker.yml
index 5f536005d..bc52ab60c 100644
--- a/roles/docker/tasks/package_docker.yml
+++ b/roles/docker/tasks/package_docker.yml
@@ -93,7 +93,7 @@
dest: /etc/sysconfig/docker
regexp: '^OPTIONS=.*$'
line: "OPTIONS='\
- {% if ansible_selinux.status | default(None) == '''enabled''' and docker_selinux_enabled | default(true) %} --selinux-enabled {% endif %}\
+ {% if ansible_selinux.status | default(None) == 'enabled' and docker_selinux_enabled | default(true) | bool %} --selinux-enabled {% endif %}\
{% if docker_log_driver is defined %} --log-driver {{ docker_log_driver }}{% endif %}\
{% if docker_log_options is defined %} {{ docker_log_options | oo_split() | oo_prepend_strings_in_list('--log-opt ') | join(' ')}}{% endif %}\
{% if docker_options is defined %} {{ docker_options }}{% endif %}\
diff --git a/roles/openshift_default_storage_class/defaults/main.yml b/roles/openshift_default_storage_class/defaults/main.yml
index bda83c933..4f371fd89 100644
--- a/roles/openshift_default_storage_class/defaults/main.yml
+++ b/roles/openshift_default_storage_class/defaults/main.yml
@@ -12,6 +12,7 @@ openshift_storageclass_defaults:
provisioner: kubernetes.io/gce-pd
type: pd-standard
+openshift_storageclass_default: "true"
openshift_storageclass_name: "{{ openshift_storageclass_defaults[openshift_cloudprovider_kind]['name'] }}"
openshift_storageclass_provisioner: "{{ openshift_storageclass_defaults[openshift_cloudprovider_kind]['provisioner'] }}"
openshift_storageclass_parameters: "{{ openshift_storageclass_defaults[openshift_cloudprovider_kind]['parameters'] }}"
diff --git a/roles/openshift_default_storage_class/tasks/main.yml b/roles/openshift_default_storage_class/tasks/main.yml
index fd5e4fabe..82cab6746 100644
--- a/roles/openshift_default_storage_class/tasks/main.yml
+++ b/roles/openshift_default_storage_class/tasks/main.yml
@@ -2,9 +2,8 @@
# Install default storage classes in GCE & AWS
- name: Ensure storageclass object
oc_storageclass:
- kind: storageclass
name: "{{ openshift_storageclass_name }}"
- default_storage_class: "true"
+ default_storage_class: "{{ openshift_storageclass_default | default('true') | string}}"
parameters:
type: "{{ openshift_storageclass_parameters.type | default('gp2') }}"
encrypted: "{{ openshift_storageclass_parameters.encrypted | default('false') | string }}"
diff --git a/roles/openshift_health_checker/library/search_journalctl.py b/roles/openshift_health_checker/library/search_journalctl.py
new file mode 100644
index 000000000..3631f71c8
--- /dev/null
+++ b/roles/openshift_health_checker/library/search_journalctl.py
@@ -0,0 +1,150 @@
+#!/usr/bin/python
+"""Interface to journalctl."""
+
+from time import time
+import json
+import re
+import subprocess
+
+from ansible.module_utils.basic import AnsibleModule
+
+
+class InvalidMatcherRegexp(Exception):
+ """Exception class for invalid matcher regexp."""
+ pass
+
+
+class InvalidLogEntry(Exception):
+ """Exception class for invalid / non-json log entries."""
+ pass
+
+
+class LogInputSubprocessError(Exception):
+ """Exception class for errors that occur while executing a subprocess."""
+ pass
+
+
+def main():
+ """Scan a given list of "log_matchers" for journalctl messages containing given patterns.
+ "log_matchers" is a list of dicts consisting of three keys that help fine-tune log searching:
+ 'start_regexp', 'regexp', and 'unit'.
+
+ Sample "log_matchers" list:
+
+ [
+ {
+ 'start_regexp': r'Beginning of systemd unit',
+ 'regexp': r'the specific log message to find',
+ 'unit': 'etcd',
+ }
+ ]
+ """
+ module = AnsibleModule(
+ argument_spec=dict(
+ log_count_limit=dict(type="int", default=500),
+ log_matchers=dict(type="list", required=True),
+ ),
+ )
+
+ timestamp_limit_seconds = time() - 60 * 60 # 1 hour
+
+ log_count_limit = module.params["log_count_limit"]
+ log_matchers = module.params["log_matchers"]
+
+ matched_regexp, errors = get_log_matches(log_matchers, log_count_limit, timestamp_limit_seconds)
+
+ module.exit_json(
+ changed=False,
+ failed=bool(errors),
+ errors=errors,
+ matched=matched_regexp,
+ )
+
+
+def get_log_matches(matchers, log_count_limit, timestamp_limit_seconds):
+ """Return a list of up to log_count_limit matches for each matcher.
+
+ Log entries are only considered if newer than timestamp_limit_seconds.
+ """
+ matched_regexp = []
+ errors = []
+
+ for matcher in matchers:
+ try:
+ log_output = get_log_output(matcher)
+ except LogInputSubprocessError as err:
+ errors.append(str(err))
+ continue
+
+ try:
+ matched = find_matches(log_output, matcher, log_count_limit, timestamp_limit_seconds)
+ if matched:
+ matched_regexp.append(matcher.get("regexp", ""))
+ except InvalidMatcherRegexp as err:
+ errors.append(str(err))
+ except InvalidLogEntry as err:
+ errors.append(str(err))
+
+ return matched_regexp, errors
+
+
+def get_log_output(matcher):
+ """Return an iterator on the logs of a given matcher."""
+ try:
+ cmd_output = subprocess.Popen(list([
+ '/bin/journalctl',
+ '-ru', matcher.get("unit", ""),
+ '--output', 'json',
+ ]), stdout=subprocess.PIPE)
+
+ return iter(cmd_output.stdout.readline, '')
+
+ except subprocess.CalledProcessError as exc:
+ msg = "Could not obtain journalctl logs for the specified systemd unit: {}: {}"
+ raise LogInputSubprocessError(msg.format(matcher.get("unit", "<missing>"), str(exc)))
+ except OSError as exc:
+ raise LogInputSubprocessError(str(exc))
+
+
+def find_matches(log_output, matcher, log_count_limit, timestamp_limit_seconds):
+ """Return log messages matched in iterable log_output by a given matcher.
+
+ Ignore any log_output items older than timestamp_limit_seconds.
+ """
+ try:
+ regexp = re.compile(matcher.get("regexp", ""))
+ start_regexp = re.compile(matcher.get("start_regexp", ""))
+ except re.error as err:
+ msg = "A log matcher object was provided with an invalid regular expression: {}"
+ raise InvalidMatcherRegexp(msg.format(str(err)))
+
+ matched = None
+
+ for log_count, line in enumerate(log_output):
+ if log_count >= log_count_limit:
+ break
+
+ try:
+ obj = json.loads(line)
+
+ # don't need to look past the most recent service restart
+ if start_regexp.match(obj["MESSAGE"]):
+ break
+
+ log_timestamp_seconds = float(obj["__REALTIME_TIMESTAMP"]) / 1000000
+ if log_timestamp_seconds < timestamp_limit_seconds:
+ break
+
+ if regexp.match(obj["MESSAGE"]):
+ matched = line
+ break
+
+ except ValueError:
+ msg = "Log entry for systemd unit {} contained invalid json syntax: {}"
+ raise InvalidLogEntry(msg.format(matcher.get("unit"), line))
+
+ return matched
+
+
+if __name__ == '__main__':
+ main()
diff --git a/roles/openshift_health_checker/openshift_checks/docker_storage.py b/roles/openshift_health_checker/openshift_checks/docker_storage.py
index e80691ef3..d2227d244 100644
--- a/roles/openshift_health_checker/openshift_checks/docker_storage.py
+++ b/roles/openshift_health_checker/openshift_checks/docker_storage.py
@@ -1,5 +1,6 @@
"""Check Docker storage driver and usage."""
import json
+import os.path
import re
from openshift_checks import OpenShiftCheck, OpenShiftCheckException, get_var
from openshift_checks.mixins import DockerHostMixin
@@ -20,10 +21,27 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
storage_drivers = ["devicemapper", "overlay", "overlay2"]
max_thinpool_data_usage_percent = 90.0
max_thinpool_meta_usage_percent = 90.0
+ max_overlay_usage_percent = 90.0
+
+ # TODO(lmeyer): mention these in the output when check fails
+ configuration_variables = [
+ (
+ "max_thinpool_data_usage_percent",
+ "For 'devicemapper' storage driver, usage threshold percentage for data. "
+ "Format: float. Default: {:.1f}".format(max_thinpool_data_usage_percent),
+ ),
+ (
+ "max_thinpool_meta_usage_percent",
+ "For 'devicemapper' storage driver, usage threshold percentage for metadata. "
+ "Format: float. Default: {:.1f}".format(max_thinpool_meta_usage_percent),
+ ),
+ (
+ "max_overlay_usage_percent",
+ "For 'overlay' or 'overlay2' storage driver, usage threshold percentage. "
+ "Format: float. Default: {:.1f}".format(max_overlay_usage_percent),
+ ),
+ ]
- # pylint: disable=too-many-return-statements
- # Reason: permanent stylistic exception;
- # it is clearer to return on failures and there are just many ways to fail here.
def run(self, tmp, task_vars):
msg, failed, changed = self.ensure_dependencies(task_vars)
if failed:
@@ -34,17 +52,17 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
}
# attempt to get the docker info hash from the API
- info = self.execute_module("docker_info", {}, task_vars=task_vars)
- if info.get("failed"):
+ docker_info = self.execute_module("docker_info", {}, task_vars=task_vars)
+ if docker_info.get("failed"):
return {"failed": True, "changed": changed,
"msg": "Failed to query Docker API. Is docker running on this host?"}
- if not info.get("info"): # this would be very strange
+ if not docker_info.get("info"): # this would be very strange
return {"failed": True, "changed": changed,
- "msg": "Docker API query missing info:\n{}".format(json.dumps(info))}
- info = info["info"]
+ "msg": "Docker API query missing info:\n{}".format(json.dumps(docker_info))}
+ docker_info = docker_info["info"]
# check if the storage driver we saw is valid
- driver = info.get("Driver", "[NONE]")
+ driver = docker_info.get("Driver", "[NONE]")
if driver not in self.storage_drivers:
msg = (
"Detected unsupported Docker storage driver '{driver}'.\n"
@@ -53,26 +71,34 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
return {"failed": True, "changed": changed, "msg": msg}
# driver status info is a list of tuples; convert to dict and validate based on driver
- driver_status = {item[0]: item[1] for item in info.get("DriverStatus", [])}
+ driver_status = {item[0]: item[1] for item in docker_info.get("DriverStatus", [])}
+
+ result = {}
+
if driver == "devicemapper":
- if driver_status.get("Data loop file"):
- msg = (
- "Use of loopback devices with the Docker devicemapper storage driver\n"
- "(the default storage configuration) is unsupported in production.\n"
- "Please use docker-storage-setup to configure a backing storage volume.\n"
- "See http://red.ht/2rNperO for further information."
- )
- return {"failed": True, "changed": changed, "msg": msg}
- result = self._check_dm_usage(driver_status, task_vars)
- result['changed'] = result.get('changed', False) or changed
- return result
+ result = self.check_devicemapper_support(driver_status, task_vars)
- # TODO(lmeyer): determine how to check usage for overlay2
+ if driver in ['overlay', 'overlay2']:
+ result = self.check_overlay_support(docker_info, driver_status, task_vars)
- return {"changed": changed}
+ result['changed'] = result.get('changed', False) or changed
+ return result
- def _check_dm_usage(self, driver_status, task_vars):
- """
+ def check_devicemapper_support(self, driver_status, task_vars):
+ """Check if dm storage driver is supported as configured. Return: result dict."""
+ if driver_status.get("Data loop file"):
+ msg = (
+ "Use of loopback devices with the Docker devicemapper storage driver\n"
+ "(the default storage configuration) is unsupported in production.\n"
+ "Please use docker-storage-setup to configure a backing storage volume.\n"
+ "See http://red.ht/2rNperO for further information."
+ )
+ return {"failed": True, "msg": msg}
+ result = self.check_dm_usage(driver_status, task_vars)
+ return result
+
+ def check_dm_usage(self, driver_status, task_vars):
+ """Check usage thresholds for Docker dm storage driver. Return: result dict.
Backing assumptions: We expect devicemapper to be backed by an auto-expanding thin pool
implemented as an LV in an LVM2 VG. This is how docker-storage-setup currently configures
devicemapper storage. The LV is "thin" because it does not use all available storage
@@ -83,7 +109,7 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
could run out of space first; so we check both.
"""
vals = dict(
- vg_free=self._get_vg_free(driver_status.get("Pool Name"), task_vars),
+ vg_free=self.get_vg_free(driver_status.get("Pool Name"), task_vars),
data_used=driver_status.get("Data Space Used"),
data_total=driver_status.get("Data Space Total"),
metadata_used=driver_status.get("Metadata Space Used"),
@@ -93,7 +119,7 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
# convert all human-readable strings to bytes
for key, value in vals.copy().items():
try:
- vals[key + "_bytes"] = self._convert_to_bytes(value)
+ vals[key + "_bytes"] = self.convert_to_bytes(value)
except ValueError as err: # unlikely to hit this from API info, but just to be safe
return {
"failed": True,
@@ -131,10 +157,12 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
vals["msg"] = "\n".join(messages or ["Thinpool usage is within thresholds."])
return vals
- def _get_vg_free(self, pool, task_vars):
- # Determine which VG to examine according to the pool name, the only indicator currently
- # available from the Docker API driver info. We assume a name that looks like
- # "vg--name-docker--pool"; vg and lv names with inner hyphens doubled, joined by a hyphen.
+ def get_vg_free(self, pool, task_vars):
+ """Determine which VG to examine according to the pool name. Return: size vgs reports.
+ Pool name is the only indicator currently available from the Docker API driver info.
+ We assume a name that looks like "vg--name-docker--pool";
+ vg and lv names with inner hyphens doubled, joined by a hyphen.
+ """
match = re.match(r'((?:[^-]|--)+)-(?!-)', pool) # matches up to the first single hyphen
if not match: # unlikely, but... be clear if we assumed wrong
raise OpenShiftCheckException(
@@ -163,7 +191,8 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
return size
@staticmethod
- def _convert_to_bytes(string):
+ def convert_to_bytes(string):
+ """Convert string like "10.3 G" to bytes (binary units assumed). Return: float bytes."""
units = dict(
b=1,
k=1024,
@@ -183,3 +212,87 @@ class DockerStorage(DockerHostMixin, OpenShiftCheck):
raise ValueError("Cannot convert to a byte size: " + string)
return float(number) * multiplier
+
+ def check_overlay_support(self, docker_info, driver_status, task_vars):
+ """Check if overlay storage driver is supported for this host. Return: result dict."""
+ # check for xfs as backing store
+ backing_fs = driver_status.get("Backing Filesystem", "[NONE]")
+ if backing_fs != "xfs":
+ msg = (
+ "Docker storage drivers 'overlay' and 'overlay2' are only supported with\n"
+ "'xfs' as the backing storage, but this host's storage is type '{fs}'."
+ ).format(fs=backing_fs)
+ return {"failed": True, "msg": msg}
+
+ # check support for OS and kernel version
+ o_s = docker_info.get("OperatingSystem", "[NONE]")
+ if "Red Hat Enterprise Linux" in o_s or "CentOS" in o_s:
+ # keep it simple, only check enterprise kernel versions; assume everyone else is good
+ kernel = docker_info.get("KernelVersion", "[NONE]")
+ kernel_arr = [int(num) for num in re.findall(r'\d+', kernel)]
+ if kernel_arr < [3, 10, 0, 514]: # rhel < 7.3
+ msg = (
+ "Docker storage drivers 'overlay' and 'overlay2' are only supported beginning with\n"
+ "kernel version 3.10.0-514; but Docker reports kernel version {version}."
+ ).format(version=kernel)
+ return {"failed": True, "msg": msg}
+ # NOTE: we could check for --selinux-enabled here but docker won't even start with
+ # that option until it's supported in the kernel so we don't need to.
+
+ return self.check_overlay_usage(docker_info, task_vars)
+
+ def check_overlay_usage(self, docker_info, task_vars):
+ """Check disk usage on OverlayFS backing store volume. Return: result dict."""
+ path = docker_info.get("DockerRootDir", "/var/lib/docker") + "/" + docker_info["Driver"]
+
+ threshold = get_var(task_vars, "max_overlay_usage_percent", default=self.max_overlay_usage_percent)
+ try:
+ threshold = float(threshold)
+ except ValueError:
+ return {
+ "failed": True,
+ "msg": "Specified 'max_overlay_usage_percent' is not a percentage: {}".format(threshold),
+ }
+
+ mount = self.find_ansible_mount(path, get_var(task_vars, "ansible_mounts"))
+ try:
+ free_bytes = mount['size_available']
+ total_bytes = mount['size_total']
+ usage = 100.0 * (total_bytes - free_bytes) / total_bytes
+ except (KeyError, ZeroDivisionError):
+ return {
+ "failed": True,
+ "msg": "The ansible_mount found for path {} is invalid.\n"
+ "This is likely to be an Ansible bug. The record was:\n"
+ "{}".format(path, json.dumps(mount, indent=2)),
+ }
+
+ if usage > threshold:
+ return {
+ "failed": True,
+ "msg": (
+ "For Docker OverlayFS mount point {path},\n"
+ "usage percentage {pct:.1f} is higher than threshold {thresh:.1f}."
+ ).format(path=mount["mount"], pct=usage, thresh=threshold)
+ }
+
+ return {}
+
+ # TODO(lmeyer): migrate to base class
+ @staticmethod
+ def find_ansible_mount(path, ansible_mounts):
+ """Return the mount point for path from ansible_mounts."""
+
+ mount_for_path = {mount['mount']: mount for mount in ansible_mounts}
+ mount_point = path
+ while mount_point not in mount_for_path:
+ if mount_point in ["/", ""]: # "/" not in ansible_mounts???
+ break
+ mount_point = os.path.dirname(mount_point)
+
+ try:
+ return mount_for_path[mount_point]
+ except KeyError:
+ known_mounts = ', '.join('"{}"'.format(mount) for mount in sorted(mount_for_path)) or 'none'
+ msg = 'Unable to determine mount point for path "{}". Known mount points: {}.'
+ raise OpenShiftCheckException(msg.format(path, known_mounts))
diff --git a/roles/openshift_health_checker/openshift_checks/etcd_traffic.py b/roles/openshift_health_checker/openshift_checks/etcd_traffic.py
new file mode 100644
index 000000000..40c87873d
--- /dev/null
+++ b/roles/openshift_health_checker/openshift_checks/etcd_traffic.py
@@ -0,0 +1,47 @@
+"""Check that scans journalctl for messages caused as a symptom of increased etcd traffic."""
+
+from openshift_checks import OpenShiftCheck, get_var
+
+
+class EtcdTraffic(OpenShiftCheck):
+ """Check if host is being affected by an increase in etcd traffic."""
+
+ name = "etcd_traffic"
+ tags = ["health", "etcd"]
+
+ @classmethod
+ def is_active(cls, task_vars):
+ """Skip hosts that do not have etcd in their group names."""
+ group_names = get_var(task_vars, "group_names", default=[])
+ valid_group_names = "etcd" in group_names
+
+ version = get_var(task_vars, "openshift", "common", "short_version")
+ valid_version = version in ("3.4", "3.5", "1.4", "1.5")
+
+ return super(EtcdTraffic, cls).is_active(task_vars) and valid_group_names and valid_version
+
+ def run(self, tmp, task_vars):
+ is_containerized = get_var(task_vars, "openshift", "common", "is_containerized")
+ unit = "etcd_container" if is_containerized else "etcd"
+
+ log_matchers = [{
+ "start_regexp": r"Starting Etcd Server",
+ "regexp": r"etcd: sync duration of [^,]+, expected less than 1s",
+ "unit": unit
+ }]
+
+ match = self.execute_module("search_journalctl", {
+ "log_matchers": log_matchers,
+ }, task_vars)
+
+ if match.get("matched"):
+ msg = ("Higher than normal etcd traffic detected.\n"
+ "OpenShift 3.4 introduced an increase in etcd traffic.\n"
+ "Upgrading to OpenShift 3.6 is recommended in order to fix this issue.\n"
+ "Please refer to https://access.redhat.com/solutions/2916381 for more information.")
+ return {"failed": True, "msg": msg}
+
+ if match.get("failed"):
+ return {"failed": True, "msg": "\n".join(match.get("errors"))}
+
+ return {}
diff --git a/roles/openshift_health_checker/test/docker_storage_test.py b/roles/openshift_health_checker/test/docker_storage_test.py
index bb25e3f66..99c529054 100644
--- a/roles/openshift_health_checker/test/docker_storage_test.py
+++ b/roles/openshift_health_checker/test/docker_storage_test.py
@@ -23,7 +23,8 @@ def test_is_active(is_containerized, group_names, is_active):
assert DockerStorage.is_active(task_vars=task_vars) == is_active
-non_atomic_task_vars = {"openshift": {"common": {"is_atomic": False}}}
+def non_atomic_task_vars():
+ return {"openshift": {"common": {"is_atomic": False}}}
@pytest.mark.parametrize('docker_info, failed, expect_msg', [
@@ -56,7 +57,7 @@ non_atomic_task_vars = {"openshift": {"common": {"is_atomic": False}}}
(
dict(info={
"Driver": "overlay2",
- "DriverStatus": []
+ "DriverStatus": [("Backing Filesystem", "xfs")],
}),
False,
[],
@@ -64,6 +65,27 @@ non_atomic_task_vars = {"openshift": {"common": {"is_atomic": False}}}
(
dict(info={
"Driver": "overlay",
+ "DriverStatus": [("Backing Filesystem", "btrfs")],
+ }),
+ True,
+ ["storage is type 'btrfs'", "only supported with\n'xfs'"],
+ ),
+ (
+ dict(info={
+ "Driver": "overlay2",
+ "DriverStatus": [("Backing Filesystem", "xfs")],
+ "OperatingSystem": "Red Hat Enterprise Linux Server release 7.2 (Maipo)",
+ "KernelVersion": "3.10.0-327.22.2.el7.x86_64",
+ }),
+ True,
+ ["Docker reports kernel version 3.10.0-327"],
+ ),
+ (
+ dict(info={
+ "Driver": "overlay",
+ "DriverStatus": [("Backing Filesystem", "xfs")],
+ "OperatingSystem": "CentOS",
+ "KernelVersion": "3.10.0-514",
}),
False,
[],
@@ -85,8 +107,9 @@ def test_check_storage_driver(docker_info, failed, expect_msg):
return docker_info
check = dummy_check(execute_module=execute_module)
- check._check_dm_usage = lambda status, task_vars: dict() # stub out for this test
- result = check.run(tmp=None, task_vars=non_atomic_task_vars)
+ check.check_dm_usage = lambda status, task_vars: dict() # stub out for this test
+ check.check_overlay_usage = lambda info, task_vars: dict() # stub out for this test
+ result = check.run(tmp=None, task_vars=non_atomic_task_vars())
if failed:
assert result["failed"]
@@ -146,8 +169,8 @@ not_enough_space = {
])
def test_dm_usage(task_vars, driver_status, vg_free, success, expect_msg):
check = dummy_check()
- check._get_vg_free = lambda pool, task_vars: vg_free
- result = check._check_dm_usage(driver_status, task_vars)
+ check.get_vg_free = lambda pool, task_vars: vg_free
+ result = check.check_dm_usage(driver_status, task_vars)
result_success = not result.get("failed")
assert result_success is success
@@ -195,10 +218,10 @@ def test_vg_free(pool, command_returns, raises, returns):
check = dummy_check(execute_module=execute_module)
if raises:
with pytest.raises(OpenShiftCheckException) as err:
- check._get_vg_free(pool, {})
+ check.get_vg_free(pool, {})
assert raises in str(err.value)
else:
- ret = check._get_vg_free(pool, {})
+ ret = check.get_vg_free(pool, {})
assert ret == returns
@@ -209,7 +232,7 @@ def test_vg_free(pool, command_returns, raises, returns):
("12g", 12.0 * 1024**3),
])
def test_convert_to_bytes(string, expect_bytes):
- got = DockerStorage._convert_to_bytes(string)
+ got = DockerStorage.convert_to_bytes(string)
assert got == expect_bytes
@@ -219,6 +242,70 @@ def test_convert_to_bytes(string, expect_bytes):
])
def test_convert_to_bytes_error(string):
with pytest.raises(ValueError) as err:
- DockerStorage._convert_to_bytes(string)
+ DockerStorage.convert_to_bytes(string)
assert "Cannot convert" in str(err.value)
assert string in str(err.value)
+
+
+ansible_mounts_enough = [{
+ 'mount': '/var/lib/docker',
+ 'size_available': 50 * 10**9,
+ 'size_total': 50 * 10**9,
+}]
+ansible_mounts_not_enough = [{
+ 'mount': '/var/lib/docker',
+ 'size_available': 0,
+ 'size_total': 50 * 10**9,
+}]
+ansible_mounts_missing_fields = [dict(mount='/var/lib/docker')]
+ansible_mounts_zero_size = [{
+ 'mount': '/var/lib/docker',
+ 'size_available': 0,
+ 'size_total': 0,
+}]
+
+
+@pytest.mark.parametrize('ansible_mounts, threshold, expect_fail, expect_msg', [
+ (
+ ansible_mounts_enough,
+ None,
+ False,
+ [],
+ ),
+ (
+ ansible_mounts_not_enough,
+ None,
+ True,
+ ["usage percentage", "higher than threshold"],
+ ),
+ (
+ ansible_mounts_not_enough,
+ "bogus percent",
+ True,
+ ["is not a percentage"],
+ ),
+ (
+ ansible_mounts_missing_fields,
+ None,
+ True,
+ ["Ansible bug"],
+ ),
+ (
+ ansible_mounts_zero_size,
+ None,
+ True,
+ ["Ansible bug"],
+ ),
+])
+def test_overlay_usage(ansible_mounts, threshold, expect_fail, expect_msg):
+ check = dummy_check()
+ task_vars = non_atomic_task_vars()
+ task_vars["ansible_mounts"] = ansible_mounts
+ if threshold is not None:
+ task_vars["max_overlay_usage_percent"] = threshold
+ docker_info = dict(DockerRootDir="/var/lib/docker", Driver="overlay")
+ result = check.check_overlay_usage(docker_info, task_vars)
+
+ assert expect_fail == bool(result.get("failed"))
+ for msg in expect_msg:
+ assert msg in result["msg"]
diff --git a/roles/openshift_health_checker/test/etcd_traffic_test.py b/roles/openshift_health_checker/test/etcd_traffic_test.py
new file mode 100644
index 000000000..287175e29
--- /dev/null
+++ b/roles/openshift_health_checker/test/etcd_traffic_test.py
@@ -0,0 +1,80 @@
+import pytest
+
+from openshift_checks.etcd_traffic import EtcdTraffic
+
+
+@pytest.mark.parametrize('group_names,version,is_active', [
+ (['masters'], "3.5", False),
+ (['masters'], "3.6", False),
+ (['nodes'], "3.4", False),
+ (['etcd'], "3.4", True),
+ (['etcd'], "3.5", True),
+ (['etcd'], "3.1", False),
+ (['masters', 'nodes'], "3.5", False),
+ (['masters', 'etcd'], "3.5", True),
+ ([], "3.4", False),
+])
+def test_is_active(group_names, version, is_active):
+ task_vars = dict(
+ group_names=group_names,
+ openshift=dict(
+ common=dict(short_version=version),
+ ),
+ )
+ assert EtcdTraffic.is_active(task_vars=task_vars) == is_active
+
+
+@pytest.mark.parametrize('group_names,matched,failed,extra_words', [
+ (["masters"], True, True, ["Higher than normal", "traffic"]),
+ (["masters", "etcd"], False, False, []),
+ (["etcd"], False, False, []),
+])
+def test_log_matches_high_traffic_msg(group_names, matched, failed, extra_words):
+ def execute_module(module_name, args, task_vars):
+ return {
+ "matched": matched,
+ "failed": failed,
+ }
+
+ task_vars = dict(
+ group_names=group_names,
+ openshift=dict(
+ common=dict(service_type="origin", is_containerized=False),
+ )
+ )
+
+ check = EtcdTraffic(execute_module=execute_module)
+ result = check.run(tmp=None, task_vars=task_vars)
+
+ for word in extra_words:
+ assert word in result.get("msg", "")
+
+ assert result.get("failed", False) == failed
+
+
+@pytest.mark.parametrize('is_containerized,expected_unit_value', [
+ (False, "etcd"),
+ (True, "etcd_container"),
+])
+def test_systemd_unit_matches_deployment_type(is_containerized, expected_unit_value):
+ task_vars = dict(
+ openshift=dict(
+ common=dict(is_containerized=is_containerized),
+ )
+ )
+
+ def execute_module(module_name, args, task_vars):
+ assert module_name == "search_journalctl"
+ matchers = args["log_matchers"]
+
+ for matcher in matchers:
+ assert matcher["unit"] == expected_unit_value
+
+ return {"failed": False}
+
+ check = EtcdTraffic(execute_module=execute_module)
+ check.run(tmp=None, task_vars=task_vars)
+
+
+def fake_execute_module(*args):
+ raise AssertionError('this function should not be called')
diff --git a/roles/openshift_health_checker/test/search_journalctl_test.py b/roles/openshift_health_checker/test/search_journalctl_test.py
new file mode 100644
index 000000000..724928aa1
--- /dev/null
+++ b/roles/openshift_health_checker/test/search_journalctl_test.py
@@ -0,0 +1,157 @@
+import pytest
+import search_journalctl
+
+
+def canned_search_journalctl(get_log_output=None):
+ """Create a search_journalctl object with canned get_log_output method"""
+ module = search_journalctl
+ if get_log_output:
+ module.get_log_output = get_log_output
+ return module
+
+
+DEFAULT_TIMESTAMP = 1496341364
+
+
+def get_timestamp(modifier=0):
+ return DEFAULT_TIMESTAMP + modifier
+
+
+def get_timestamp_microseconds(modifier=0):
+ return get_timestamp(modifier) * 1000000
+
+
+def create_test_log_object(stamp, msg):
+ return '{{"__REALTIME_TIMESTAMP": "{}", "MESSAGE": "{}"}}'.format(stamp, msg)
+
+
+@pytest.mark.parametrize('name,matchers,log_input,expected_matches,expected_errors', [
+ (
+ 'test with valid params',
+ [
+ {
+ "start_regexp": r"Sample Logs Beginning",
+ "regexp": r"test log message",
+ "unit": "test",
+ },
+ ],
+ [
+ create_test_log_object(get_timestamp_microseconds(), "test log message"),
+ create_test_log_object(get_timestamp_microseconds(), "Sample Logs Beginning"),
+ ],
+ ["test log message"],
+ [],
+ ),
+ (
+ 'test with invalid json in log input',
+ [
+ {
+ "start_regexp": r"Sample Logs Beginning",
+ "regexp": r"test log message",
+ "unit": "test-unit",
+ },
+ ],
+ [
+ '{__REALTIME_TIMESTAMP: ' + str(get_timestamp_microseconds()) + ', "MESSAGE": "test log message"}',
+ ],
+ [],
+ [
+ ["invalid json", "test-unit", "test log message"],
+ ],
+ ),
+ (
+ 'test with invalid regexp',
+ [
+ {
+ "start_regexp": r"Sample Logs Beginning",
+ "regexp": r"test [ log message",
+ "unit": "test",
+ },
+ ],
+ [
+ create_test_log_object(get_timestamp_microseconds(), "test log message"),
+ create_test_log_object(get_timestamp_microseconds(), "sample log message"),
+ create_test_log_object(get_timestamp_microseconds(), "fake log message"),
+ create_test_log_object(get_timestamp_microseconds(), "dummy log message"),
+ create_test_log_object(get_timestamp_microseconds(), "Sample Logs Beginning"),
+ ],
+ [],
+ [
+ ["invalid regular expression"],
+ ],
+ ),
+], ids=lambda argval: argval[0])
+def test_get_log_matches(name, matchers, log_input, expected_matches, expected_errors):
+ def get_log_output(matcher):
+ return log_input
+
+ module = canned_search_journalctl(get_log_output)
+ matched_regexp, errors = module.get_log_matches(matchers, 500, 60 * 60)
+
+ assert set(matched_regexp) == set(expected_matches)
+ assert len(expected_errors) == len(errors)
+
+ for idx, partial_err_set in enumerate(expected_errors):
+ for partial_err_msg in partial_err_set:
+ assert partial_err_msg in errors[idx]
+
+
+@pytest.mark.parametrize('name,matcher,log_count_lim,stamp_lim_seconds,log_input,expected_match', [
+ (
+ 'test with matching log message, but out of bounds of log_count_lim',
+ {
+ "start_regexp": r"Sample Logs Beginning",
+ "regexp": r"dummy log message",
+ "unit": "test",
+ },
+ 3,
+ get_timestamp(-100 * 60 * 60),
+ [
+ create_test_log_object(get_timestamp_microseconds(), "test log message"),
+ create_test_log_object(get_timestamp_microseconds(), "sample log message"),
+ create_test_log_object(get_timestamp_microseconds(), "fake log message"),
+ create_test_log_object(get_timestamp_microseconds(), "dummy log message"),
+ create_test_log_object(get_timestamp_microseconds(), "Sample Logs Beginning"),
+ ],
+ None,
+ ),
+ (
+ 'test with matching log message, but with timestamp too old',
+ {
+ "start_regexp": r"Sample Logs Beginning",
+ "regexp": r"dummy log message",
+ "unit": "test",
+ },
+ 100,
+ get_timestamp(-10),
+ [
+ create_test_log_object(get_timestamp_microseconds(), "test log message"),
+ create_test_log_object(get_timestamp_microseconds(), "sample log message"),
+ create_test_log_object(get_timestamp_microseconds(), "fake log message"),
+ create_test_log_object(get_timestamp_microseconds(-1000), "dummy log message"),
+ create_test_log_object(get_timestamp_microseconds(-1000), "Sample Logs Beginning"),
+ ],
+ None,
+ ),
+ (
+ 'test with matching log message, and timestamp within time limit',
+ {
+ "start_regexp": r"Sample Logs Beginning",
+ "regexp": r"dummy log message",
+ "unit": "test",
+ },
+ 100,
+ get_timestamp(-1010),
+ [
+ create_test_log_object(get_timestamp_microseconds(), "test log message"),
+ create_test_log_object(get_timestamp_microseconds(), "sample log message"),
+ create_test_log_object(get_timestamp_microseconds(), "fake log message"),
+ create_test_log_object(get_timestamp_microseconds(-1000), "dummy log message"),
+ create_test_log_object(get_timestamp_microseconds(-1000), "Sample Logs Beginning"),
+ ],
+ create_test_log_object(get_timestamp_microseconds(-1000), "dummy log message"),
+ ),
+], ids=lambda argval: argval[0])
+def test_find_matches_skips_logs(name, matcher, log_count_lim, stamp_lim_seconds, log_input, expected_match):
+ match = search_journalctl.find_matches(log_input, matcher, log_count_lim, stamp_lim_seconds)
+ assert match == expected_match
diff --git a/roles/openshift_service_catalog/files/kubeservicecatalog_roles_bindings.yml b/roles/openshift_service_catalog/files/kubeservicecatalog_roles_bindings.yml
index bcc7fb590..c30c15778 100644
--- a/roles/openshift_service_catalog/files/kubeservicecatalog_roles_bindings.yml
+++ b/roles/openshift_service_catalog/files/kubeservicecatalog_roles_bindings.yml
@@ -137,6 +137,12 @@ objects:
- serviceclasses
verbs:
- create
+ - delete
+ - update
+ - patch
+ - get
+ - list
+ - watch
- apiGroups:
- settings.k8s.io
resources: