Oracle® Big Data Appliance Configuration Worksheets Release 1 (1.1) Part Number E36163-02 |
|
PDF · Mobi · ePub |
Configuration Worksheets
Release 1 (1.1)
E36163-02
September 2012
The Oracle Big Data Appliance Configuration Worksheets identify the information and decisions required when installing Oracle Big Data Appliance. This document contains the following sections:
The configuration worksheets will help you plan your installation of Oracle Big Data Appliance with the network and database administrators and your Oracle representative. It is important that you complete the worksheets and provide them to your Oracle representative before receiving your Oracle Big Data Appliance to avoid delaying the installation.
Your Oracle representative uses the information entered in these worksheets to create an Oracle Big Data Appliance Installation Template. Use the Installation Template to confirm that the configuration settings of your Oracle Big Data Appliance were entered correctly. You can make site-specific adjustments to the Installation Template in consultation with your Oracle representative. Many of these settings cannot be changed after the appliance is operational, so make your decisions carefully.
Note:
Oracle Big Data Appliance uses Cloudera's Distribution including Apache Hadoop (CDH). A Hadoop cluster on Oracle Big Data Appliance is called a CDH cluster.
The terms appliance and rack refer to Oracle Big Data Appliance.
To configure Oracle Big Data Appliance:
Read this document and the Oracle Big Data Appliance Owner's Guide to understand the networking requirements.
Identify your current management network.
Identify your current client access network.
If you have multiple appliances, then choose between connecting them into a single CDH cluster or creating an independent CDH cluster on each appliance.
Open the Oracle Big Data Appliance Configuration Worksheets in Adobe Reader or Adobe Acrobat. The document is in editable form.
Note:
Do not use a PDF reader such as Foxit or Preview to edit the file. They may corrupt the file, such that other programs cannot read it.Complete one copy of the configuration worksheets for each CDH cluster. For example, if you are configuring multiple appliances as a single CDH cluster, then complete just one copy of the worksheets. If you are configuring multiple appliances as two CDH clusters, then complete two copies of the worksheets, and so on.
Give the completed configuration worksheets to your Oracle representative. The information in these worksheets is used to generate the Installation Template, which your Oracle representative will give to you.
Verify that the information in the Installation Template is correct. To make any adjustments, contact your Oracle representative.
Configure your existing network to use the new IP addresses and host names supplied in the Installation Template.
Review the network and IP address requirements described in the Oracle Big Data Appliance Owner's Guide.
Run the network connections to the planned location for Oracle Big Data Appliance.
Inform your Oracle representative when you have completed these steps.
Note:
The network administrator should be available on-site during the installation of Oracle Big Data Appliance (1 to 2 days). Otherwise, a problem with the customer network may cause extended delays.See Also:
Oracle Big Data Appliance Owner's Guide for more information about the configuration settings
Oracle Big Data Appliance Site Checklists for site preparations before installation
The network configuration worksheets identify the naming conventions for the hardware and the IP addresses used on your network:
In Table 1, enter the names for the hardware components and other basic information.
All racks in a CDH cluster must have the same CDH cluster name, which is used in the assignment of standardized host names for all Oracle Big Data Appliance servers. By default, the host name for all 18 servers in the rack is in this format:
rackname
node
NN
.domain
In this syntax:
rackname by default is clusternameM, and M is the index of the rack in the cluster.
NN is the position number of the server in the rack (01 to 18).
domain is the domain name.
For example, if the cluster name is bda
and the domain name is example.com
, then the fully qualified host name of the server at the bottom of the first rack is bda1node01.example.com
. For the top server in the third rack of this cluster, the host name is bda3node18.example.com
.
The host names must have fewer than 38 characters, which can be ASCII letters (a to z and A to Z), numbers (0 to 9), and hyphens (-) only. Do not begin or end the name with a hyphen. You use the host names to address the servers over the client network interface.
The host names on the other networks have the following formats. You use these formats when connecting over the other networks:
For short host names over the administration network:
rackname
node
NN
-adm
For the private InfiniBand network host names:
rackname
node
NN
-priv
For the Oracle Integrated Lights Out Manager (ILOM) host names:
clusternameM
node
NN
-c
For the switch host names:
clusternameM
sw-ib
N
In this syntax:
rackname by default is clusternameM, and M is the index of the rack in the cluster.
NN is the position number of the server in the rack (01 to 18).
N is 1, 2, or 3, depending on the switch location in the rack.
If you do not want to use the cluster name in the host names, then you must provide a rack name. You also can change the suffixes used for the different network interfaces.
To assign a unique host name to each server, provide your Oracle representative with a list of all individual host names. You must also enter the domain name and time zone fields in Table 1.
Table 1 General Rack Configuration Worksheet
Property | Example | Setting |
---|---|---|
Customer NameFoot 1 |
|
|
CDH Cluster NameFootref 1 |
|
|
Rack Names: Enter the primary rack name first. |
|
|
Server Base Name |
|
|
Admin Access Suffix |
|
|
Private Name Suffix |
|
|
ILOM Name Suffix |
|
|
Switch Base Name |
|
|
Domain NameFootref 1 |
|
|
Time ZoneFootref 1 |
|
Footnote 1 Required. All other properties in this worksheet have default values.
In Table 2, enter up to four IP addresses for the Domain Name System (DNS) and Network Time Protocol (NTP) servers and up to four search domains.
In Table 3, enter the network properties for the individual servers.
Each server in the appliance is assigned an IP address for the administrative 1 gigabit Ethernet (1 GbE) network, the private InfiniBand network, the 10 GbE client access network, and Oracle ILOM. The IP addresses of the 18 servers are assigned sequentially beginning with the first IP address for node01.
If you cannot allocate 18 sequential IP addresses for a network, then provide your Oracle representative with a list of 18 IP addresses instead of just the starting IP address.
The Oracle ILOM IP addresses are on the administrative network. The administrative and Oracle ILOM IP addresses must be on the same subnet and use the same netmask and gateway.
Table 3 Network Configuration Worksheet for the Servers
Property | Setting |
---|---|
Administrative Network: First of 18 sequential IP addresses, multiplied by the number of racks in the cluster. For example, if you configure three racks in a cluster, then enter the first of 54 (18 x 3) sequential IP addresses. |
|
ILOM Network: First of 18 sequential IP addresses, multiplied by the number of racks. |
|
Administrative Network and ILOM Netmask |
|
Administrative Network and ILOM Gateway |
|
Client Network: First of 18 sequential IP addresses, multiplied by the number of racks. |
|
Client Network Netmask |
|
Client Network Gateway |
|
Are you connecting this rack to another rack through the InfiniBand fabric, such as another Oracle Big Data Appliance rack or an Oracle Exadata Database Machine? |
|
Private InfiniBand Network: First of 18 sequential IP addresses, multiplied by the number of racks. Optional: Defaults to 192.168.10.1Foot 1 |
|
Private InfiniBand Network Netmask Optional: Defaults to 255.255.255.0Footref 1 |
Footnote 1 If you are connecting this rack to another rack through the InfiniBand fabric, see the following note.
Note:
If you are connecting this rack to another rack through the InfiniBand fabric, then ensure that:The InfiniBand IP addresses of all servers are unique, including the servers in other Oracle engineered systems
All InfiniBand IP addresses are on the same network
For example, if you connect an Oracle Big Data Appliance rack to Oracle Exadata Database Machine, then you must use the same netmask on the InfiniBand networks for both systems. Moreover, after you apply this netmask to the InfiniBand IP addresses of the Exadata compute nodes, the Exadata storage servers, and the Oracle Big Data Appliance servers, all IP addresses are in the same subnet.
In Table 4, enter the IP addresses of the keyboard-video-mouse (KVM) switch, the Cisco Catalyst Ethernet switch, and the first of three Sun InfiniBand switches. The IP addresses of the switches are on the administrative network.
Table 4 Network Configuration Worksheet for the Switches
Property | Setting |
---|---|
KVM Switch IP: One IP address, multiplied by the number of racks. For example, if you configure three racks in a cluster, then enter the first of three consecutive IP addresses. |
|
Cisco Switch IP: One IP address, multiplied by the number of racks. |
|
InfiniBand Switch IP: First of three sequential IP addresses, multiplied by the number of racks in the cluster. For example, if you configure three racks in a cluster, then enter the first of nine consecutive IP addresses. |
In Table 5, enter the IP addresses for the two power distribution units (PDUs). The IP addresses of the PDUs are on the administrative network.
You can configure and activate the following optional software components for Oracle Big Data Appliance:
Note:
If you are extending a CDH cluster and these worksheets are not for the primary rack, then complete only Table 8. Leave the other Software Configuration Worksheets blank.See Also:
Oracle Big Data Appliance Owner's Guide for a description of the software configuration optionsIn Table 6, identify the applications to activate and configure.
Oracle Big Data Connectors and Oracle NoSQL Database are optional components. You must activate them now if you plan to use them. Otherwise, you can save disk space for other uses by not activating them. Disk space allocated to Oracle NoSQL Database is not available for the Hadoop Distributed File System (HDFS).
Oracle NoSQL Database Community Edition is included in the license for Oracle Big Data Appliance.
Oracle Big Data Connectors requires a separate license. You must have this license to install the connectors on Oracle Big Data Appliance. If you have a license, then you can decide whether to configure Oracle Data Integrator agent during the initial software installation, so that it is up and running immediately.
Oracle Big Data Appliance is configured to have four copies of the critical NameNode data spread across four disks on two servers. This default of four copies inside Oracle Big Data Appliance and no external copy is the Oracle recommended configuration. It is sufficient for most disaster recovery scenarios. Oracle recommends that you keep this default backup configuration.
Nonetheless, if you want to maintain a copy on a separate server outside of Oracle Big Data Appliance, you can provide an external network file system (NFS) mount point for that purpose. Oracle Big Data Appliance then maintains two copies of the NameNode data on the appliance and one copy on the external mount point. Be aware that when you provide an external mount point, the number of copies inside Oracle Big Data Appliance is reduced from four copies to two copies.
Note:
To maintain an external copy of the NameNode data, you must manually define a mount point before you install the Oracle Big Data Appliance software.To set up an external NFS mount point for backing up the NameNode data:
Log in to the NFS host. You must have root
privileges.
Create the hdfs
user in the hadoop
group. The hdfs
user ID (UID) must the same as the hdfs
UID on Oracle Big Data Appliance, and the hadoop
group ID (GID) must be the same as the hadoop
GID on Oracle Big Data Appliance.
As the root
user, create a subdirectory with the same name as the cluster under the new directory path.
Verify that this subdirectory is owned by root
.
As the hdfs
user, create two subdirectories named nn and snn under the new subdirectory.
Verify that these subdirectories are owned by user hdfs
and group hadoop
.
Log in as the root
user to node01 of Oracle Big Data Appliance.
Create a mount point by issuing a command like the following:
mount nfs-host:/backup/namenode /mnt/nfs-backup
In this example, NFS-HOST is the name of the external NFS host, /backup/namenode is the top-level directory for the backup files, and /mnt/nfs-backup is the mount point on Oracle Big Data Appliance.
In Table 6, enter the full path of the top-level directory from Step 3.
For example, if you enter the NFS directory as nfs-host:/backup/namenode
and the cluster name is bda, then:
The /backup/namenode/bda directory must exist on NFS-HOST and be owned by root
.
The /backup/namenode/bda/nn and /backup/namenode/bda/snn directories must exist on NFS-HOST and be owned by hdfs
in group hadoop
.
The directory structure on NFS-HOST looks like this:
/backup /namenode /bda /nn /snn
Note:
Complete this worksheet for the only rack or the primary rack in a CDH cluster.Table 6 Software Components Configuration Worksheet
Property | Example | Setting |
---|---|---|
Are Oracle Big Data Connectors licensed? |
|
|
Configure Oracle Data Integrator agent? |
|
|
Install Oracle NoSQL Database, Community Edition? |
|
|
Total disk space to allocate for Oracle NoSQL Database (in terabytes) |
|
|
External backup network file system (NFS) directory |
|
In Table 7, enter the configuration settings for Auto Service Request (ASR).
ASR monitors the health of Oracle Big Data Appliance hardware and automatically submits a service request when it detects a fault. Although you can opt out of this program, Oracle recommends that you enable ASR.
ASR Manager must be configured to run on a separate server outside of Oracle Big Data Appliance. Software on Oracle Big Data Appliance must be able to connect to ASR Manager and route to the Internet, either directly or through a proxy, to send event information that automatically opens service requests.
Note:
Complete this worksheet for the only rack or the primary rack in a CDH cluster.The installation software defines users, groups, and passwords. If you do not provide the passwords in Table 8 for the configuration files, then you must enter them manually during the installation.
The Oracle IDs must match those of a connected Oracle Exadata Database Machine to support the network file system (NFS) protocol between the two systems.
Note:
If you are extending a CDH cluster and these worksheets are not for the primary rack, then complete only the password fields in Table 8. Leave the ID fields blank.Table 8 Users and Groups Configuration Worksheet
Description | Example | Value |
---|---|---|
Cloudera Manager |
||
|
||
|
||
|
|
|
|
|
|
|
|
|
MySQL Database administration passwordFootref 1 |
||
MySQL Database password for Oracle Data IntegratorFootref 1 |
Footnote 1 Required for all racks. Leave blank only if you will be available during the installation.
In Table 9, provide information about the email server (SMTP) on the network.
This information enables Cloudera Manager to send email alerts when it detects a problem in the CDH cluster. If this information is omitted, then no email alerts will be sent.
Note:
Complete this worksheet for the only rack or the primary rack in a CDH cluster.Following is an example of an Installation Template.
Property | Value |
---|---|
Cluster Name | bda |
Rack Name | bda1 |
Primary Appliance | Yes |
Country | America |
Timezone | America/New_York |
Domain | example.com |
DNS Server | 172.16.100.45 |
NTP Server | 172.16.215.62 |
Search Domains | example.com us.example.com |
Type | Netmask | Gateway |
---|---|---|
Administrative - eth0 | 255.255.255.0 | 10.18.113.1 |
Private - bondib0 | 255.255.255.0 | |
Client Access - bondeth0 | 255.255.255.0 | 10.18.114.1 |
Server Position | eth0 hostname | eth0 IP address | bondib0 IP address | ... |
---|---|---|---|---|
01 | bda1node01-adm | 10.18.113.10 | bda1node01-priv | ... |
02 | bda1node02-adm | 10.18.113.11 | bda1node02-priv | ... |
03 | bda1node03-adm | 10.18.113.12 | bda1node03-priv | ... |
... | ... | ... | ... | ... |
17 | bda1node17-adm | 10.18.113.26 | bda1node17-priv | ... |
18 | bda1node18-adm | 10.18.113.27 | bda1node18-priv | ... |
Type | Hostname | IP Address |
---|---|---|
KVM Switch | bda1sw-kvm | 10.18.113.143 |
Cisco Switch | bda1sw-ip | 10.18.113.144 |
NM2-36p Spine Switch | bda1sw-ib1 | 10.18.113.145 |
NM2-GW Leaf Switch #1 | bda1sw-ib2 | 10.18.113.146 |
NM2-GW Leaf Switch #2 | bda1sw-ib3 | 10.18.113.147 |
Type | Hostname | IP Address |
---|---|---|
PDU A | bda1-pdu1 | 10.18.113.151 |
PDU B | bda1-pdu2 | 10.18.113.152 |
Property | Value |
---|---|
Oracle Big Data Connectors | are being licensed |
Oracle Data Integrator Agent | will be configured and started |
Oracle NoSQL Database | will be installed with 54 TB total diskspace |
External NFS Directory | no external NFS mount point requested |
Property | Value |
---|---|
Oracle Auto Service Requests | will be configured |
Oracle ASR server | asr-host.example.com:162 |
Oracle ASR Server root password | password |
Property | Value |
---|---|
Cloudera Manager admin password | password |
root operating system password | password |
oracle operating system password | password |
MySQL administration password | password |
MySQL password for Oracle Data Integrator | password |
Property | Value | Property | Value | Property | Value |
---|---|---|---|---|---|
oracle user ID | 1000 | oinstall group ID | 1001 | dba group ID | 1002 |
Property | Value |
---|---|
Email server (SMTP) host | router.example.com:25 does not use SSL |
Email server (SMTP) user name | smtp.user |
Email server (SMTP) | password |
Email alert recipients | admin@example.com |
For information about Oracle's commitment to accessibility, visit the Oracle Accessibility Program website at http://www.oracle.com/pls/topic/lookup?ctx=acc&id=docacc
.
Oracle customers have access to electronic support through My Oracle Support. For information, visit http://www.oracle.com/pls/topic/lookup?ctx=acc&id=info
or visit http://www.oracle.com/pls/topic/lookup?ctx=acc&id=trs
if you are hearing impaired.
Oracle Big Data Appliance Configuration Worksheets, Release 1 (1.1)
E36163-02
Copyright © 2012, Oracle and/or its affiliates. All rights reserved.
This software and related documentation are provided under a license agreement containing restrictions on use and disclosure and are protected by intellectual property laws. Except as expressly permitted in your license agreement or allowed by law, you may not use, copy, reproduce, translate, broadcast, modify, license, transmit, distribute, exhibit, perform, publish, or display any part, in any form, or by any means. Reverse engineering, disassembly, or decompilation of this software, unless required by law for interoperability, is prohibited.
The information contained herein is subject to change without notice and is not warranted to be error-free. If you find any errors, please report them to us in writing.
If this is software or related documentation that is delivered to the U.S. Government or anyone licensing it on behalf of the U.S. Government, the following notice is applicable:
U.S. GOVERNMENT END USERS: Oracle programs, including any operating system, integrated software, any programs installed on the hardware, and/or documentation, delivered to U.S. Government end users are "commercial computer software" pursuant to the applicable Federal Acquisition Regulation and agency-specific supplemental regulations. As such, use, duplication, disclosure, modification, and adaptation of the programs, including any operating system, integrated software, any programs installed on the hardware, and/or documentation, shall be subject to license terms and license restrictions applicable to the programs. No other rights are granted to the U.S. Government.
This software or hardware is developed for general use in a variety of information management applications. It is not developed or intended for use in any inherently dangerous applications, including applications that may create a risk of personal injury. If you use this software or hardware in dangerous applications, then you shall be responsible to take all appropriate fail-safe, backup, redundancy, and other measures to ensure its safe use. Oracle Corporation and its affiliates disclaim any liability for any damages caused by use of this software or hardware in dangerous applications.
Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.
Intel and Intel Xeon are trademarks or registered trademarks of Intel Corporation. All SPARC trademarks are used under license and are trademarks or registered trademarks of SPARC International, Inc. AMD, Opteron, the AMD logo, and the AMD Opteron logo are trademarks or registered trademarks of Advanced Micro Devices. UNIX is a registered trademark of The Open Group.
This software or hardware and documentation may provide access to or information on content, products, and services from third parties. Oracle Corporation and its affiliates are not responsible for and expressly disclaim all warranties of any kind with respect to third-party content, products, and services. Oracle Corporation and its affiliates will not be responsible for any loss, costs, or damages incurred due to your access to or use of third-party content, products, or services.
Cloudera, Cloudera CDH, and Cloudera Manager are registered and unregistered trademarks of Cloudera, Inc.