You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@trafodion.apache.org by sa...@apache.org on 2017/03/03 07:48:37 UTC
[2/3] incubator-trafodion git commit: [TRAFODION-2482] documentation
for python installer
[TRAFODION-2482] documentation for python installer
Project: http://git-wip-us.apache.org/repos/asf/incubator-trafodion/repo
Commit: http://git-wip-us.apache.org/repos/asf/incubator-trafodion/commit/92c80fd3
Tree: http://git-wip-us.apache.org/repos/asf/incubator-trafodion/tree/92c80fd3
Diff: http://git-wip-us.apache.org/repos/asf/incubator-trafodion/diff/92c80fd3
Branch: refs/heads/release2.1
Commit: 92c80fd3750d7b90b147c62aa095b6d63ecf4a50
Parents: 3e0d531
Author: Eason <hf...@gmail.com>
Authored: Fri Feb 24 12:52:46 2017 +0800
Committer: Eason <hf...@gmail.com>
Committed: Fri Feb 24 12:52:46 2017 +0800
----------------------------------------------------------------------
.../src/asciidoc/_chapters/enable_security.adoc | 19 +-
.../src/asciidoc/_chapters/introduction.adoc | 435 +++++---------
.../src/asciidoc/_chapters/prepare.adoc | 110 ++--
.../src/asciidoc/_chapters/quickstart.adoc | 485 +--------------
.../src/asciidoc/_chapters/requirements.adoc | 37 +-
.../src/asciidoc/_chapters/script_install.adoc | 599 ++++++++-----------
.../src/asciidoc/_chapters/script_remove.adoc | 50 +-
.../src/asciidoc/_chapters/script_upgrade.adoc | 286 +--------
8 files changed, 520 insertions(+), 1501 deletions(-)
----------------------------------------------------------------------
http://git-wip-us.apache.org/repos/asf/incubator-trafodion/blob/92c80fd3/docs/provisioning_guide/src/asciidoc/_chapters/enable_security.adoc
----------------------------------------------------------------------
diff --git a/docs/provisioning_guide/src/asciidoc/_chapters/enable_security.adoc b/docs/provisioning_guide/src/asciidoc/_chapters/enable_security.adoc
index 7ce495f..df6ced7 100644
--- a/docs/provisioning_guide/src/asciidoc/_chapters/enable_security.adoc
+++ b/docs/provisioning_guide/src/asciidoc/_chapters/enable_security.adoc
@@ -86,28 +86,25 @@ TBD - A future update will include details on how tickets can be managed at the
=== Kerberos installation
The {project-name} installation scripts automatically determine if Kerberos is enabled on the node. If it is enabled,
-then the environment variable SECURE_HADOOP is set to "Y".
+then the environment variable SECURE_HADOOP is set to "Y".
The following are questions that will be asked related to Kerberos:
* Enter KDC server address, default is []: \u2013 no default
* Enter admin principal (include realm), default is []: - no default
-* Enter fully qualified name for HBase keytab, default is []: - Installer searches for a valid keytab based on the distribution
-* Enter fully qualified name for HDFS keytab, default is []: - Installer searches for a valid keytab based on the distribution
-* Enter max lifetime for Trafodion principal (valid format required), default is [24hours]:
-* Enter renew lifetime for Trafodion principal (valid format required), default is [7days]:
-* Enter Trafodion keytab name, default is []: - Installer determines default name based on the distribution
-* Enter keytab location, default is []: - Installer determins default name based on the distribution
+* Enter password for admin principal:
-NOTE: The {project-name} installer always asked for the KDC admin password when enabling Kerberos independent on whether running in Automated
-of Guided mode. It does not save this password.
+NOTE: KDC admin password will be saved only in configuration file `db_config.bakYYMMDD_HHMM` in installer folder when installation completed.
+You can delete this file for secure perspective.
+NOTE: Keytab files are auto detected by installer in CDH/HDP cluster.
+NOTE: Installer doesn't support kerberos enabled Apache Hadoop for this release.
[[enable-security-ldap]]
== Configuring LDAP
{project-name} does not manage user names and passwords internally but supports authentication via directory servers using
the OpenLDAP protocol, also known as LDAP servers. You can configure the LDAP servers during installation by answering the {project-name}
Installer's prompts. To configure LDAP after installation run the {project-name} security installer directly. Installing LDAP also enables database
-authorization (privilege support).
+authorization (privilege support).
Once authentication and authorization are enabled, {project-name} allows users to be registered in the database and allows privileges
on objects to be granted to users and roles (which are granted to users). {project-name} also supports component-level (or system-level)
@@ -510,7 +507,7 @@ SQL USER CONNECTED user not connected
SQL USER DB NAME SQLUSER1
SQL USER ID 33367
TERMINAL CHARSET ISO88591
-TRANSACTION ID
+TRANSACTION ID
TRANSACTION STATE not in progress
WARNINGS on
```
http://git-wip-us.apache.org/repos/asf/incubator-trafodion/blob/92c80fd3/docs/provisioning_guide/src/asciidoc/_chapters/introduction.adoc
----------------------------------------------------------------------
diff --git a/docs/provisioning_guide/src/asciidoc/_chapters/introduction.adoc b/docs/provisioning_guide/src/asciidoc/_chapters/introduction.adoc
index 0351129..ca3e543 100644
--- a/docs/provisioning_guide/src/asciidoc/_chapters/introduction.adoc
+++ b/docs/provisioning_guide/src/asciidoc/_chapters/introduction.adoc
@@ -50,9 +50,9 @@ respective environment's configuration settings per {project-name} requirements.
<<requirements-trafodion-provisioning-user,{project-name} Provisioning User>> for more information
about the requirements and usage associated with this user ID.
-* *Runtime User*: A Linux-level user under which the {project-name} software runs. This user ID must be registered
-as a user in the Hadoop Distributed File System (HDFS) to store and access objects in HDFS, HBase, and Hive.
-In addition, this user ID requires passwordless access among the nodes where {project-name} is installed.
+* *Runtime User*: A Linux-level user under which the {project-name} software runs, default name is `trafodion`. This user ID must be registered
+as a user in the Hadoop Distributed File System (HDFS) to store and access objects in HDFS, HBase, and Hive.
+In addition, this user ID requires passwordless access among the nodes where {project-name} is installed.
Refer to <<requirements-trafodion-runtime-user,{project-name} Runtime User>> for more information about this user ID.
* *{project-name} Database Users*: {project-name} users are managed by {project-name} security features (grant, revoke, etc.),
@@ -142,13 +142,13 @@ include basic management tasks such as starting and checking the status of the {
* *<<enable-security,Enable Security>>*: Activities related to enabling security features on an already installed
{project-name} installation. These activities include tasks such as adding Kerberos principals and keytabs,
-and setting up the LDAP configuration files.
+and setting up the LDAP configuration files. *Only support in bash installer for now*
[[introduction-provisioning-master-node]]
== Provisioning Master Node
-All provisioning tasks are performed from a single node in the cluster, which must be part
-of the Hadoop environment you're adding {project-name} to. This node is referred to as the
-"*Provisioning Master Node*" in this guide.
+All provisioning tasks are performed from a single node in the cluster, which can be any node
+as long as it has access to the Hadoop environment you're adding {project-name} to.
+This node is referred to as the "*Provisioning Master Node*" in this guide.
The {project-name} Provisioning User must have access to all other nodes from the Provisioning
Master Node in order to perform provisioning tasks on the cluster.
@@ -165,24 +165,23 @@ Next, you unpack the tar file.
```
$ mkdir $HOME/trafodion-installer
$ cd $HOME/trafodion-downloads
-$ tar -zxf apache-trafodion-installer-1.3.0-incubating-bin.tar.gz -C $HOME/trafodion-installer
-$
+$ tar -zxf apache-trafodion-pyinstaller-2.1.0-incubating.tar.gz -C $HOME/trafodion-installer
+$
```
<<<
The {project-name} Installer supports two different modes:
1. *Guided Setup*: Prompts for information as it works through the installation/upgrade process. This mode is recommended for new users.
-2. *Automated Setup*: Required information is provided in a pre-formatted bash-script configuration file, which is provided
-via a command argument when running the {project-name} Installer thereby suppressing all prompts. There is one exception,
-if Kerberos is enabled on the cluster, then you will always be prompted for the KDC admin password. We do not store the
-KDC admin password as part of installation anywhere.
+2. *Automated Setup*: Required information is provided in a pre-formatted ini configuration file, which is provided
+via a command argument when running the {project-name} Installer thereby suppressing all prompts. This ini configuration file only exists
+on the *Provisioning Master Node*, please secure this file or delete it after you installed {project-name} successfully.
+
-A template of the configuration file is available here within the installer directory: `trafodion_config_default`.
+A template of the configuration file is available here within the installer directory: `configs/db_config_default.ini`.
Make a copy of the file in your directory and populate the needed information.
+
Automated Setup is recommended since it allows you to record the required provisioning information ahead of time.
-Refer to <<introduction-trafodion-automated-setup,Automated Setup>> for information about how to
+Refer to <<introduction-trafodion-installer-automated-setup,Automated Setup>> for information about how to
populate this file.
[[introduction-trafodion-installer-usage]]
@@ -191,21 +190,30 @@ populate this file.
The following shows help for the {project-name} Installer.
```
-./trafodion_install --help
-
-This script will install Trafodion. It will create a configuration
-file (if one has not been created), setup of the environment needed
-for Trafodion, configure HBase with Hbase-trx and co-processors needed,
-and install a specified Trafodion build.
+$ ./db_install.py -h
+**********************************
+ Trafodion Installation ToolKit
+**********************************
+Usage: db_install.py [options]
+ Trafodion install main script.
Options:
- --help Print this message and exit
- --accept_license If provided, the user agrees to accept all the
- provisions in the Trafodion license. This allows
- for automation by skipping the display and prompt of
- the Trafodion license.
- --config_file If provided, all install prompts will be
- taken from this file and not prompted for.
+ -h, --help show this help message and exit
+ -c FILE, --config-file=FILE
+ Json format file. If provided, all install prompts
+ will be taken from this file and not prompted for.
+ -u USER, --remote-user=USER
+ Specify ssh login user for remote server,
+ if not provided, use current login user as default.
+ -v, --verbose Verbose mode, will print commands.
+ --silent Do not ask user to confirm configuration result
+ --enable-pwd Prompt SSH login password for remote hosts.
+ If set, 'sshpass' tool is required.
+ --build Build the config file in guided mode only.
+ --reinstall Reinstall Trafodion without restarting Hadoop.
+ --apache-hadoop Install Trafodion on top of Apache Hadoop.
+ --offline Enable local repository for offline installing
+ Trafodion.
```
<<<
@@ -217,6 +225,8 @@ or an upgrade by looking for the {project-name} Runtime User in the `/etc/passwd
* If the user ID doesn't exist, then the {project-name} Installer runs in install mode.
* If the user ID exists, then the {project-name} Installer runs in upgrade mode.
+* If `--reinstall` option is specified, then the {project-name} Installer will not restart Hadoop. It's only available when
+you reinstall the same release version, otherwise an error will be reported during installation.
[[introduction-trafodion-installer-guided-setup]]
@@ -233,16 +243,16 @@ Refer to the following sections for examples:
[[introduction-trafodion-installer-automated-setup]]
=== Automated Setup
-The `--config_file` option runs the {project-name} in Automated Setup mode.
+The `--config-file` option runs the {project-name} in Automated Setup mode.
Before running the {project-name} Installer with this option, you do the following:
-1. Copy the `trafodion_config_default` file.
+1. Copy the `db_config_default.ini` file.
+
*Example*
+
```
-cp trafodion_config_default my_config
+cp configs/db_config_default.ini my_config
```
2. Edit the new file using information you collect in the
@@ -254,258 +264,137 @@ section in the <<prepare,Prepare>> chapter.
*Example*
+
```
-./trafodion_installer --config_file my_config
+./db_install.py --config-file my_config
```
NOTE: Your {project-name} Configuration File contains the password for the {project-name} Runtime User
and for the Distribution Manager. Therefore, we recommend that you secure the file in a manner
-that matches the security policies of your organization.
+that matches the security policies of your organization.
-NOTE: If you are installing {project-name} on a version of Hadoop that has been instrumented with Kerberos,
-you will be asked for a password associated with a Kerberos administrator.
+==== Example: Quick start using a {project-name} Configuration File
+The {project-name} Installer supports a minimum configuration to quick start your installation in two steps.
+1. Copy {project-name} server binary file to your installer directory.
+```
+cp /path/to/apache-trafodion_server-2.1.0-RH-x86_64-incubating.tar.gz python-installer/
+```
+2. Modify configuration file `my_config`, add the Hadoop Distribution Manager URL in `mgr_url`.
+```
+mgr_url = 192.168.0.1:8080
+```
+Once completed, run the {project-name} Installer with the --config-file option.
==== Example: Creating a {project-name} Configuration File
Using the instructions in <<prepare-gather-configuration-information,Gather Configuration Information>>
-in the <<prepare,Prepare>> chapter, you record the following information.
-
-[cols="30%l,50%,20%",options="header"]
-|===
-| ID | Information | Setting
-| ADMIN | Administrator user name for Apache Ambari or Cloudera Manager. | admin
-| ADMIN_PRINCIPAL | Kerberos principal for the KDC admin user including the realm. |
-| BACKUP_DCS_NODES | List of nodes where to start the backup DCS Master components. |
-| CLOUD_CONFIG | Whether you're installing {project-name} on a cloud environment. | N
-| CLOUD_TYPE | What type of cloud environment you're installing {project-name} on. |
-| CLUSTER_NAME | The name of the Hadoop Cluster. | Cluster 1
-| DB_ROOT_NAME | LDAP name used to connect as database root user | trafodion
-| DCS_BUILD | Tar file containing the DCS component. |
-| DCS_PRIMARY_MASTER_NODE | The node where the primary DCS should run. |
-| DCS_SERVER_PARM | Number of concurrent client sessions per node. | 8
-| ENABLE_HA | Whether to run DCS in high-availability (HA) mode. | N
-| EPEL_RPM | Location of EPEL RPM. Specify if you don't have access to the Internet. |
-| FLOATING_IP | IP address if running DCS in HA mode. |
-| HADOOP_TYPE | The type of Hadoop distribution you're installing {project-name} on. | cloudera
-| HBASE_GROUP | Linux group name for the HBASE administrative user. | hbase
-| HBASE_KEYTAB | Kerberos service keytab for HBase admin principal. | Default based on distribution
-| HBASE_USER | Linux user name for the HBASE administrative user. | hbase
-| HDFS_KEYTAB | Kerberos service keytab for HDFS admin principal. | Default based on distribution
-| HDFS_USER | Linux user name for the HDFS administrative user. | hdfs
-| HOME_DIR | Root directory under which the `trafodion` home directory should be created. | /home
-| INIT_TRAFODION | Whether to automatically initialize the {project-name} database. | Y
-| INTERFACE | Interface type used for $FLOATING_IP. |
-| JAVA_HOME | Location of Java 1.7.0_65 or higher (JDK). | /usr/java/jdk1.7.0_67-cloudera
-| KDC_SERVER | Location of Kerberos server for admin access |
-| LDAP_CERT | Full path to TLS certificate. |
-| LDAP_HOSTS | List of nodes where LDAP Identity Store servers are running. |
-| LDAP_ID | List of LDAP unique identifiers. |
-| LDAP_LEVEL | LDAP Encryption Level. |
-| LDAP_PASSWORD | Password for LDAP_USER. |
-| LDAP_PORT | Port used to communicate with LDAP Identity Store. |
-| LDAP_SECURITY | Whether to enable LDAP authentication. | N
-| LDAP_USER | LDAP Search user name. |
-| LOCAL_WORKDIR | The directory where the {project-name} Installer is located. | /home/centos/trafodion-installer/installer
-| MANAGEMENT_ENABLED | Whether your installation uses separate management nodes. | N
-| MANAGEMENT_NODES | The FQDN names of management nodes, if any. |
-| MAX_LIFETIME | Kerberos ticket lifetime for {project-name} principal | 24hours
-| NODE_LIST | The FQDN names of the nodes where {project-name} will be installed. | trafodion-1 trafodion-2
-| PASSWORD | Administrator password for Apache Ambari or Cloudera Manager. | admin
-| RENEW_LIFETIME | Kerberos ticket renewal lifetime for {project-name} principal | 7days
-| REST_BUILD | Tar file containing the REST component. |
-| SECURE_HADOOP | Indicates whether Hadoop has Kerberos enabled | Based on whether Kerberos is enabled for your Hadoop installation
-| TRAF_HOME | Target directory for the {project-name} software. | /home/trafodion/apache-trafodion-1.3.0-incubating-bin
-| START | Whether to start {project-name} after install/upgrade. | Y
-| SUSE_LINUX | Whether your installing {project-name} on SUSE Linux. | false
-| TRAF_PACKAGE | The location of the {project-name} installation package tar file or core installation tar file. | /home/centos/trafodion-download/apache-trafodion-1.3.0-incubating-bin.tar.gz
-| TRAF_KEYTAB | Kerberos keytab for `trafodion` principal. | Default keytab based on distribution
-| TRAF_KEYTAB_DIR | Location of Kerberos keytab for the `trafodion` principal. | Default location based on distribution
-| TRAF_USER | The {project-name} runtime user ID. Must be `trafodion` in this release. | trafodion
-| TRAF_USER_PASSWORD | The password used for the `trafodion:trafodion` user ID. | traf123
-| URL | FQDN and port for the Distribution Manager's REST API. | trafodion-1.apache.org:7180
-|===
-
-Next, you edit `my_config` to contain the following:
+in the <<prepare,Prepare>> chapter, record the information and edit `my_config` to contain the following:
```
-#!/bin/bash
-# @@@ START COPYRIGHT @@@
-#
-# Licensed to the Apache Software Foundation (ASF) under one
-# or more contributor license agreements. See the NOTICE file
-# distributed with this work for additional information
-# regarding copyright ownership. The ASF licenses this file
-# to you 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.
-#
-# @@@ END COPYRIGHT @@@
-
-#====================================================
-# Trafodion Configuration File
-# This file contains default values for the installer.
-
-# Users can also edit this file and provide values for all parameters
-# and then specify this file on the run line of trafodion_install.
-# Example:
-# ./trafodion_install --config_file <Trafodion-config-file>
-# WARNING: This mode is for advanced users!
-#
-#=====================================================
-
-
-#=====================================================
-#Must be set to 'true' if on a SUSE linux system. If on another type of system
-#this must be set to false.
-
-export SUSE_LINUX="false"
-
-# The working directory where Trafodion installer untars files, etc.
-# do not change this unless you really know what you are doing
-export TRAF_WORKDIR="/usr/lib/trafodion"
-
-# This is the directory where the installer scripts were untarred to
-export LOCAL_WORKDIR="/home/centos/trafodion-installer/installer"
-
-# The maximum number of dcs servers, i.e. client connections
-export DCS_SERVERS_PARM="8"
-
-# "true" if this is an upgrade
-export UPGRADE_TRAF="false"
-
-# Trafodion userid, This is the userid the Trafodion instance will run under
-export TRAF_USER="trafodion"
-
-# Trafodion userid's password
-export TRAF_USER_PASSWORD="traf123"
-
-# a blank separated list of nodes in your cluster
-# node names should include full domain names
-#This can not be left blank!
-export NODE_LIST="trafodion-1 trafodion-2"
-
-# count of nodes in node list
-export node_count="2"
-
-# another list of the same nodes in NODE_LIST but specified in a pdsh usable format
-# i.e. "-w centos-cdh[1-6]" or "-w node1 -w node2 -w node3"
-export MY_NODES="-w trafodion-[1-2]"
-
-# the directory prefix for the trafodion userid's $HOME directory
-# i.e. /opt/home, not /opt/home/trafodion
-export HOME_DIR="/home"
-
-#JAVA HOME must be a JDK. Must include FULL Path. Must be 1.7.0_65 or higher.
-
-export JAVA_HOME="/usr/java/jdk1.7.0_67-cloudera"
-
-# If your machine doesn't have external internet access then you must
-# specify the location of the EPEL rpm, otherwise leave blank and it
-# will be installed from the internet
-export EPEL_RPM=""
-
-# full path of the Trafodion package tar file
-export TRAF_PACKAGE="/home/centos/trafodion-download/apache-trafodion-1.3.0-incubating-bin.tar.gz"
-
-# if TRAF_PACKAGE wasn't specified then these two values must be specified
-# TRAF_BUILD is the trafodion_server tar file
-# DCS_BUILD is the DCS tar file
-# REST_BUILD is the REST tar file
-export TRAF_BUILD=""
-export DCS_BUILD=""
-export REST_BUILD=""
-# Either "cloudera" or "hortonworks" (all lowercase)
-export HADOOP_TYPE="cloudera"
-
-# The URL for Cloudera/Hortonworks REST API (i.e. node1.host.com:8080)
-export URL="trafodion-1.apache.org:7180"
-
-# Cloudera/Hortonworks UI admin's userid and password
-export ADMIN="admin"
-export PASSWORD="admin"
-
-# hadoop cluster name
-export CLUSTER_NAME=""
-
-# the Hadoop HDFS userid
-export HDFS_USER="hdfs"
-
-# the Hadoop HBase userid and group
-export HBASE_USER="hbase"
-export HBASE_GROUP="hbase"
-
-# The hadoop HBase service name
-export HBASE="hbase"
-
-# full path of where to install Trafodion to
-# Example is used below. If $HOME_DIR or $TRAF_USER have been changed
-# then this will need to be changed.
-# On an upgrade, it is recommend to choose a different directory.
-# First time install : /home/trafodion/traf
-# On Upgrade: /home/trafodion/traf_<date>
-# By doing this the previous version will remain and allow for an easier rollback.
-export TRAF_HOME="/home/trafodion/apache-trafodion-1.3.0-incubating-bin"
-
-# Start Trafodion after install completes
-export START="Y"
-
-# initialize trafodion after starting
-export INIT_TRAFODION="Y"
-
-# full path to the sqconfig file
-# Default is to leave as is and this file will be created.
-export SQCONFIG=""
-
-#----------------- security configuration information -----------------
-#Enter in Kerberos details if Kerberos is enabled on your cluster
-
-#Indicate Kerberos is enabled
-export SECURE_HADOOP="N"
-
-#Location of Kerberos server for admin access
-export KDC_SERVER=""
-
-#Kerberos Admin principal used to create Trafodion principals and keytabs
-#Please include realm, for example: trafadmin/admin@MYREALM.COM
-export ADMIN_PRINCIPAL=""
-
-#Keytab for HBase admin user, used to grant Trafodion user CRWE privilege
-export HBASE_KEYTAB=""
-
-#Keytab for HDFS admin user, used to create data directories for Trafodion
-export HDFS_KEYTAB=""
-
-#Kerberos ticket defaults for the Trafodion user
-export MAX_LIFETIME="24hours"
-export RENEW_LIFETIME="7days"
-
-#Trafodion keytab information
-export TRAF_KEYTAB=""
-export TRAF_KEYTAB_DIR=""
-
-#Enter in LDAP configuration information
-#Turn on authentication - MUST have existing LDAP configured.
-export LDAP_SECURITY="Y"
-
-#Name of LDAP Config file
-export LDAP_AUTH_FILE="traf_authentication_config_`hostname -s`"
-
-#LDAP name to map to database user DB__ROOT
-DB_ROOT_NAME="trafodion"
-#----------------- end security configuration -----------------
-
-export CONFIG_COMPLETE="true"
+[dbconfigs]
+# NOTICE: if you are using CDH/HDP hadoop distro,
+# you can only specifiy management url address for a quick install
+
+##################################
+# Common Settings
+##################################
+
+# trafodion username and password
+traf_user = trafodion
+traf_pwd = traf123
+# trafodion user's home directory
+home_dir = /home
+# the directory location of trafodion binary
+# if not provided, the default value will be {package_name}-{version}
+traf_dirname =
+
+# trafodion used java(JDK) path on trafodion nodes
+# if not provided, installer will auto detect installed JDK
+java_home =
+
+# cloudera/ambari management url(i.e. http://192.168.0.1:7180 or just 192.168.0.1)
+# if 'http' or 'https' prefix is not provided, the default one is 'http'
+# if port is not provided, the default port is cloudera port '7180'
+mgr_url = 192.168.0.1:8080
+# user name for cloudera/ambari management url
+mgr_user = admin
+# password for cloudera/ambari management url
+mgr_pwd = admin
+# set the cluster number if multiple clusters managed by one Cloudera manager
+# ignore it if only one cluster being managed
+cluster_no = 1
+
+# trafodion tar package file location
+# no need to provide it if the package can be found in current installer's directory
+traf_package =
+# the number of dcs servers on each node
+dcs_cnt_per_node = 4
+
+# scratch file location, seperated by comma if more than one
+scratch_locs = $TRAF_HOME/tmp
+
+# start trafodion instance after installation completed
+traf_start = Y
+
+##################################
+# DCS HA configuration
+##################################
+
+# set it to 'Y' if enable DCS HA
+dcs_ha = N
+# if HA is enabled, provide floating ip, network interface and the hostname of backup dcs master nodes
+dcs_floating_ip =
+# network interface that dcs used
+dcs_interface =
+# backup dcs master nodes, seperated by comma if more than one
+dcs_backup_nodes =
+
+##################################
+# Offline installation setting
+##################################
+
+# set offline mode to Y if no internet connection
+offline_mode = N
+# if offline mode is set, you must provide a local repository directory with all needed RPMs
+local_repo_dir =
+
+##################################
+# LDAP security configuration
+##################################
+
+# set it to 'Y' if enable LDAP security
+ldap_security = N
+# LDAP user name and password to be assigned as DB admin privilege
+db_admin_user = admin
+db_admin_pwd = traf123
+# LDAP user to be assigned DB root privileges (DB__ROOT)
+db_root_user = trafodion
+# if LDAP security is enabled, provide the following items
+ldap_hosts =
+# 389 for no encryption or TLS, 636 for SSL
+ldap_port = 389
+ldap_identifiers =
+ldap_encrypt = 0
+ldap_certpath =
+
+# set to Y if user info is needed
+ldap_userinfo = N
+# provide if ldap_userinfo = Y
+ladp_user =
+ladp_pwd =
+
+##################################
+# Kerberos security configuration
+##################################
+# if kerberos is enabled in your hadoop system, provide below info
+
+# KDC server address
+kdc_server =
+# include realm, i.e. admin/admin@EXAMPLE.COM
+admin_principal =
+# admin password for admin principal, it is used to create trafodion user's principal and keytab
+kdcadmin_pwd =
```
-Once completed, run the {project-name} Installer with the `--config_file` option.
+Once completed, run the {project-name} Installer with the `--config-file` option.
Refer to the following sections for examples:
@@ -518,9 +407,3 @@ Refer to the following sections for examples:
{project-name} stores its provisioning information in the following directories on each node in the cluster:
* `/etc/trafodion`: Configuration information.
-* `/usr/lib/trafodion`: Copies of the files required by the installer.
-
-
-
-
-
http://git-wip-us.apache.org/repos/asf/incubator-trafodion/blob/92c80fd3/docs/provisioning_guide/src/asciidoc/_chapters/prepare.adoc
----------------------------------------------------------------------
diff --git a/docs/provisioning_guide/src/asciidoc/_chapters/prepare.adoc b/docs/provisioning_guide/src/asciidoc/_chapters/prepare.adoc
index c14e79c..34ca1a8 100644
--- a/docs/provisioning_guide/src/asciidoc/_chapters/prepare.adoc
+++ b/docs/provisioning_guide/src/asciidoc/_chapters/prepare.adoc
@@ -101,7 +101,7 @@ are met for each node in the cluster where you intend to install {project-name}.
[[prepare-configure-kerberos]]
== Configure Kerberos
-If your Hadoop installation has enabled Kerberos, then {project-name} needs to have Kerberos enabled. If not,
+If your Hadoop installation has enabled Kerberos, then {project-name} needs to have Kerberos enabled. If not,
then {project-name} will not run. If you plan to enable Kerberos in {project-name}, then you need to have access to a KDC (Kerberos Key Distribution
Center) and administration credentials so you can create the necessary {project-name} principals and keytabs.
@@ -125,76 +125,48 @@ You need to gather/decide information about your environment to aid installation
[cols="25%l,25%,15%l,35%",options="header"]
|===
-| ID^1^ | Information | Default | Notes
-| ADMIN | Administrator user name for Apache Ambari or Cloudera Manager. | admin | A user that can change configuration and restart services via the
-distribution manager's REST API.
-| ADMIN_PRINCIPAL^2^ | Kerberos admin principal to manage principals and keytabs | None | Required if Kerberos is enabled.
-| BACKUP_DCS_NODES | List of nodes where to start the backup DCS Master components. | None | Blank separated FQDN list. Not needed if $ENABLE_HA = N.
-| CLOUD_CONFIG | Whether you're installing {project-name} on a cloud environment. | N | N = bare-metal or VM installation.
-| CLOUD_TYPE | What type of cloud environment you're installing {project-name} on. | None | { AWS \| OpenStack \| Other }. Not applicable for bare-metal or VM installation.
-| CLUSTER_NAME | The name of the Hadoop Cluster. | None | From Apache Ambari or Cloudera Manager.
-| DB_ROOT_NAME^2^ | LDAP name used to connect as database root user | trafodion | Required when LDAP is enabled.
-| DCS_BUILD | Tar file containing the DCS component. | None | Not needed if using a {project-name} package installation tar file.
-| DCS_PRIMARY_MASTER_NODE | The node where the primary DCS should run. | None | The DCS Master handles JDBC and ODBC connection requests.
-| DCS_SERVER_PARM | Number of concurrent client sessions per node. | 16 | This number specifies the concurrent sessions per node to be supported. Each session could require up to 1GB of physical memory. The number can be changed post-installation. For more information,
+| ID | Information | Default | Notes
+| admin_principal | Kerberos admin principal to manage principals and keytabs | None | Required if Kerberos is enabled.
+| cluster_no | Cluster number if multiple clusters managed in ClouderaManager | 1 | Not required in Hortonworks distro
+| dcs_cnt_per_node | Number of concurrent client sessions per node. | 4 | This number specifies the concurrent sessions per node to be supported. Each session could require up to 1GB of physical memory. The number can be changed post-installation. For more information,
refer to the {docs-url}/client_install/index.html[{project-name} Client Installation Guide].
-| ENABLE_HA | Whether to run DCS in high-availability (HA) mode. | N | You need the floating IP address, the interface, and the backup nodes for DCS Master if enabling this feature.
-| EPEL_RPM | Location of EPEL RPM. | None | Specify if you don't have access to the Internet.
-Downloaded automatically by the {project-name} Installer.
-| FLOATING_IP | IP address if running DCS in HA mode. | None | Not needed if $ENABLE_HA = N. An FQDN name or IP address.
-| HADOOP_TYPE | The type of Hadoop distribution you're installing {project-name} on. | None | Lowercase. cloudera or hadoop.
-| HBASE_GROUP | Linux group name for the HBASE administrative user. | hbase | Required in order to provide access to select HDFS directories to this user ID.
-| HBASE_KEYTAB^2^ | HBase credentials used to grant {project-name} CRWE privileges | based on distribution | Required if Kerberos is enabled.
-| HBASE_USER | Linux user name for the HBASE administrative user. | hbase | Required in order to provide access to select HDFS directories to this user ID.
-| HDFS_KEYTAB^2^ | HDFS credentials used to set privileges on HDFS directories. . | based on distribution | Required if Kerberos is enabled.
-| HDFS_USER | Linux user name for the HDFS administrative user. | hdfs | The {project-name} Installer uses `sudo su` to make HDFS
-configuration changes under this user.
-| HOME_DIR | Root directory under which the `trafodion` home directory should be created. | /home | *Example* +
- +
-If the home directory of the `trafodion` user is
-`/opt/home/trafodion`, then specify the root directory as `/opt/home`.
-| INIT_TRAFODION | Whether to automatically initialize the {project-name} database. | N | Applies if $START=Y only.
-| INTERFACE | Interface type used for $FLOATING_IP. | None | Not needed if $ENABLE_HA = N.
-| JAVA_HOME | Location of Java 1.7.0_65 or higher (JDK). | $JAVA_HOME setting | Fully qualified path of the JDK. For example:
-`/usr/java/jdk1.7.0_67-cloudera`
-| KDC_SERVER^2^ | Location of host where Kerberos server exists | None | Required if Kerberos enabled.
-| LDAP_CERT^2^ | Full path to TLS certificate. | None | Required if $LDAP_LEVEL = 1 or 2.
-| LDAP_HOSTS^2^ | List of nodes where LDAP Identity Store servers are running. | None | Blank separated. FQDN format.
-| LDAP_ID^2^ | List of LDAP unique identifiers. | None | Blank separated.
-| LDAP_LEVEL^2^ | LDAP Encryption Level. | 0 | 0: Encryption not used, 1: SSL, 2: TLS
-| LDAP_PASSWORD^2^ | Password for LDAP_USER. | None | If LDAP_USER is required only.
-| LDAP_PORT^2^ | Port used to communicate with LDAP Identity Store. | None | Examples: 389 for no encryption or TLS, 636 for SSL.
-| LDAP_SECURITY^2^ | Whether to enable simple LDAP authentication. | N | If Y, then you need to provide LDAP_HOSTS.
-| LDAP_USER^2^ | LDAP Search user name. | None | Required if you need additional LDAP functionally such as LDAPSearch. If so, must provide LDAP_PASSWORD, too.
-| LOCAL_WORKDIR | The directory where the {project-name} Installer is located. | None | Full path, no environmental variables.
-| MANAGEMENT_ENABLED | Whether your installation uses separate management nodes. | N | Y if using separate management nodes for Apache Ambari or Cloudera Manager.
-| MANAGEMENT_NODES | The FQDN names of management nodes, if any. | None | Provide a blank-separated list of node names.
-| MAX_LIFETIME^2^ | Kerberos ticket lifetime for Trafodion principal | 24hours | Can be specified when Kerberos is enabled.
-| NODE_LIST | The FQDN names of the nodes where {project-name} will be installed. | None | Provide a blank-separated list of node names. The {project-name}
-Provisioning ID must have passwordless and `sudo` access to these nodes.
-| PASSWORD | Administrator password for Apache Ambari or Cloudera Manager. | admin | A user that can change configuration and restart services via the
+| dcs_ha | Whether to run DCS in high-availability (HA) mode. | N | If Y, you need to provide below dcs configurations.
+| db_admin_user | LDAP name used to connect as database admin user | admin | Required when LDAP is enabled.
+| db_root_user | LDAP name used to connect as database root user | trafodion | Required when LDAP is enabled.
+| dcs_backup_nodes | List of nodes where to start the backup DCS Master components. | None | Required when LDAP is enabled. Comma separated FQDN list.
+| dcs_floating_ip | IP address if running DCS in HA mode. | None | Required when LDAP is enabled. An FQDN name or IP address.
+| dcs_interface | Interface type used for dcs_floating_ip. | None | Required when LDAP is enabled. For example, eth0.
+| home_dir | Root directory under which the `trafodion` home directory should be created. | /home | *Example* +
+If the home directory of the `trafodion` user is `/opt/home/trafodion`, then specify the root directory as `/opt/home`.
+| java_home | Location of Java 1.7.0_65 or higher (JDK). | auto detected | Fully qualified path of the JDK. For example: `/usr/java/jdk1.7.0_67-cloudera`
+| kdcadmin_pwd^1^ | Password for kerberos admin principal | None | Should be removed from configuration file or secured after install.
+| kdc_server^1^ | Location of host where Kerberos server exists | None | Required if Kerberos enabled.
+| ldap_security^1^ | Whether to enable simple LDAP authentication. | N | If Y, then you need to provide below ldap configurations.
+| ldap_encrypt^1^ | LDAP Encryption Level. | 0 | 0: Encryption not used, 1: SSL, 2: TLS
+| ldap_certpath ^1^ | Full path to TLS certificate. | None | Required if ldap_encrypt = 1 or 2.
+| ldap_hosts^1^ | List of nodes where LDAP Identity Store servers are running. | None | Comma separated. FQDN format.
+| ldap_identifiers^1^| List of LDAP unique identifiers. | None | Comma separated.
+| ldap_port^1^ | Port used to communicate with LDAP Identity Store. | None | Examples: 389 for no encryption or TLS, 636 for SSL.
+| ldap_userinfo | Whether to use LDAP Search user name. | N | If Y, then you need to provide ldap_user and ldap_pwd.
+| ladp_user^1^ | LDAP Search user name. | None | Required if you need additional LDAP functionally such as LDAPSearch. If so, must provide ldap_pwd too.
+| ladp_pwd^1^ | Password for ldap_user. | None | If ldap_userinfo is required.
+| local_repo_dir | folder location of Trafodion local repository | None | Required if offline_mode = Y. A local folder with all trafodion rpm dependencies and repodata. For example: `/opt/trafodion_repo`
+| mgr_url | FQDN and port for the Distribution Manager's REST API. | None | Include `http://` or `https://` as applicable. If no prefix, default is `http://`.
+Specify in the form: `<IP-address>:<port>` or `<node name>:<port>` Example: `https://vm-1.yourcompany.local:8080`
+| mgr_user | Administrator user name for Apache Ambari or Cloudera Manager. | admin | A user that can change configuration and restart services via the
+distribution manager's REST API.
+| mgr_pwd | Administrator password for Apache Ambari or Cloudera Manager. | admin | A user that can change configuration and restart services via the
distribution manager's REST API.
-| RENEW_LIFETIME^2^ | Number times Kerberos ticket is for the Trafodion principal | 7days | Can be specified when Kerberos is enabled.
-| REST_BUILD | Tar file containing the REST component. | None | Not needed if using a {project-name} package installation tar file.
-| SECURE_HADOOP^2^ | Indicates whether Hadoop has enabled Kerberos | Y only if Kerberos enabled | Based on whether Kerberos is enabled for your Hadoop installation
-| TRAF_HOME | Target directory for the {project-name} software. | $HOME_DIR/trafodion | {project-name} is installed in this directory on all nodes in `$NODE_LIST`.
-| START | Whether to start {project-name} after install/upgrade. | N |
-| SUSE_LINUX | Whether your installing {project-name} on SUSE Linux. | false | Auto-detected by the {project-name} Installer.
-| TRAF_KEYTAB^2^ | Name to use when specifying {project-name} keytab | based on distribution | Required if Kerberos is enabled.
-| TRAF_KEYTAB_DIR^2^ | Location of {project-name} keytab | based on distribution | Required if Kerberos is enabled.
-| TRAF_PACKAGE | The location of the {project-name} installation package tar file or core installation tar file. | None | The package file contains the {project-name} server,
-DCS, and REST software while the core installation file contains the {project-name} server software only. If you're using a core installation file, then you need to
-record the location of the DCS and REST installation tar files, too. Normally, you perform {project-name} provisioning using a {project-name} package installation tar file.
-| TRAF_USER | The {project-name} runtime user ID. | trafodion | Must be `trafodion` in this release.
-| TRAF_USER_PASSWORD | The password used for the `trafodion:trafodion` user ID. | traf123 | Must be 6-8 characters long.
-| URL | FQDN and port for the Distribution Manager's REST API. | None | Include `http://` or `https://` as applicable. Specify in the form:
-`<IP-address>:<port>` or `<node name>:<port>` Example: `https://susevm-1.yourcompany.local:8080`
+| offline_mode | Whether to install Trafodion without internet connection. | N | If Y, then you need to provide local directory in local_repo_dir
+| scratch_locs | Overflow scratch files location for large queries that cannot fit in memory. | $TRAF_HOME/tmp | Comma seperated if more than one folder, it should be set in a large disk
+| traf_dirname | Target folder name for the {project-name} software. | apache-trafodion-{version} | {project-name} is installed in this directory under `$HOME` on all nodes in `$NODE_LIST`.
+| traf_package | The location of the {project-name} server package tar file. | auto detected in installer folder | The package file contains the {project-name} server, DCS, and REST software
+| traf_pwd | The password used for the {project-name} runtime user ID. | traf123 | Must be 6-8 characters long.
+| traf_start | Whether to start {project-name} after install/upgrade. | Y
+| traf_user | The {project-name} runtime user ID. | trafodion | Must be `trafodion` in this release.
|===
-1. The ID matches the environmental variables used in the {project-name} Installation configuration file. Refer to <<install-trafodion-installer,{project-name} Installer>>
-for more information.
-2. Refer to <<enable-security,Enable Security>> for more information about these security settings.
-
+1. Refer to <<enable-security,Enable Security>> for more information about these security settings.
<<<
[[prepare-install-required-software-packages]]
@@ -212,13 +184,13 @@ If none of these situations exist, then we highly recommend that you use the {pr
You perform this step as a user with `root` or `sudo` access.
-Install the packages listed in <<requirements-software-packages,Software Packages>> above on all nodes in the cluster.
+Install the packages listed in <<requirements-software-packages,Software Packages>> above on all nodes in the cluster.
<<<
[[prepare-download-trafodion-binaries]]
== Download {project-name} Binaries
-You download the {project-name} binaries from the {project-name} {download-url}[Download] page.
+You download the {project-name} binaries from the {project-name} {download-url}[Download] page.
Download the following packages:
Command-line Installation
http://git-wip-us.apache.org/repos/asf/incubator-trafodion/blob/92c80fd3/docs/provisioning_guide/src/asciidoc/_chapters/quickstart.adoc
----------------------------------------------------------------------
diff --git a/docs/provisioning_guide/src/asciidoc/_chapters/quickstart.adoc b/docs/provisioning_guide/src/asciidoc/_chapters/quickstart.adoc
index 8abb145..fbdac4f 100644
--- a/docs/provisioning_guide/src/asciidoc/_chapters/quickstart.adoc
+++ b/docs/provisioning_guide/src/asciidoc/_chapters/quickstart.adoc
@@ -48,60 +48,55 @@ NOTE: You can download and install the {project-name} Clients once you've instal
*Example*
+Download the Trafodion Installer and Server binaries:
```
$ mkdir $HOME/trafodion-download
$ cd $HOME/trafodion-download
$ # Download the Trafodion Installer binaries
-$ wget http://apache.cs.utah.edu/incubator/trafodion/trafodion-1.3.0.incubating/apache-trafodion-installer-1.3.0-incubating-bin.tar.gz
+$ wget http://apache.cs.utah.edu/incubator/trafodion/trafodion-2.1.0.incubating/apache-trafodion-pyinstaller-2.1.0-incubating.tar.gz
Resolving http://apache.cs.utah.edu... 192.168.1.56
Connecting to http://apache.cs.utah.edu|192.168.1.56|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 68813 (67K) [application/x-gzip]
-Saving to: "apache-trafodion-installer-1.3.0-incubating-bin.tar.gz"
+Saving to: "apache-trafodion-pyinstaller-2.1.0-incubating.tar.gz"
100%[=====================================================================================================================>] 68,813 124K/s in 0.5s
-2016-02-14 04:19:42 (124 KB/s) - "apache-trafodion-installer-1.3.0-incubating-bin.tar.gz" saved [68813/68813]
-```
+2016-02-14 04:19:42 (124 KB/s) - "apache-trafodion-pyinstaller-2.1.0-incubating.tar.gz" saved [68813/68813]
-<<<
-```
-$ # Download the Trafodion Server binaries
-$ wget http://apache.cs.utah.edu/incubator/trafodion/trafodion-1.3.0.incubating/apache-trafodion-1.3.0-incubating-bin.tar.gz
+$ wget http://apache.cs.utah.edu/incubator/trafodion/trafodion-2.1.0.incubating/apache-trafodion_server-2.1.0-RH-x86_64-incubating.tar.gz
Resolving http://apache.cs.utah.edu... 192.168.1.56
Connecting to http://apache.cs.utah.edu|192.168.1.56|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 214508243 (205M) [application/x-gzip]
-Saving to: "apache-trafodion-1.3.0-incubating-bin.tar.gz"
+Saving to: "apache-trafodion_server-2.1.0-RH-x86_64-incubating.tar.gz"
100%[=====================================================================================================================>] 214,508,243 3.90M/s in 55s
-2016-02-14 04:22:14 (3.72 MB/s) - "apache-trafodion-1.3.0-incubating-bin.tar.gz" saved [214508243/214508243]
+2016-02-14 04:22:14 (3.72 MB/s) - "apache-trafodion_server-2.1.0-RH-x86_64-incubating.tar.gz" saved [214508243/214508243]
$ ls -l
-total 209552
--rw-rw-r-- 1 centos centos 214508243 Jan 12 20:10 apache-trafodion-1.3.0-incubating-bin.tar.gz
--rw-rw-r-- 1 centos centos 68813 Jan 12 20:10 apache-trafodion-installer-1.3.0-incubating-bin.tar.gz
+-rw-rw-r--. 1 centos centos 74237 Feb 13 14:53 apache-trafodion_pyinstaller-2.1.0-incubating.tar.gz
+-rw-rw-r--. 1 centos centos 183114066 Feb 10 22:34 apache-trafodion_server-2.1.0-RH-x86_64-incubating.tar.gz
$
```
[[quickstart-unpack-installer]]
-== Unpack Installer
+== Unpack Installer and Server package
The first step in the installation process is to unpack the {project-name} Installer tar file.
+{project-name} server package tar file can be auto detected by installer if put it in installer's folder.
*Example*
```
$ mkdir $HOME/trafodion-installer
$ cd $HOME/trafodion-downloads
-$ tar -zxf apache-trafodion-installer-1.3.0-incubating-bin.tar.gz -C $HOME/trafodion-installer
-$ ls $HOME/trafodion-installer/installer
-bashrc_default tools traf_config_check trafodion_apache_hadoop_install traf_package_setup
-build-version-1.3.0.txt traf_add_user traf_config_setup trafodion_config_default traf_setup
-dcs_installer traf_apache_hadoop_config_setup traf_create_systemdefaults trafodion_install traf_sqconfig
-rest_installer traf_authentication_conf_default traf_getHadoopNodes trafodion_license traf_start
-setup_known_hosts.exp traf_cloudera_mods98 traf_hortonworks_mods98 trafodion_uninstaller
+$ tar -zxf apache-trafodion-pyinstaller-2.1.0-incubating.tar.gz -C $HOME/trafodion-installer
+$ cp -f apache-trafodion_server-2.1.0-RH-x86_64-incubating.tar.gz $HOME/trafodion-installer
+$ ls $HOME/trafodion-installer/python-installer
+apache-trafodion_server-2.1.0-RH-x86_64-incubating.tar.gz db_install.py DISCLAIMER LICENSE prettytable.py scripts
+configs db_uninstall.py discovery.py NOTICE README.md
$
```
@@ -110,19 +105,15 @@ $
Collect/decide the following information:
-=== Location of {project-name} Server-Side Binary
-You need the fully-qualified name of the {project-name} server-side binary.
+=== Java Location
-*Example*
+Java location can be automatically detected by installer. You need to provide the java location only if installer cannot detect it.
-```
-/home/trafodion-downloads/apache-trafodion-installer-1.3.0-incubating-bin.tar.gz
-```
-
-=== Java Location
+How to detect java location manually:
-You need to record the location of the Java. For example, use `ps -ef | grep java | grep hadoop | grep hbase` to determine what version HBase is running.
+1. Login to trafodion' node
+2. Use `ps -ef | grep java | grep hadoop | grep hbase` to determine what version HBase is running.
*Example*
@@ -136,7 +127,9 @@ The Java location is: `/usr/jdk64/jdk1.7.0_67`
<<<
=== Data Nodes
-{project-name} is installed on all data nodes in your Hadoop cluster. You need to record the fully-qualified domain name node for each node.
+{project-name} is installed on all data nodes in your Hadoop cluster. Data nodes can be automatically detected by installer while installing on a HDP/CDH cluster.
+
+You need to record hostname for each node when you install {project-name} on Apache Hadoop.
For example, refer to `/etc/hosts`.
*Example*
@@ -146,29 +139,16 @@ $ cat /etc/hosts
127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6
-172.31.56.238 ip-172-31-56-238.ec2.internal node01
-172.31.61.110 ip-172-31-61-110.ec2.internal node02
-172.31.57.143 ip-172-31-57-143.ec2.internal node03
-```
-
-Record the node names in a space-separated list.
-
-*Example*
-
-```
-ip-172-31-56-238.ec2.internal ip-172-31-61-110.ec2.internal ip-172-31-57-143.ec2.internal
+172.31.56.238 node-1.internal node-1
+172.31.61.110 node-2.internal node-2
```
-=== {project-name} Runtime User Home Directory
-
-The Installer creates the `trafodion` user ID. You need to decide the home directory for this user.
-
-The default is: `/home`
+Record the node names in a comma-separated list `node-1, node-2` or in regular expression mode `node-[1-2]`
=== Distribution Manager URL
The Installer interacts with the Distribution Manager (for example, Apache Ambari or Cloudera Manager) to modify the
-Hadoop configuration.
+Hadoop configuration.
*Example*
@@ -182,412 +162,7 @@ http://myhost.com:8080
[[quickstart-run-installer]]
== Run Installer
-You run the Installer once you've collected the base information as described in
+You run the Installer once you've collected the base information as described in
<<quickstart-collect-information, Collect Information>> above.
-The following example shows a guided install of {project-name} on a three-node Hortonworks Hadoop cluster.
-
-NOTE: By default, the {project-name} Installer invokes `sqlci` so that you can enter the `initialize trafodion;` command.
-This is shown in the example below.
-
-*Example*
-
-1. Run the {project-name} Installer in guided mode.
-+
-```
-$ cd $HOME/trafodion-installer/installer
-$ ./trafodion_install 2>&1 | tee install.log
-******************************
- TRAFODION INSTALLATION START
-******************************
-
-***INFO: testing sudo access
-***INFO: Log file located at /var/log/trafodion/trafodion_install_2016-06-30-21-02-38.log
-***INFO: Config directory: /etc/trafodion
-***INFO: Working directory: /usr/lib/trafodion
-
-************************************
- Trafodion Configuration File Setup
-************************************
-
-***INFO: Please press [Enter] to select defaults.
-
-Is this a cloud environment (Y/N), default is [N]: N
-Enter trafodion password, default is [traf123]:
-Enter list of data nodes (blank separated), default []: ip-172-31-56-238.ec2.internal ip-172-31-61-110.ec2.internal ip-172-31-57-143.ec2.internal
-Do you h ave a set of management nodes (Y/N), default is N: N
-Enter Trafodion userid's home directory prefix, default is [/home]: /opt
-Specify location of Java 1.7.0_65 or higher (JDK), default is []: /usr/jdk64/jdk1.7.0_67
-Enter full path (including .tar or .tar.gz) of trafodion tar file []: /home/trafodion-downloads/apache-trafodion_server-2.0.1-incubating.tar.gz
-Enter Backup/Restore username (can be Trafodion), default is [trafodion]:
-Specify the Hadoop distribut ion installed (1: Cloudera, 2: Hortonworks, 3: Other): 2
-Enter Hadoop admin username, default is [admin]: Enter Hadoop admin pas sword, default is [admin]:
-Enter full Hadoop external network URL:port (include 'http://' or 'https://), default is []: http://ip-172-31-56-238.ec2.internal:8080
-Enter HDFS username or username running HDFS, default is [hdfs]:
-Enter HBase username or username running HBase, default is [hbase]:
-Enter HBase group, default is [hbase]:
-Enter Zookeeper username or username running Zookeeper, default is [zookeeper]:
-Enter directory to install trafodion to, default is [/opt/trafodion/apache-trafodion_server-2.0.1-incubating]:
-Start Trafodion after install (Y/N), default is Y:
-Total number of client connections per cluster, default [24]: 96
-Enter the node of primary DcsMaste r, default [ip-172-31-56-238.ec2.internal]:
-Enable High Availability (Y/N), default is N:
-Enable simple LDAP security (Y/N), d efault is N:
-***INFO: Trafodion configuration setup complete
-***INFO: Trafodion Configuration File Check
-***INFO: Testing sudo access on node ip-172-31-56-238
-***INFO: Testing sudo access on node ip-172-31-61-110
-***INFO: Testing sudo access on node ip-172-31-57-143
-***INFO: Testing ssh on ip-172-31-56-238
-***INFO: Testing ssh on ip-172-31-61-110
-***INFO: Testing ssh on ip-172-31-57-143
-#!/bin/bash
-#
-# @@@ START COPYRIGHT @@@
-#
-# Licensed to the Apache Software Foundation (ASF) under one
-# or more contributor license agreements. See the NOTICE file
-# distributed with this work for additional information
-# regarding copyright ownership. The ASF licenses this file
-# to you 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.
-#
-.
-.
-.
-9. Accepting Warranty or Additional Liability. While redistributing
-the Work or Derivative Works thereof, You may choose to offer, and
-charge a fee for, acceptance of support, warranty, indemnity, or
-other liability obligations and/or rights consistent with this
-License. However, in accepting such obligations, You may act only
-on Your own behalf and on Your sole responsibility, not on behalf
-of any other Contributor, and only if You agree to indemnify, defend,
-and hold each Contributor harmless for any liability incurred by,
-or claims asserted against, such Contributor by reason of your
-accepting any such warranty or additional liability.
-
-END OF TERMS AND CONDITIONS
-
-BY TYPING "ACCEPT" YOU AGREE TO THE TERMS OF THIS AGREEMENT: ***INFO: testing sudo access
-***INFO: Starting Trafodion Package Setup (2016-06-30-21-06-40)
-***INFO: Installing required packages
-***INFO: Log file located in /var/log/trafodion
-***INFO: ... pdsh on node ip-172-31-56-238
-***INFO: ... pdsh on node ip-172-31-61-110
-***INFO: ... pdsh on node ip-172-31-57-143
-***INFO: Checking if apr is installed ...
-***INFO: Checking if apr-util is installed ...
-***INFO: Checking if sqlite is installed ...
-***INFO: Checking if expect is installed ...
-***INFO: Checking if perl-DBD-SQLite* is installed ...
-***INFO: Checking if protobuf is installed ...
-***INFO: Checking if xerces-c is installed ...
-***INFO: Checking if perl-Params-Validate is installed ...
-***INFO: Checking if perl-Time-HiRes is installed ...
-***INFO: Checking if gzip is installed ...
-***INFO: Checking if lzo is installed ...
-***INFO: Checking if lzop is installed ...
-***INFO: Checking if unzip is installed ...
-***INFO: modifying limits in /usr/lib/trafodion/trafodion.conf on all nodes
-***INFO: create Trafodion userid "trafodion"
-***INFO: Trafodion userid's (trafodion) home directory: /opt/trafodion
-***INFO: testing sudo access
-Generating public/private rsa key pair.
-Created directory '/opt/trafodion/.ssh'.
-Your identification has been saved in /opt/trafodion/.ssh/id_rsa.
-Your public key has been saved in /opt/trafodion/.ssh/id_rsa.pub.
-The key fingerprint is:
-12:59:ab:d7:59:a2:0e:e8:38:1c:e9:e1:86:f6:18:23 trafodion@ip-172-31-56-238
-The key's randomart image is:
-+--[ RSA 2048]----+
-| . |
-| o . |
-| o . . . |
-| . . o o + |
-| + . + S o |
-| = = = |
-|E+B . . |
-|o.=. |
-| . . |
-+-----------------+
-***INFO: creating .bashrc file
-***INFO: Setting up userid trafodion on all other nodes in cluster
-***INFO: Creating known_hosts file for all nodes
-ip-172-31-56-238
-ip-172-31-56-238 ip-172-31-61-110 ip-172-31-57-143
-ip-172-31-61-110
-ip-172-31-56-238 ip-172-31-61-110 ip-172-31-57-143
-ip-172-31-57-143
-ip-172-31-56-238 ip-172-31-61-110 ip-172-31-57-143
-***INFO: trafodion user added successfully
-***INFO: Trafodion environment setup completed
-***INFO: creating sqconfig file
-***INFO: Reserving DCS ports
-
-***INFO: Creating trafodion sudo access file
-
-
-******************************
- TRAFODION MODS
-******************************
-
-***INFO: Hortonworks installed will run traf_hortonworks_mods
-***INFO: copying hbase-trx-hdp2_3-*.jar to all nodes
-***INFO: hbase-trx-hdp2_3-*.jar copied correctly! Huzzah.
-USERID=admin
-PASSWORD=admin
-PORT=:8080
-{
- "resources" : [
- {
- "href" : "http://ip-172-31-56-238.ec2.internal:8080/api/v1/clusters/trafodion/configurations/service_config_versions?ser
-vice_name=HBASE&service_config_version=2",
-.
-.
-.
- {
- "href" : "http://ip-172-31-56-238.ec2.internal:8080/api/v1/clusters/trafodion/requests/12/tasks/128",
- "Tasks" : {
- "cluster_name" : "trafodion",
- "id" : 128,
- "request_id" : 12,
- "stage_id" : 2
- }
- },
- {
- "href" : "http://ip-172-31-56-238.ec2.internal:8080/api/v1/clusters/trafodion/requests/12/tasks/129",
- "Tasks" : {
- "cluster_name" : "trafodion",
- "id" : 129,
- "request_id" : 12,
- "stage_id" : 2
- }
- },
- {
- "href" : "http://ip-172-31-56-238.ec2.internal:8080/api/v1/clusters/trafodion/requests/12/tasks/130",
- "Tasks" : {
- "cluster_name" : "trafodion",
- "id" : 130,
- "request_id" : 12,
- "stage_id" : 2
- }
- }
- ],
- "stages" : [
- {
- "href" : "http://ip-172-31-56-238.ec2.internal:8080/api/v1/clusters/trafodion/requests/12/stages/0",
- "Stage" : {
- "cluster_name" : "trafodion",
- "request_id" : 12,
- "stage_id" : 0
- }
- },
- {
- "href" : "http://ip-172-31-56-238.ec2.internal:8080/api/v1/clusters/trafodion/requests/12/stages/1",
- "Stage" : {
- "cluster_name" : "trafodion",
- "request_id" : 12,
- "stage_id" : 1
- }
- },
- {
- "href" : "http://ip-172-31-56-238.ec2.internal:8080/api/v1/clusters/trafodion/requests/12/stages/2",
- "Stage" : {
- "cluster_name" : "trafodion",
- "request_id" : 12,
- "stage_id" : 2
- }
- }
- ]
-}***INFO: ...polling every 30 seconds until HBase start is completed.
-***INFO: HBase restart completed
-***INFO: Setting HDFS ACLs for snapshot scan support
-cp: `trafodion_config' and `/home/trafinstall/trafodion-2.0.1/installer/trafodion_config' are the same file
-***INFO: Trafodion Mods ran successfully.
-
-******************************
- TRAFODION CONFIGURATION
-******************************
-
-/usr/lib/trafodion/installer/..
-/opt/trafodion/apache-trafodion_server-2.0.1-incubating
-***INFO: untarring file to /opt/trafodion/apache-trafodion_server-2.0.1-incubating
-***INFO: modifying .bashrc to set Trafodion environment variables
-***INFO: copying .bashrc file to all nodes
-***INFO: copying sqconfig file (/opt/trafodion/sqconfig) to /opt/trafodion/apache-trafodion_server-2.0.1-incubating/sql/script
-s/sqconfig
-***INFO: Creating /opt/trafodion/apache-trafodion_server-2.0.1-incubating directory on all nodes
-***INFO: Start of DCS install
-***INFO: DCS Install Directory: /opt/trafodion/apache-trafodion_server-2.0.1-incubating/dcs-2.0.1
-***INFO: modifying /opt/trafodion/apache-trafodion_server-2.0.1-incubating/dcs-2.0.1/conf/dcs-env.sh
-***INFO: modifying /opt/trafodion/apache-trafodion_server-2.0.1-incubating/dcs-2.0.1/conf/dcs-site.xml
-***INFO: creating /opt/trafodion/apache-trafodion_server-2.0.1-incubating/dcs-2.0.1/conf/servers file
-***INFO: End of DCS install.
-***INFO: Start of REST Server install
-***INFO: Rest Install Directory: /opt/trafodion/apache-trafodion_server-2.0.1-incubating/rest-2.0.1
-***INFO: modifying /opt/trafodion/apache-trafodion_server-2.0.1-incubating/rest-2.0.1/conf/rest-site.xml
-***INFO: End of REST Server install.
-***INFO: starting sqgen
-ip-172-31-56-238,ip-172-31-57-143,ip-172-31-61-110
-
-Creating directories on cluster nodes
-/usr/bin/pdsh -R exec -w ip-172-31-56-238,ip-172-31-57-143,ip-172-31-61-110 -x ip-172-31-56-238 ssh -q -n %h mkdir -p /opt/tra
-fodion/apache-trafodion_server-2.0.1-incubating/etc
-/usr/bin/pdsh -R exec -w ip-172-31-56-238,ip-172-31-57-143,ip-172-31-61-110 -x ip-172-31-56-238 ssh -q -n %h mkdir -p /opt/tra
-fodion/apache-trafodion_server-2.0.1-incubating/logs
-/usr/bin/pdsh -R exec -w ip-172-31-56-238,ip-172-31-57-143,ip-172-31-61-110 -x ip-172-31-56-238 ssh -q -n %h mkdir -p /opt/tra
-fodion/apache-trafodion_server-2.0.1-incubating/tmp
-/usr/bin/pdsh -R exec -w ip-172-31-56-238,ip-172-31-57-143,ip-172-31-61-110 -x ip-172-31-56-238 ssh -q -n %h mkdir -p /opt/tra
-fodion/apache-trafodion_server-2.0.1-incubating/sql/scripts
-
-Generating SQ environment variable file: /opt/trafodion/apache-trafodion_server-2.0.1-incubating/etc/ms.env
-
-Note: Using cluster.conf format type 2.
-
-Generating SeaMonster environment variable file: /opt/trafodion/apache-trafodion_server-2.0.1-incubating/etc/seamonster.env
-
-
-Generated SQ startup script file: ./gomon.cold
-Generated SQ startup script file: ./gomon.warm
-Generated SQ cluster config file: /opt/trafodion/apache-trafodion_server-2.0.1-incubating/tmp/cluster.conf
-Generated SQ Shell file: sqshell
-Generated RMS Startup file: rmsstart
-Generated RMS Stop file: rmsstop
-Generated RMS Check file: rmscheck.sql
-Generated SSMP Startup file: ssmpstart
-Generated SSMP Stop file: ssmpstop
-Generated SSCP Startup file: sscpstart
-Generated SSCP Stop file: sscpstop
-
-
-Copying the generated files to all the nodes in the cluster
-.
-.
-.
-SQ Startup script (/opt/trafodion/apache-trafodion_server-2.0.1-incubating/sql/scripts/gomon.cold) ran successfully. Performin
-g further checks...
-Checking if processes are up.
-Checking attempt: 1; user specified max: 2. Execution time in seconds: 0.
-
-The SQ environment is up!
-
-
-Process Configured Actual Down
-------- ---------- ------ ----
-DTM 3 3
-RMS 6 6
-DcsMaster 1 0 1
-DcsServer 3 0 3
-mxosrvr 96 0 96
-
-Thu Jun 30 21:15:29 UTC 2016
-Checking if processes are up.
-Checking attempt: 1; user specified max: 1. Execution time in seconds: 0.
-
-The SQ environment is up!
-
-
-Process Configured Actual Down
-------- ---------- ------ ----
-DTM 3 3
-RMS 6 6
-DcsMaster 1 0 1
-DcsServer 3 0 3
-mxosrvr 96 0 96
-
-Starting the DCS environment now
-starting master, logging to /opt/trafodion/apache-trafodion_server-2.0.1-incubating/dcs-2.0.1/bin/../logs/dcs-trafodion-1-mast
-er-ip-172-31-56-238.out
-ip-172-31-56-238: starting server, logging to /opt/trafodion/apache-trafodion_server-2.0.1-incubating/dcs-2.0.1/bin/../logs/dc
-s-trafodion-1-server-ip-172-31-56-238.out
-ip-172-31-57-143: starting server, logging to /opt/trafodion/apache-trafodion_server-2.0.1-incubating/dcs-2.0.1/bin/../logs/dc
-s-trafodion-3-server-ip-172-31-57-143.out
-ip-172-31-61-110: starting server, logging to /opt/trafodion/apache-trafodion_server-2.0.1-incubating/dcs-2.0.1/bin/../logs/dc
-s-trafodion-2-server-ip-172-31-61-110.out
-Checking if processes are up.
-Checking attempt: 1; user specified max: 2. Execution time in seconds: 1.
-
-The SQ environment is up!
-
-
-Process Configured Actual Down
-------- ---------- ------ ----
-DTM 3 3
-RMS 6 6
-DcsMaster 1 1
-DcsServer 3 3
-mxosrvr 96 7 89
-
-Starting the REST environment now
-starting rest, logging to /opt/trafodion/apache-trafodion_server-2.0.1-incubating/rest-2.0.1/bin/../logs/rest-trafodion-1-rest
--ip-172-31-56-238.out
-
-
-
-Zookeeper listen port: 2181
-DcsMaster listen port: 23400
-
-Configured Primary DcsMaster: "ip-172-31-56-238.ec2.internal"
-Active DcsMaster : "ip-172-31-56-238"
-
-Process Configured Actual Down
---------- ---------- ------ ----
-DcsMaster 1 1
-DcsServer 3 3
-mxosrvr 96 94 2
-
-
-You can monitor the SQ shell log file : /opt/trafodion/apache-trafodion_server-2.0.1-incubating/logs/sqmon.log
-
-
-Startup time 0 hour(s) 2 minute(s) 19 second(s)
-Apache Trafodion Conversational Interface 2.0.1
-Copyright (c) 2015-2016 Apache Software Foundation
->>
---- SQL operation complete.
->>
-
-End of MXCI Session
-
-***INFO: Installation setup completed successfully.
-
-******************************
- TRAFODION INSTALLATION END
-******************************
-```
-
-2. Switch to the {project-name} Runtime User and check the status of {project-name}.
-+
-```
-$ sudo su - trafodion
-$ sqcheck
-Checking if processes are up.
-Checking attempt: 1; user specified max: 2. Execution time in seconds: 0.
-
-The SQ environment is up!
-
-
-Process Configured Actual Down
-------- ---------- ------ ----
-DTM 3 3
-RMS 6 6
-DcsMaster 1 1
-DcsServer 3 3
-mxosrvr 96 96
-$
-```
-
-{project-name} is now running on your Hadoop cluster. Please refer to the <<activate,Activate>> chapter for
-basic instructions on how to verify the {project-name} management and how to perform basic management
-operations.
-
+Please refer to <<install-guided-install, Guided Install>> for the *example* of installing {project-name} on a two-node Cloudera Hadoop cluster.
http://git-wip-us.apache.org/repos/asf/incubator-trafodion/blob/92c80fd3/docs/provisioning_guide/src/asciidoc/_chapters/requirements.adoc
----------------------------------------------------------------------
diff --git a/docs/provisioning_guide/src/asciidoc/_chapters/requirements.adoc b/docs/provisioning_guide/src/asciidoc/_chapters/requirements.adoc
index e33a15e..0d038dc 100644
--- a/docs/provisioning_guide/src/asciidoc/_chapters/requirements.adoc
+++ b/docs/provisioning_guide/src/asciidoc/_chapters/requirements.adoc
@@ -86,8 +86,8 @@ Please verify these requirements on each node you will install {project-name} on
[cols="20%a,40%a,40%a",options="header"]
|===
| Function | Requirement | Verification Guidance
-| Linux | 64-bit version of Red Hat 6.5 or later, or SUSE SLES 11.3 or later. |
-| sshd | The `ssh` daemon is running on each node in the cluster. |
+| Linux | 64-bit version of Red Hat(RHEL) or CentOS 6.5 -6.8 | `cat /etc/redhat-release`
+| sshd | The `ssh` daemon is running on each node in the cluster. |
• `ps aux \| grep sshd` +
• `sudo netstat -plant \| grep :22`
| ntpd | The `ntp` daemon is running and synchronizing time on each node in the cluster. |
@@ -107,16 +107,12 @@ the port is *not* open.
| passwordless ssh | The user name used to provision {project-name} must have passwordless ssh access to all nodes. | ssh to the nodes, ensure that no password prompt appears.
| sudo privileges | The user name used to provision {project-name} must sudo access to a number of root functions . | `sudo echo "test"` on each node.
| bash | Available for shell-script execution. | `bash --version`
-| java | Available to run the {project-name} software. Same version as HBase is using. | `java --version`
+| java | Available to run the {project-name} software. Same version as HBase is using. | `java --version`
| perl | Available for script execution. | `perl --version`
| python | Available for script execution. | `python --version`
| yum | Available for installs, updates, and removal of software packages. | `yum --version`
| rpm | Available for installs, updates, and removal of software packages. | `rpm --version`
| scp | Available to copy files among nodes in the cluster. | `scp --help`
-| curl | Available to transfer data with URL syntax. | `curl --version`
-| wget | Available to download files from the Web. | `wget --version`
-| pdsh | Available to run shell commands in parallel. | `pdsh -V`
-| pdcp | Available to copy files among nodes in parallel. part of the `pdsh` package. | `pdcp -V`
|===
@@ -214,19 +210,17 @@ http://download.fedoraproject.org/pub/epel/6/x86_64/epel-release-6-8.noarch[Fedo
| protobuf | Data serialization. | yum install protobuf
| xerces-c | C++ XML parsing. | yum install xerces-c
| gzip | Data compress/decompress. | yum install gzip
-| rpm-build^2^ | Build binary and source software packages. | yum install rpm-build
-| apr-devel^2^ | Support files used to build applications using the APR library. | yum install apr-devel
-| apr-util-devel^2^ | Support files used to build applications using the APR utility library. | yum install apr-util-devel
-| doxygen^2^ | Generate documentation from annotated C++ sources. | yum install doxygen
-| gcc^2^ | GNU Compiler Collection | yum install gcc
-| gcc_c++^2^ | GNU C++ compiler. | yum install gcc_c++
+| apr-devel | Support files used to build applications using the APR library. | yum install apr-devel
+| apr-util-devel | Support files used to build applications using the APR utility library. | yum install apr-util-devel
|===
-1. `log4c++` was recently withdrawn from public repositories. Therefore, you will need to build the `log4c++` RPM
-on your system and then install the RPM using the procedure described in <<log4cplusplus_installation,log4c++ Installation>>.
-2. Software package required to build `log4c++`. Not required otherwise. These packages are *not* installed by the {project-name} Installer in this release.
-
-The {project-name} Installer requires Internet access to install the required software packages.
+The {project-name} Installer requires both Internet access/Offline mode to install the required software packages.
+Specify `db_install.py --offline` to use the offline install feature. Before that, you need to prepare a local repository
+folder for all the above dependencies.
+To create a local repository, be sure the `createrepo` package is installed, then run createrepo command in your rpm folder.
+```
+$ createrepo -d .
+```
[[requirements-trafodion-user-ids-and-their-privileges]]
== {project-name} User IDs and Their Privileges
@@ -265,17 +259,14 @@ The user ID that performs the {project-name} installation steps. Typically, this
** Run Java version command on each node in the cluster.
** Run Hadoop version command on each node in the cluster.
** Run HBase version command on each node in the cluster.
-** Create directories and files in:
-*** `/etc`
-*** `/usr/lib`
-*** `/var/log`
+** Create directories and files in `/etc/trafodion`:
** Invoke `su` to execute commands as other users; for example, `trafodion`.
** Edit `sysctl.conf` and activate changes using `sysctl -p`:
*** Modify kernel limits.
*** Reserve IP ports.
^1^ `sudo` is *required* in the current release of {project-name}. This restriction may be relaxed in later releases.
-Alternative mechanisms for privileged access (such as running as `root` or `sudo` alternative commands) are not supported.
+Alternative mechanisms for privileged access (such as `sudo` alternative commands) are not supported.
[[requirements-distribution-manager-user]]
==== Distribution Manager User