greenplum configs

README.md

@@ -21,3 +21,4 @@ Building blocks for the following big data project use cases are supported in th
 - Spark
 - Kafka
 - Elasticsearch
+- GreenplumDB

docker_bigdata/Dockerfile

@@ -1,7 +1,7 @@
 # Distributed under the terms of the Modified BSD License.
 
 ARG BASE_NAMESPACE
-ARG BASE_IMG="base"
+ARG BASE_IMG="jdk11"
 FROM ${BASE_NAMESPACE:+$BASE_NAMESPACE/}${BASE_IMG}
 
 LABEL maintainer="haobibo@gmail.com"
@@ -20,7 +20,6 @@ RUN source /opt/utils/script-setup.sh \
  && echo "Install mysql client:" && setup_mysql_client \
  && echo "Install mongosh:" && setup_mongosh_client \
  && echo "Install redis-cli:" && setup_redis_client \
- && echo "Install JDK:" && VERSION_JDK=11 setup_java_base \
  && echo "Install pyflink:" && install_pip /opt/utils/list_install_pip_pyflink.txt \
  && echo "Install pyspark:" && install_pip /opt/utils/list_install_pip_pyspark.txt \
  && echo "Clean up" && list_installed_packages && install__clean

docker_greenplum/Dockerfile

@@ -1,39 +1,69 @@
 # Distributed under the terms of the Modified BSD License.
 ARG BASE_NAMESPACE
 ARG BASE_IMG="base"
+FROM ${BASE_NAMESPACE:+$BASE_NAMESPACE/}${BASE_IMG} AS builder
+
+COPY rootfs /
+
+RUN source /opt/utils/script-utils.sh \
+ && install_apt /opt/utils/install_list_greenplum.apt \
+ && apt-get -qq install -yq --no-install-recommends  gcc g++ bison flex cmake pkg-config ccache ninja-build \
+ && install_tar_gz https://github.com/greenplum-db/gpdb/releases/download/7.0.0-beta.2/7.0.0-beta.2-src-full.tar.gz \
+ && cd /opt/gpdb_src \
+ && PYTHON=/opt/conda/bin/python3 ./configure --prefix=/opt/gpdb --with-perl --with-python --with-libxml --with-gssapi --with-openssl \
+ && sudo make -j16 && sudo make install -j16
+
 FROM ${BASE_NAMESPACE:+$BASE_NAMESPACE/}${BASE_IMG}
 
 LABEL maintainer="haobibo@gmail.com"
 
-COPY work /opt/utils/
+COPY rootfs /
+COPY --from=builder --chown=1000:1000 /opt/gpdb /opt/gpdb
 
-ENV GPDB_HOME="/opt/gpdb" \
-    GPDB_DATA="/data/gpdb" \
-    GPDB_USER="gpadmin"
+ENV GPHOME="/opt/gpdb" \
+    GPDATA="/data/gpdb" \
+    GPUSER="gpadmin"
 
 RUN source /opt/utils/script-utils.sh \
+ && echo "source ${GPHOME}/greenplum_path.sh" >> /etc/profile \
+ && useradd -u 1000 ${GPUSER} -s /bin/bash -d /home/${GPUSER} \
+ && usermod -aG root ${GPUSER} \
+ && echo "${GPUSER} ALL=(ALL) NOPASSWD:ALL" >> /etc/sudoers \
+ && echo "${GPUSER}:${GPUSER}" | chpasswd \
+ && mkdir -pv ${GPHOME}/conf ${GPDATA} /home/${GPUSER}/.ssh \
+ && chown -R ${GPUSER}:${GPUSER} ${GPHOME} ${GPDATA} /home/${GPUSER} \
  && install_apt /opt/utils/install_list_greenplum.apt \
- && useradd ${GPDB_USER} \
- && usermod -aG root ${GPDB_USER} \
- && echo "${GPDB_USER} ALL=(ALL) NOPASSWD:ALL" >> /etc/sudoers \
- && mkdir -p /home/${GPDB_USER}/.ssh \
- && chown -R ${GPDB_USER}:${GPDB_USER} /home/${GPDB_USER} \
- && echo "${GPDB_USER}:${GPDB_USER}" | chpasswd \
- && install_tar_gz https://github.com/greenplum-db/gpdb/releases/download/7.0.0-beta.2/7.0.0-beta.2-src-full.tar.gz \
- && cd /opt/gpdb_src \
- && export PATH="${PATH}":${GPDB_HOME}/bin \
- && PYTHON=/opt/conda/bin/python3 ./configure --prefix=${GPDB_HOME} --with-perl --with-python --with-libxml --with-gssapi --with-openssl \
- && sudo make -j16 && sudo make install -j16 \
- && mkdir -pv ${GPDB_HOME}/conf ${GPDB_DATA} \
- && sudo chown -R ${GPDB_USER}:${GPDB_USER} ${GPDB_HOME} ${GPDB_DATA} \
- && cp ${GPDB_HOME}/docs/cli_help/gpconfigs/gpinitsystem_config ${GPDB_HOME}/conf/gpinitsystem_config \
+ && export PATH="${PATH}:${GPHOME}/bin" && cd ${GPHOME}/bin && ls -alh \
  && PYTHON_SITE=$(python3 -c 'import sys;print(list(filter(lambda s: "site" in s, sys.path))[0])') \
- && sudo ln -s ${GPDB_HOME}/lib/python/* ${PYTHON_SITE}/ \
- && pip install conan psutil pygresql \
- && cp -rf /opt/utils/etc / && rm -rf /opt/utils/etc /opt/gpdb_src \
- && echo "source ${GPDB_HOME}/greenplum_path.sh" >> /etc/profile \
+ && sudo ln -s ${GPHOME}/lib/python/* ${PYTHON_SITE}/ \
+ && pip install -U psutil pygresql pyyaml \
+ && cp ${GPHOME}/docs/cli_help/gpconfigs/gpinitsystem_config ${GPHOME}/conf/gpinitsystem_config \
+ # config sshd
+ && sed -i 's/PermitRootLogin prohibit-password/PermitRootLogin yes/' /etc/ssh/sshd_config \
+ && sed -i -r 's/^.*StrictHostKeyChecking\s+\w+/StrictHostKeyChecking no/' /etc/ssh/ssh_config \
+ && sed 's@session\s*required\s*pam_loginuid.so@session optional pam_loginuid.so@g' -i /etc/pam.d/sshd \
+ && mkdir -pv /run/sshd \
+ && ulimit -n 65536 65536 \
  && echo "Clean up" && list_installed_packages && install__clean
 
-USER ${GPDB_USER}
+USER ${GPUSER}
+RUN [ -e ~/.ssh/id_rsa.pub ] || ssh-keygen -t rsa -b 4096 -N "" -C GreenplumDB -f ~/.ssh/id_rsa \
+ && cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys \
+ && ssh-keygen -A -v \
+ && chmod 600 ~/.ssh/authorized_keys \
+ && chmod +x ${GPHOME}/*.sh
+
+ENV PATH="$PATH:${GPHOME}/bin" \
+    USER=${GPUSER} \
+    LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:${GPHOME}/lib" \
+    COORDINATOR_DIRECTORY="${GPDATA}/coordinator" \
+    COORDINATOR_DATA_DIRECTORY="${GPDATA}/coordinator/gpseg-1" \
+    PGPORT=5432 \
+    PGUSER=gpadmin \
+    PGDATABASE=postgres
+
+WORKDIR /home/${GPUSER}
 EXPOSE 5432 22
-VOLUME ["${GPDB_HOME}"]
+VOLUME ["${GPDATA}", "${GPHOME}/conf"]
+ENTRYPOINT ["tini", "-g", "--"]
+CMD ["/bin/bash", "-c", "${GPHOME}/entrypoint.sh"]

docker_greenplum/README-gpdb-install.md

@@ -2,7 +2,7 @@
 
 本文档介绍GPDB的安装，以目前Github上开源的最新版本为例。
 
-GPDB集群为主从模式，主结点称为master结点或者coordinator结点，从结点成为segment结点。
+GPDB集群为主从模式，主结点称为master结点或者coordinator结点，从结点称为segment结点。
 
 ## 准备工作（均为必须）
 
@@ -32,7 +32,7 @@ ssh-keygen -t rsa -b 4096 -N "" -C `hostname` -f ~/.ssh/id_rsa
 
 在主节点上，也即第1台服务器上，授权通过ssh访问各台服务器：
 建议先安装sshpass: apt-get install sshpass，然后再从第1台服务器上执行下面的命令（最后一个参数为主机名）
-```sshpass -p "P@ssw0rd!" ssh-copy-id -o StrictHostKeyChecking=no gpadmin@KGDB-001```
+```sshpass -p "P@ssw0rd!" ssh-copy-id -o StrictHostKeyChecking=no gpadmin@GPDB-001```
 
 ### 1.3 安装依赖
 
@@ -58,16 +58,16 @@ sudo pip3 install conan
 
 ```txt
 # IP Address    Node Name
-30.23.109.100   KGDB-001
-30.23.109.107   KGDB-002
-30.23.109.109   KGDB-003
+30.23.109.100   GPDB-001
+30.23.109.101   GPDB-002
+30.23.109.102   GPDB-003
 ```
 
 ## 2 编译安装GPDB
 
 在第一台服务器上，以gpadmin用户，从gpdb的源代码编译安装GPDB到/opt/gpdb/
 本例中，下载的源代码来自于gpdb官方代码库的master branch: https://github.com/greenplum-db/gpdb.git
-master branch编译出的版本为基于PostgreSQL 12.0的版本，且该版本可以只使用Python3。
+master branch编译出的版本为基于PostgreSQL 12.x的版本，且该版本可以只使用Python3。
 
 ### 2.1 准备Python依赖
 
@@ -121,15 +121,15 @@ PYTHON_SITE=$(python3 -c 'import sys;print(list(filter(lambda s: "site" in s, sy
 
 ```txt
 # hostlist
-KGDB-001
-KGDB-002
-KDGB-003
+GPDB-001
+GPDB-002
+GPDB-003
 ```
 
 ```txt
 # seg_host
-KDGB-002
-KGDB-003
+GPDB-002
+GPDB-003
 ```
 
 ### 2.6 授权服务器之间SSH互访
@@ -149,7 +149,7 @@ gpscp -f seg_host /opt/conda /opt/gpdb =:~/
 例如通过下面的echo命令来追加到/etc/hosts文件（参照步骤1.4）：
 
 ```shell
-sudo echo "30.23.109.100  KGDB-001" >> /etc/hosts
+sudo echo "30.23.109.100  GPDB-001" >> /etc/hosts
 ```
 
 ### 2.8. 修改Linux系统配置
@@ -223,14 +223,14 @@ mkdir -pv /data/gpdb/primary1 /data/gpdb/primary2 /data/gpdb/mirror1 /data/gpdb/
 
 ```conf
 #集群名称
-ARRAY_NAME="KGDB"
+ARRAY_NAME="GPDB"
 
 SEG_PREFIX="gpseg"
 
 MACHINE_LIST_FILE="/opt/gpdb/conf/seg_host"
 
 # Master结点主机名
-COORDINATOR_HOSTNAME=GP7-001
+COORDINATOR_HOSTNAME=GPDB-001
 
 #master的数据目录
 COORDINATOR_DIRECTORY=/data/gpdb/coordinator

docker_greenplum/README.md

@@ -1,9 +1,20 @@
 ```bash
-docker build -t qpod/gpdb --build-arg "BASE_NAMESPACE=qpod" .
+docker build -t qpod0dev/greenplum --build-arg "BASE_NAMESPACE=qpod" .
 
 docker run -it \
   -p 15432:5432  -p 10022:22 \
-  -v $(pwd):/opt/dev \
-  docker.io/qpod/greenplum \
+  -v /data/database/gpdb:/data/gpdb \
+  -h gpdb-cdw \
+  --name gpdb-cdw \
+  docker.io/qpod0dev/greenplum \
   bash
+
+docker run -d \
+  -p 15432:5432  -p 10022:22 \
+  -v /data/database/gpdb:/data/gpdb \
+  -h gpdb-cdw \
+  --name gpdb-cdw \
+  docker.io/qpod0dev/greenplum
+
+/bin/bash -c ${GPHOME}/entrypoint.sh
 ```

docker_greenplum/example/gpdb-single-node/docker-compose.yml

@@ -0,0 +1,22 @@
+version: "3"
+
+networks:
+  greenplum:
+    ipam:
+      config:
+      - subnet: 10.188.0.0/24
+        gateway: 10.188.0.1
+
+services:
+  cdw:
+    image: "docker.io/qpod/greenplum"
+    container_name: gpdb-cdw
+    hostname: gpdb-cdw
+    ports:
+      - "10022:22"
+      - "15432:5432"
+    tty: true
+    networks:
+      - greenplum
+    volumes:
+    - /data/database/gpdb:/data/gpdb

docker_greenplum/example/gpdb-single-vm/conf/gpinitsystem.conf

@@ -0,0 +1,58 @@
+# reference file:/opt/gpdb/docs/cli_help/gpconfigs/gpinitsystem_config
+
+# Optional - GPDB Cluster Name
+ARRAY_NAME="GPDB"
+
+################################################
+#### REQUIRED PARAMETERS
+################################################
+
+#### Naming convention for utility-generated data directories.
+SEG_PREFIX=gpseg
+
+#### Base number by which primary segment port numbers are calculated.
+PORT_BASE=6000
+
+#### File system location(s) where primary segment data directories will be created.
+#### The number of locations in the list dictate the number of primary segments that will get created per physical host.
+#### (If multiple addresses for a host are listed in the hostfile, the number of segments will be spread evenly across the specified interface addresses).
+declare -a DATA_DIRECTORY=(${GPDATA}/primary1 ${GPDATA}/primary1 ${GPDATA}/primary1 ${GPDATA}/primary2 ${GPDATA}/primary2 ${GPDATA}/primary2)
+
+#### OS-configured hostname or IP address of the coordinator host.
+COORDINATOR_HOSTNAME=gpdb-cdw
+
+#### File system location where the coordinator data directory will be created.
+COORDINATOR_DIRECTORY=/data/gpdb/coordinator
+
+#### Port number for the coordinator instance.
+COORDINATOR_PORT=5432
+
+#### Shell utility used to connect to remote hosts.
+TRUSTED_SHELL=ssh
+
+#### Default server-side character set encoding.
+ENCODING=UNICODE
+
+CHECK_POINT_SEGMENTS=8
+
+################################################
+#### OPTIONAL MIRROR PARAMETERS
+################################################
+
+#### Base number by which mirror segment port numbers are calculated.
+#MIRROR_PORT_BASE=7000
+
+#### File system location(s) where mirror segment data directories will be created.
+#### The number of mirror locations must equal the number of primary locations as specified in the DATA_DIRECTORY parameter.
+#declare -a MIRROR_DATA_DIRECTORY=(${GPDATA}/mirror1 ${GPDATA}/mirror1 ${GPDATA}/mirror1 ${GPDATA}/mirror2 ${GPDATA}/mirror2 ${GPDATA}/mirror2)
+
+
+################################################
+#### OTHER OPTIONAL PARAMETERS
+################################################
+
+#### Create a database of this name after initialization.
+#DATABASE_NAME=name_of_database
+
+#### Specify the location of the host address file here instead of with the -h option of gpinitsystem.
+MACHINE_LIST_FILE="/opt/gpdb/conf/seg_host"

docker_greenplum/example/gpdb-single-vm/conf/seg_host

@@ -0,0 +1,3 @@
+gpdb-cdw
+gpdb-sdw1
+gpdb-sdw2

docker_greenplum/example/gpdb-single-vm/docker-compose.yml

@@ -0,0 +1,51 @@
+version: "3"
+
+networks:
+  greenplum:
+    ipam:
+      config:
+      - subnet: 10.189.0.0/24
+        gateway: 10.189.0.1
+
+services:
+  cdw:
+    image: "docker.io/qpod/greenplum"
+    container_name: gpdb-cdw
+    hostname: gpdb-cdw
+    ports:
+      - "10022:22"
+      - "15432:5432"
+    tty: true
+    networks:
+      - greenplum
+    environment:
+      - ROLE="COORDINATOR"
+    volumes:
+    - ./conf:/opt/gpdb/conf
+    - /data/database/greenplum:/data/gpdb
+
+  sdw1:
+    image: "docker.io/qpod/greenplum"
+    container_name: gpdb-sdw1
+    hostname: gpdb-sdw1
+    tty: true
+    networks:
+      - greenplum
+    environment:
+      - ROLE="SEGMENT"
+    volumes:
+    - ./conf:/opt/gpdb/conf
+    - /data/database/greenplum:/data/gpdb
+
+  sdw2:
+    image: "docker.io/qpod/greenplum"
+    container_name: gpdb-sdw2
+    hostname: gpdb-sdw2
+    tty: true
+    networks:
+      - greenplum
+    environment:
+      - ROLE="SEGMENT"
+    volumes:
+      - ./conf:/opt/gpdb/conf
+      - /data/database/greenplum:/data/gpdb

docker_greenplum/rootfs/etc/security/limits.d/90-greenplum.conf

@@ -0,0 +1,4 @@
+* soft nofile 1048576
+* hard nofile 1048576
+* soft nproc 1048576
+* hard nproc 1048576

docker_greenplum/rootfs/etc/sysctl.conf

@@ -0,0 +1,15 @@
+kernel.shmmax = 5000000000000
+kernel.shmmni = 32768
+kernel.shmall = 40000000000
+kernel.sem = 1000 32768000 1000 32768
+kernel.msgmnb = 1048576
+kernel.msgmax = 1048576
+kernel.msgmni = 32768
+
+net.core.netdev_max_backlog = 80000
+net.core.rmem_default = 2097152
+net.core.rmem_max = 16777216
+net.core.wmem_max = 16777216
+
+vm.overcommit_memory = 2
+vm.overcommit_ratio = 95