Create dockerfile and utility scripts for docker image (#773)

* Add section to readme with info on how to consume docker image
* Docker image automatically starts demo cluster and sources relevant env files

Dockerfile

+FROM pivotaldata/gpdb-devel
+
+WORKDIR /workspace
+
+ADD . gpdb/
+
+WORKDIR gpdb
+
+RUN ./configure --with-python --with-perl --enable-mapreduce --with-libxml --prefix=/usr/local/gpdb
+RUN time make -j4
+RUN make install
+
+RUN chown -R gpadmin:gpadmin /workspace/gpdb
+RUN chown -R gpadmin:gpadmin /usr/local/gpdb

README.md

@@ -204,6 +204,63 @@ make installcheck-bugbuster
   some tests are known to fail with Greenplum. The
   __installcheck-good__ schedule excludes those tests.
 
+## Development with Docker (alpha)
+
+We provide a docker image with all dependencies required to compile and test
+GPDB. You can view the dependency dockerfile at `./docker/base/Dockerfile`.
+The image is hosted on docker hub at `pivotaldata/gpdb-devel`. This docker
+image is currently under heavy development.
+
+Known issues:
+* The `installcheck-good` make target has at least 4 failures, some of which
+  are non-deterministic
+
+### Running regression tests with Docker
+
+1. Create a docker host with 8gb RAM and 4 cores
+    ```bash
+    docker-machine create -d virtualbox --virtualbox-cpu-count 4 --virtualbox-disk-size 50000 --virtualbox-memory 8192 gpdb
+    eval $(docker-machine env gpdb)
+    ```
+
+1. Build your code on gpdb-devel rootfs
+    ```bash
+    cd [path/to/gpdb]
+    docker build .
+    # image beefc4f3 built
+    ```
+    The top level Dockerfile will automatically sync your current working
+    directory into the docker image. This means that any code you are working
+    on will automatically be built and ready for testing in the docker context
+
+1. Log into docker image
+    ```bash
+    docker run -it beefc4f3
+    ```
+
+1. As `gpadmin` user run `installcheck-good`
+    ```bash
+    su gpadmin
+    cd /workspace/gpdb
+    make installcheck-good
+    ```
+
+### Caveats
+
+* No Space Left On Device
+    On macOS the docker-machine vm can periodically become full with unused images.
+    You can clear these images with a combination of docker commands.
+    ```bash
+    # assuming no currently running containers
+    # remove all stopped containers from cache
+    docker ps -aq | xargs -n 1 docker rm
+    # remove all untagged images
+    docker images -aq --filter dangling=true | xargs -n 1 docker rmi
+    ```
+
+    Alternatively you can use the (beta) Native macOS docker client now available
+    in docker 1.12.
+
 ## Glossary
 
 * __QD__

docker/base/Dockerfile

+FROM centos:6
+
+RUN yum -y update
+RUN yum -y groupinstall "Development tools"
+RUN yum -y install ed
+RUN yum -y install readline-devel
+RUN yum -y install zlib-devel
+RUN yum -y install curl-devel
+RUN yum -y install bzip2-devel
+RUN yum -y install python-devel
+RUN yum -y install apr-devel
+RUN yum -y install libevent-devel
+RUN yum -y install openssl-libs openssl-devel
+RUN yum -y install libyaml libyaml-devel
+RUN yum -y install epel-release
+RUN yum -y install htop
+RUN yum -y install ccache
+RUN yum -y install libffi-devel
+RUN yum -y install iproute
+RUN yum -y install net-tools
+RUN yum -y install openssh-server
+RUN yum -y install perl-devel
+RUN yum -y install perl-CPAN
+RUN yum -y install perl-ExtUtils-Embed
+RUN yum -y install libxml2-devel
+
+ADD https://bootstrap.pypa.io/get-pip.py /
+RUN python /get-pip.py
+RUN pip install psutil lockfile paramiko setuptools epydoc
+RUN rm /get-pip.py
+
+# Misc
+RUN yum -y install vim mc psmisc
+
+# cmake 3.0
+WORKDIR /root
+
+ADD http://www.cmake.org/files/v3.0/cmake-3.0.0.tar.gz ./
+RUN tar -zxvf cmake-3.0.0.tar.gz
+WORKDIR cmake-3.0.0
+RUN ./bootstrap
+RUN make
+RUN make install
+ENV PATH=/usr/local/bin:$PATH
+
+# compile GPDB
+
+WORKDIR /workspace
+
+ADD . gpdb/
+ADD ./docker/dotfiles/root.bashrc /root/.bashrc
+ADD ./docker/dotfiles/gpadmin.bashrc /home/gpadmin/.bashrc
+
+WORKDIR gpdb
+
+RUN ./configure --with-python --with-perl --enable-mapreduce --with-libxml --prefix=/usr/local/gpdb
+RUN time make -j4
+RUN make install
+
+RUN ./docker/prep_for_ssh.bash
+
+RUN chown -R gpadmin:gpadmin /home/gpadmin
+
+RUN chmod u+s `which ping`

docker/dotfiles/gpadmin.bashrc

+#!/bin/bash
+
+source /usr/local/gpdb/greenplum_path.sh
+
+if [ ! -e /tmp/demo_sem ]; then
+  touch /tmp/demo_sem
+
+  pushd /workspace/gpdb/gpAux/gpdemo
+    make cluster
+  popd
+fi
+
+source /workspace/gpdb/gpAux/gpdemo/gpdemo-env.sh

docker/dotfiles/root.bashrc

+#!/bin/bash
+
+if [ ! -e /tmp/ssh_sem ]; then
+  touch /tmp/ssh_sem
+  /workspace/gpdb/docker/start_ssh.bash
+fi

docker/prep_for_ssh.bash

+#!/bin/bash
+
+# Based on install_hawq_toolchain.bash in Pivotal-DataFabric/ci-infrastructure repo
+
+setup_ssh_for_user() {
+  local user="${1}"
+  local home_dir
+  home_dir=$(eval echo "~${user}")
+
+  mkdir -p "${home_dir}"/.ssh
+  touch "${home_dir}/.ssh/authorized_keys" "${home_dir}/.ssh/known_hosts" "${home_dir}/.ssh/config"
+  ssh-keygen -t rsa -N "" -f "${home_dir}/.ssh/id_rsa"
+  cat "${home_dir}/.ssh/id_rsa.pub" >> "${home_dir}/.ssh/authorized_keys"
+  chmod 0600 "${home_dir}/.ssh/authorized_keys"
+  cat << 'NOROAMING' >> "${home_dir}/.ssh/config"
+Host *
+  UseRoaming no
+NOROAMING
+  chown -R "${user}" "${home_dir}/.ssh"
+}
+
+transfer_ownership() {
+  chown -R gpadmin:gpadmin /workspace/gpdb
+  chown -R gpadmin:gpadmin /usr/local/gpdb
+}
+
+setup_gpadmin_user() {
+  /usr/sbin/useradd gpadmin
+  echo -e "password\npassword" | passwd gpadmin
+  groupadd supergroup
+  usermod -a -G supergroup gpadmin
+  setup_ssh_for_user gpadmin
+  transfer_ownership
+}
+
+setup_sshd() {
+  test -e /etc/ssh/ssh_host_key || ssh-keygen -f /etc/ssh/ssh_host_key -N '' -t rsa1
+  test -e /etc/ssh/ssh_host_rsa_key || ssh-keygen -f /etc/ssh/ssh_host_rsa_key -N '' -t rsa
+  test -e /etc/ssh/ssh_host_dsa_key || ssh-keygen -f /etc/ssh/ssh_host_dsa_key -N '' -t dsa
+
+  # See https://gist.github.com/gasi/5691565
+  sed -ri 's/UsePAM yes/UsePAM no/g' /etc/ssh/sshd_config
+  # Disable password authentication so builds never hang given bad keys
+  sed -ri 's/PasswordAuthentication yes/PasswordAuthentication no/g' /etc/ssh/sshd_config
+
+  setup_ssh_for_user root
+}
+
+_main() {
+  setup_gpadmin_user
+  setup_sshd
+}
+
+_main "$@"
+

docker/start_ssh.bash

+#!/bin/bash
+
+set -ex
+
+ssh_keyscan_for_user() {
+  local user="${1}"
+  local home_dir
+  home_dir=$(eval echo "~${user}")
+
+  {
+    ssh-keyscan localhost
+    ssh-keyscan 0.0.0.0
+    ssh-keyscan github.com
+  } >> "${home_dir}/.ssh/known_hosts"
+}
+
+/usr/sbin/sshd -D &
+
+ssh_keyscan_for_user gpadmin
+ssh_keyscan_for_user root
+