#!/bin/bash # Copyright 2014 Google Inc. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # Common utilities, variables and checks for all build scripts. set -o errexit set -o nounset set -o pipefail KUBE_ROOT=$(dirname "${BASH_SOURCE}")/.. cd "${KUBE_ROOT}" # This'll canonicalize the path KUBE_ROOT=$PWD source hack/lib/init.sh # Incoming options # readonly KUBE_SKIP_CONFIRMATIONS="${KUBE_SKIP_CONFIRMATIONS:-n}" readonly KUBE_BUILD_RUN_IMAGES="${KUBE_BUILD_RUN_IMAGES:-n}" readonly KUBE_GCS_UPLOAD_RELEASE="${KUBE_GCS_UPLOAD_RELEASE:-n}" readonly KUBE_GCS_NO_CACHING="${KUBE_GCS_NO_CACHING:-y}" readonly KUBE_GCS_MAKE_PUBLIC="${KUBE_GCS_MAKE_PUBLIC:-y}" # KUBE_GCS_RELEASE_BUCKET default: kubernetes-releases-${project_hash} # KUBE_GCS_RELEASE_PREFIX default: devel/ # KUBE_GCS_DOCKER_REG_PREFIX default: docker-reg/ # Constants readonly KUBE_BUILD_IMAGE_REPO=kube-build # These get set in verify_prereqs with a unique hash based on KUBE_ROOT # KUBE_BUILD_IMAGE_TAG= # KUBE_BUILD_IMAGE="${KUBE_BUILD_IMAGE_REPO}:${KUBE_BUILD_IMAGE_TAG}" # KUBE_BUILD_CONTAINER_NAME=kube-build- readonly KUBE_BUILD_IMAGE_CROSS_TAG=cross readonly KUBE_BUILD_IMAGE_CROSS="${KUBE_BUILD_IMAGE_REPO}:${KUBE_BUILD_IMAGE_CROSS_TAG}" readonly KUBE_BUILD_GOLANG_VERSION=1.3 # KUBE_BUILD_DATA_CONTAINER_NAME=kube-build-data- # Here we map the output directories across both the local and remote _output # directories: # # *_OUTPUT_ROOT - the base of all output in that environment. # *_OUTPUT_SUBPATH - location where golang stuff is built/cached. Also # persisted across docker runs with a volume mount. # *_OUTPUT_BINPATH - location where final binaries are placed. If the remote # is really remote, this is the stuff that has to be copied # back. readonly LOCAL_OUTPUT_ROOT="${KUBE_ROOT}/_output" readonly LOCAL_OUTPUT_SUBPATH="${LOCAL_OUTPUT_ROOT}/dockerized" readonly LOCAL_OUTPUT_BINPATH="${LOCAL_OUTPUT_SUBPATH}/bin" readonly LOCAL_OUTPUT_IMAGE_STAGING="${LOCAL_OUTPUT_ROOT}/images" readonly REMOTE_OUTPUT_ROOT="/go/src/${KUBE_GO_PACKAGE}/_output" readonly REMOTE_OUTPUT_SUBPATH="${REMOTE_OUTPUT_ROOT}/dockerized" readonly REMOTE_OUTPUT_BINPATH="${REMOTE_OUTPUT_SUBPATH}/bin" readonly DOCKER_MOUNT_ARGS_BASE=(--volume "${LOCAL_OUTPUT_BINPATH}:${REMOTE_OUTPUT_BINPATH}") # DOCKER_MOUNT_ARGS=("${DOCKER_MOUNT_ARGS_BASE[@]}" --volumes-from "${KUBE_BUILD_DATA_CONTAINER_NAME}") # We create a Docker data container to cache incremental build artifacts. We # need to cache both the go tree in _output and the go tree under Godeps. readonly REMOTE_OUTPUT_GOPATH="${REMOTE_OUTPUT_SUBPATH}/go" readonly REMOTE_GODEP_GOPATH="/go/src/${KUBE_GO_PACKAGE}/Godeps/_workspace/pkg" readonly DOCKER_DATA_MOUNT_ARGS=( --volume "${REMOTE_OUTPUT_GOPATH}" --volume "${REMOTE_GODEP_GOPATH}" ) readonly KUBE_RUN_IMAGE_BASE="kubernetes" readonly KUBE_RUN_IMAGES=( apiserver controller-manager proxy scheduler kubelet bootstrap ) # This is where the final release artifacts are created locally readonly RELEASE_DIR="${LOCAL_OUTPUT_ROOT}/release-tars" # --------------------------------------------------------------------------- # Basic setup functions # Verify that the right utilities and such are installed for building Kube. Set # up some dynamic constants. # # Vars set: # KUBE_ROOT_HASH # KUBE_BUILD_IMAGE_TAG # KUBE_BUILD_IMAGE # KUBE_BUILD_CONTAINER_NAME # KUBE_BUILD_DATA_CONTAINER_NAME # DOCKER_MOUNT_ARGS function kube::build::verify_prereqs() { if [[ -z "$(which docker)" ]]; then echo "Can't find 'docker' in PATH, please fix and retry." >&2 echo "See https://docs.docker.com/installation/#installation for installation instructions." >&2 exit 1 fi if kube::build::is_osx; then if [[ -z "$(which boot2docker)" ]]; then echo "It looks like you are running on Mac OS X and boot2docker can't be found." >&2 echo "See: https://docs.docker.com/installation/mac/" >&2 exit 1 fi if [[ $(boot2docker status) != "running" ]]; then echo "boot2docker VM isn't started. Please run 'boot2docker start'" >&2 exit 1 else # Reach over and set the clock. After sleep/resume the clock will skew. echo "+++ Setting boot2docker clock" boot2docker ssh sudo date -u -D "%Y%m%d%H%M.%S" --set "$(date -u +%Y%m%d%H%M.%S)" >/dev/null fi fi if ! docker info > /dev/null 2>&1 ; then { echo "Can't connect to 'docker' daemon. please fix and retry." echo echo "Possible causes:" echo " - On Mac OS X, boot2docker VM isn't installed or started" echo " - On Mac OS X, docker env variable isn't set approriately. Run:" echo " \$(boot2docker shellinit)" echo " - On Linux, user isn't in 'docker' group. Add and relogin." echo " - Something like 'sudo usermod -a -G docker ${USER-user}'" echo " - RHEL7 bug and workaround: https://bugzilla.redhat.com/show_bug.cgi?id=1119282#c8" echo " - On Linux, Docker daemon hasn't been started or has crashed" } >&2 exit 1 fi KUBE_ROOT_HASH=$(kube::build::short_hash "$KUBE_ROOT") KUBE_BUILD_IMAGE_TAG="build-${KUBE_ROOT_HASH}" KUBE_BUILD_IMAGE="${KUBE_BUILD_IMAGE_REPO}:${KUBE_BUILD_IMAGE_TAG}" KUBE_BUILD_CONTAINER_NAME="kube-build-${KUBE_ROOT_HASH}" KUBE_BUILD_DATA_CONTAINER_NAME="kube-build-data-${KUBE_ROOT_HASH}" DOCKER_MOUNT_ARGS=("${DOCKER_MOUNT_ARGS_BASE[@]}" --volumes-from "${KUBE_BUILD_DATA_CONTAINER_NAME}") } # --------------------------------------------------------------------------- # Utility functions function kube::build::is_osx() { [[ "$(uname)" == "Darwin" ]] } function kube::build::clean_output() { # Clean out the output directory if it exists. if kube::build::build_image_built ; then echo "+++ Cleaning out _output/dockerized/bin/ via docker build image" kube::build::run_build_command bash -c "rm -rf '${REMOTE_OUTPUT_BINPATH}'/*" else echo "!!! Build image not built. Cannot clean via docker build image." fi echo "+++ Removing data container" docker rm "${KUBE_BUILD_DATA_CONTAINER_NAME}" >/dev/null 2>&1 || true echo "+++ Cleaning out local _output directory" rm -rf "${LOCAL_OUTPUT_ROOT}" } # Make sure the _output directory is created and mountable by docker function kube::build::prepare_output() { mkdir -p "${LOCAL_OUTPUT_SUBPATH}" # On RHEL/Fedora SELinux is enabled by default and currently breaks docker # volume mounts. We can work around this by explicitly adding a security # context to the _output directory. # Details: https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Resource_Management_and_Linux_Containers_Guide/sec-Sharing_Data_Across_Containers.html#sec-Mounting_a_Host_Directory_to_a_Container if which selinuxenabled >/dev/null && \ selinuxenabled && \ which chcon >/dev/null ; then if [[ ! $(ls -Zd "${LOCAL_OUTPUT_ROOT}") =~ svirt_sandbox_file_t ]] ; then echo "+++ Applying SELinux policy to '_output' directory. If this fails it may be" echo " because you have root owned files under _output. Delete those and continue" chcon -Rt svirt_sandbox_file_t "${LOCAL_OUTPUT_ROOT}" fi fi } # Detect if a specific image exists # # $1 - image repo name # #2 - image tag function kube::build::docker_image_exists() { [[ -n $1 && -n $2 ]] || { echo "!!! Internal error. Image not specified in docker_image_exists." >&2 exit 2 } # We cannot just specify the IMAGE here as `docker images` doesn't behave as # expected. See: https://github.com/docker/docker/issues/8048 docker images | grep -Eq "^${1}\s+${2}\s+" } # Takes $1 and computes a short has for it. Useful for unique tag generation function kube::build::short_hash() { [[ $# -eq 1 ]] || { echo "!!! Internal error. No data based to short_hash." >&2 exit 2 } local short_hash if which md5 >/dev/null 2>&1; then short_hash=$(md5 -q -s "$1") else short_hash=$(echo -n "$1" | md5sum) fi echo ${short_hash:0:5} } # --------------------------------------------------------------------------- # Building function kube::build::build_image_built() { kube::build::docker_image_exists "${KUBE_BUILD_IMAGE_REPO}" "${KUBE_BUILD_IMAGE_TAG}" } function kube::build::ensure_golang() { kube::build::docker_image_exists golang 1.3 || { [[ ${KUBE_SKIP_CONFIRMATIONS} =~ ^[yY]$ ]] || { echo "You don't have a local copy of the golang docker image. This image is 450MB." read -p "Download it now? [y/n] " -n 1 -r echo [[ $REPLY =~ ^[yY]$ ]] || { echo "Aborting." >&2 exit 1 } } echo "+++ Pulling docker image: golang:${KUBE_BUILD_GOLANG_VERSION}" docker pull golang:${KUBE_BUILD_GOLANG_VERSION} } } # Set up the context directory for the kube-build image and build it. function kube::build::build_image() { local -r build_context_dir="${LOCAL_OUTPUT_IMAGE_STAGING}/${KUBE_BUILD_IMAGE}" local -r source=( api build cmd examples Godeps/_workspace/src Godeps/Godeps.json hack LICENSE pkg plugin README.md test third_party ) kube::build::build_image_cross mkdir -p "${build_context_dir}" tar czf "${build_context_dir}/kube-source.tar.gz" "${source[@]}" kube::version::get_version_vars kube::version::save_version_vars "${build_context_dir}/kube-version-defs" cp build/build-image/Dockerfile ${build_context_dir}/Dockerfile kube::build::docker_build "${KUBE_BUILD_IMAGE}" "${build_context_dir}" } # Build the kubernetes golang cross base image. function kube::build::build_image_cross() { kube::build::ensure_golang local -r build_context_dir="${LOCAL_OUTPUT_ROOT}/images/${KUBE_BUILD_IMAGE}/cross" mkdir -p "${build_context_dir}" cp build/build-image/cross/Dockerfile ${build_context_dir}/Dockerfile kube::build::docker_build "${KUBE_BUILD_IMAGE_CROSS}" "${build_context_dir}" } # Builds the runtime image. Assumes that the appropriate binaries are already # built and in $LOCAL_OUTPUT_BINPATH. function kube::build::run_image() { [[ ${KUBE_BUILD_RUN_IMAGES} =~ ^[yY]$ ]] || return 0 local -r build_context_base="${LOCAL_OUTPUT_IMAGE_STAGING}/${KUBE_RUN_IMAGE_BASE}" # First build the base image. This one brings in all of the binaries. mkdir -p "${build_context_base}" tar czf "${build_context_base}/kube-bins.tar.gz" \ -C "${LOCAL_OUTPUT_BINPATH}/linux/amd64" \ "${KUBE_RUN_IMAGES[@]}" cp -R build/run-images/base/* "${build_context_base}/" kube::build::docker_build "${KUBE_RUN_IMAGE_BASE}" "${build_context_base}" local b for b in "${KUBE_RUN_IMAGES[@]}" ; do local sub_context_dir="${build_context_base}-$b" mkdir -p "${sub_context_dir}" cp -R build/run-images/$b/* "${sub_context_dir}/" kube::build::docker_build "${KUBE_RUN_IMAGE_BASE}-$b" "${sub_context_dir}" done } # Build a docker image from a Dockerfile. # $1 is the name of the image to build # $2 is the location of the "context" directory, with the Dockerfile at the root. function kube::build::docker_build() { local -r image=$1 local -r context_dir=$2 local -ra build_cmd=(docker build -t "${image}" "${context_dir}") echo "+++ Building Docker image ${image}." local docker_output docker_output=$("${build_cmd[@]}" 2>&1) || { cat <&2 +++ Docker build command failed for ${image} ${docker_output} To retry manually, run: ${build_cmd[*]} EOF return 1 } } function kube::build::clean_image() { local -r image=$1 echo "+++ Deleting docker image ${image}" docker rmi ${image} 2> /dev/null || true } function kube::build::clean_images() { kube::build::clean_image "${KUBE_BUILD_IMAGE}" kube::build::clean_image "${KUBE_RUN_IMAGE_BASE}" local b for b in "${KUBE_RUN_IMAGES[@]}" ; do kube::build::clean_image "${KUBE_RUN_IMAGE_BASE}-${b}" done echo "+++ Cleaning all other untagged docker images" docker rmi $(docker images -q --filter 'dangling=true') 2> /dev/null || true } function kube::build::ensure_data_container() { if ! docker inspect "${KUBE_BUILD_DATA_CONTAINER_NAME}" >/dev/null 2>&1; then echo "+++ Creating data container" local -ra docker_cmd=( docker run "${DOCKER_DATA_MOUNT_ARGS[@]}" --name "${KUBE_BUILD_DATA_CONTAINER_NAME}" "${KUBE_BUILD_IMAGE}" true ) "${docker_cmd[@]}" fi } # Run a command in the kube-build image. This assumes that the image has # already been built. This will sync out all output data from the build. function kube::build::run_build_command() { [[ $# != 0 ]] || { echo "Invalid input." >&2; return 4; } kube::build::ensure_data_container kube::build::prepare_output local -a docker_run_opts=( "--name=${KUBE_BUILD_CONTAINER_NAME}" "${DOCKER_MOUNT_ARGS[@]}" ) # If we have stdin we can run interactive. This allows things like 'shell.sh' # to work. However, if we run this way and don't have stdin, then it ends up # running in a daemon-ish mode. So if we don't have a stdin, we explicitly # attach stderr/stdout but don't bother asking for a tty. if [[ -t 0 ]]; then docker_run_opts+=(--interactive --tty) else docker_run_opts+=(--attach=stdout --attach=stderr) fi local -ra docker_cmd=( docker run "${docker_run_opts[@]}" "${KUBE_BUILD_IMAGE}") # Remove the container if it is left over from some previous aborted run docker rm "${KUBE_BUILD_CONTAINER_NAME}" >/dev/null 2>&1 || true "${docker_cmd[@]}" "$@" # Remove the container after we run. '--rm' might be appropriate but it # appears that sometimes it fails. See # https://github.com/docker/docker/issues/3968 docker rm "${KUBE_BUILD_CONTAINER_NAME}" >/dev/null 2>&1 || true } # Test if the output directory is remote (and can only be accessed through # docker) or if it is "local" and we can access the output without going through # docker. function kube::build::is_output_remote() { rm -f "${LOCAL_OUTPUT_SUBPATH}/test_for_remote" kube::build::run_build_command touch "${REMOTE_OUTPUT_BINPATH}/test_for_remote" [[ ! -e "${LOCAL_OUTPUT_BINPATH}/test_for_remote" ]] } # If the Docker server is remote, copy the results back out. function kube::build::copy_output() { if kube::build::is_output_remote; then # When we are on the Mac with boot2docker (or to a remote Docker in any # other situation) we need to copy the results back out. Ideally we would # leave the container around and use 'docker cp' to copy the results out. # However, that doesn't work for mounted volumes currently # (https://github.com/dotcloud/docker/issues/1992). And it is just plain # broken (https://github.com/dotcloud/docker/issues/6483). # # The easiest thing I (jbeda) could figure out was to launch another # container pointed at the same volume, tar the output directory and ship # that tar over stdout. echo "+++ Syncing back _output/dockerized/bin directory from remote Docker" rm -rf "${LOCAL_OUTPUT_BINPATH}" mkdir -p "${LOCAL_OUTPUT_BINPATH}" # The ' /dev/null; then tar=gtar fi local tar_cmd=("$tar" "czf" "${tarfile}" "-C" "${stagingdir}" "kubernetes") if "$tar" --version | grep -q GNU; then tar_cmd=("${tar_cmd[@]}" "--owner=0" "--group=0") else echo " !!! GNU tar not available. User names will be embedded in output and" echo " release tars are not official. Build on Linux or install GNU tar" echo " on Mac OS X (brew install gnu-tar)" fi "${tar_cmd[@]}" } # --------------------------------------------------------------------------- # GCS Release function kube::release::gcs::release() { [[ ${KUBE_GCS_UPLOAD_RELEASE} =~ ^[yY]$ ]] || return 0 kube::release::gcs::verify_prereqs kube::release::gcs::ensure_release_bucket kube::release::gcs::push_images kube::release::gcs::copy_release_tarballs } # Verify things are set up for uploading to GCS function kube::release::gcs::verify_prereqs() { if [[ -z "$(which gsutil)" || -z "$(which gcloud)" ]]; then echo "Releasing Kubernetes requires gsutil and gcloud. Please download," echo "install and authorize through the Google Cloud SDK: " echo echo " https://developers.google.com/cloud/sdk/" return 1 fi if [[ -z "${GCLOUD_ACCOUNT-}" ]]; then GCLOUD_ACCOUNT=$(gcloud auth list 2>/dev/null | awk '/(active)/ { print $2 }') fi if [[ -z "${GCLOUD_ACCOUNT-}" ]]; then echo "No account authorized through gcloud. Please fix with:" echo echo " gcloud auth login" return 1 fi if [[ -z "${GCLOUD_PROJECT-}" ]]; then GCLOUD_PROJECT=$(gcloud config list project | awk '{project = $3} END {print project}') fi if [[ -z "${GCLOUD_PROJECT-}" ]]; then echo "No account authorized through gcloud. Please fix with:" echo echo " gcloud config set project " return 1 fi } # Create a unique bucket name for releasing Kube and make sure it exists. function kube::release::gcs::ensure_release_bucket() { local project_hash project_hash=$(kube::build::short_hash "$GCLOUD_PROJECT") KUBE_GCS_RELEASE_BUCKET=${KUBE_GCS_RELEASE_BUCKET-kubernetes-releases-${project_hash}} KUBE_GCS_RELEASE_PREFIX=${KUBE_GCS_RELEASE_PREFIX-devel/} KUBE_GCS_DOCKER_REG_PREFIX=${KUBE_GCS_DOCKER_REG_PREFIX-docker-reg/} if ! gsutil ls "gs://${KUBE_GCS_RELEASE_BUCKET}" >/dev/null 2>&1 ; then echo "Creating Google Cloud Storage bucket: $KUBE_GCS_RELEASE_BUCKET" gsutil mb -p "${GCLOUD_PROJECT}" "gs://${KUBE_GCS_RELEASE_BUCKET}" fi } function kube::release::gcs::ensure_docker_registry() { local -r reg_container_name="gcs-registry" local -r running=$(docker inspect ${reg_container_name} 2>/dev/null \ | build/json-extractor.py 0.State.Running 2>/dev/null) [[ "$running" != "true" ]] || return 0 # Grovel around and find the OAuth token in the gcloud config local -r boto=~/.config/gcloud/legacy_credentials/${GCLOUD_ACCOUNT}/.boto local refresh_token refresh_token=$(grep 'gs_oauth2_refresh_token =' "$boto" | awk '{ print $3 }') if [[ -z "$refresh_token" ]]; then echo "Couldn't find OAuth 2 refresh token in ${boto}" >&2 return 1 fi # If we have an old one sitting around, remove it docker rm ${reg_container_name} >/dev/null 2>&1 || true echo "+++ Starting GCS backed Docker registry" local -ra docker_cmd=( docker run -d "--name=${reg_container_name}" -e "GCS_BUCKET=${KUBE_GCS_RELEASE_BUCKET}" -e "STORAGE_PATH=${KUBE_GCS_DOCKER_REG_PREFIX}" -e "GCP_OAUTH2_REFRESH_TOKEN=${refresh_token}" -p 127.0.0.1:5000:5000 google/docker-registry ) "${docker[@]}" # Give it time to spin up before we start throwing stuff at it sleep 5 } function kube::release::gcs::push_images() { [[ ${KUBE_BUILD_RUN_IMAGES} =~ ^[yY]$ ]] || return 0 kube::release::gcs::ensure_docker_registry # Tag each of our run binaries with the right registry and push local b image_name for b in "${KUBE_RUN_IMAGES[@]}" ; do image_name="${KUBE_RUN_IMAGE_BASE}-${b}" echo "+++ Tagging and pushing ${image_name} to GCS bucket ${KUBE_GCS_RELEASE_BUCKET}" docker tag "${KUBE_RUN_IMAGE_BASE}-$b" "localhost:5000/${image_name}" docker push "localhost:5000/${image_name}" docker rmi "localhost:5000/${image_name}" done } function kube::release::gcs::copy_release_tarballs() { # TODO: This isn't atomic. There will be points in time where there will be # no active release. Also, if something fails, the release could be half- # copied. The real way to do this would perhaps to have some sort of release # version so that we are never overwriting a destination. local -r gcs_destination="gs://${KUBE_GCS_RELEASE_BUCKET}/${KUBE_GCS_RELEASE_PREFIX}" local gcs_options=() if [[ ${KUBE_GCS_NO_CACHING} =~ ^[yY]$ ]]; then gcs_options=("-h" "Cache-Control:private, max-age=0") fi echo "+++ Copying client tarballs to ${gcs_destination}" # First delete all objects at the destination gsutil -q rm -f -R "${gcs_destination}" >/dev/null 2>&1 || true # Now upload everything in release directory gsutil -m "${gcs_options[@]+${gcs_options[@]}}" cp -r "${RELEASE_DIR}"/* "${gcs_destination}" # TODO(jbeda): Generate an HTML page with links for this release so it is easy # to see it. For extra credit, generate a dynamic page that builds up the # release list using the GCS JSON API. Use Angular and Bootstrap for extra # extra credit. if [[ ${KUBE_GCS_MAKE_PUBLIC} =~ ^[yY]$ ]]; then gsutil acl ch -R -g all:R "${gcs_destination}" >/dev/null 2>&1 fi gsutil ls -lh "${gcs_destination}" }