This page is obsolete. It has been replaced with the Amazon sandbox: https://aws.amazon.com/marketplace/pp/B018RBMFG0?sr=0-2

The Trafodion sandbox is a virtual appliance that provides a simulation of Trafodion on a single-node cluster, giving you an area where you can become familiar with Trafodion. Download the trafodion_release.ova file and install it on a PC that is already running one of the supported hypervisors. The sandbox already has Trafodion Release 1.1 and Hortonworks Data Platform (HDP) installed.

NOTE: If you encounter problems or need assistance with the installation or use of the Trafodion sandbox, please email tbd for help.

Prerequisites

Checking System Resources

To run the Trafodion sandbox successfully, you will need at least 16 GB of memory on your PC. The sandbox uses about 5 GB of memory while the Hadoop services and Trafodion are running, so make sure that you have enough available memory on your PC to support those processes. Otherwise, the sandbox will operate very slowly.

Installing a Hypervisor

Install one of these tested and supported hypervisors on your PC:

Enabling Virtualization on Your PC

Your PC should have virtualization enabled in its BIOS settings. To enable virtualization on your PC, follow these steps:

  1. Power off your PC.
  2. When powering on your PC, press Esc to display the Startup Menu.
  3. When the Startup Menu appears, press F10 for BIOS Setup.
  4. In the BIOS Setup, select the System Configuration tab.
  5. Select Device Configurations.
  6. Scroll down and select Virtualization Technology.
  7. Click Save.
  8. When prompted to save changes, click Yes.
  9. Click Exit.
  10. When prompted again to save changes, click Yes.
  11. Restart your PC.

Installing the Trafodion Sandbox

Once your PC meets the prerequisites, follow these steps:

Downloading and Importing the OVA File

  1. Download the trafodion_release.ova file from http://downloads.trafodion.org/trafodion_release.ova to your PC.

    NOTE: The .ova file is about 6 GB in size. Therefore, it might take over an hour to download this file to your PC depending on the speed of your Internet connection. If you'd like to verify the download, the checksum iseb08cbc5bfd4b7b2bde8ba2d7b01db4b trafodion_release.ova, which you can compute and compare using the Linux command "md5sum trafodion_release.ova".

  2. Make sure that the downloaded file is a .ova file and not a .tar file. If you see trafodion_release.tar, rename it to trafodion_release.ova.
  3. Right-click the trafodion_release.ova file on your PC and select Open with and then VMware Workstation or VirtualBox.exe.

    The hypervisor will display a dialog box prompting you to import the new virtual machine.

  4. Keep the default name and local storage path (or virtual disk image) for the new virtual machine. If you are using VirtualBox, select Reinitialize the MAC address of all network cards. Click Import.

    If you see this VMware Workstation error message, click Retry:

    The import failed because <local storage path> 
did not pass OVF specification 
conformance or virtual hardware compliance checks.

Click Retry to relax OVF specification and virtual hardware 
compliance checks and try the import again, or click Cancel to 
cancel the import. If you retry the import, you might not be 
able to use the virtual machine in VMware Workstation.

    NOTE: The import process might take about 15 minutes to complete because it is transferring about 4 GB of data.

Configuring the Virtual Machine Settings

  1. Once you have imported the virtual machine but before starting it, check the virtual machine settings for memory, and change that setting based on your PC's system resources:
    1. In the hypervisor, display the virtual machine settings:
      • In VMware Workstation, click Edit virtual machine settings.
      • In VirtualBox, click Settings.
    2. If you are using VMware Workstation, select Memory and set it to the maximum recommended value.
    3. If you are using VirtualBox, disable the USB controller by clicking USB and clearing the Enable USB Controller check box. In the same Settings dialog box, click System and then click the Motherboard tab. Set the Base Memory by dragging the slider control to where the red and green meet. If you are at the maximum recommended value, you should no longer see Invalid settings detected at the bottom of the dialog box.
    4. Click OK to accept the changes.

Starting the Virtual Machine

  1. Start the virtual machine. For example, in VMware Workstation, click Power on this virtual machine. In VirtualBox, click the Start button.

    The hypervisor boots up CentOS, which might take a few minutes to complete.

  2. In the trafodion dialog box, select trafodion, enter traf123 for the password, and then click Log In.

TIP: Instead of shutting down the system, hibernate the virtual machine. When you restart it, it should return to the same state in which you had left it.

Starting the Hadoop Services

NOTE: It might take a while to start the Ambari console for the first time. Please be patient.

  1. Once you are logged in to the virtual machine, start the Hadoop services:
      1. Click the Firefox browser icon in the toolbar and enter sandbox:8080 (or localhost:8080) in the address bar to open the Ambari console.
      2. In the Ambari console, enter admin for the username and admin for the password, and click Sign in.
      3. All Ambari services should be started (green).
        • If they are not, in the left side of the console, click Actions and select Start All.
        • When prompted for confirmation, click OK. This will start the Hadoop services. Please note that it might take a while for all the Hadoop services to start, especially if your system has limited memory. (See the recommended system resources.)
        • If all the Hadoop services start up successfully, you will see a green check mark next to Start All Services, and its progress bar will be green at 100%. Please allow some time for all the services to be active and in a steady state. Once the services appear to be running, proceed to step 9.
        • If you see a red exclamation point next to Start All Services or a red exclamation point next to one or more services, click on each of the stopped services and manually try to start each one. (See step 10 for a list of the required Hadoop services.) Once you have manually started the services, proceed to step 9.

    NOTE: If any problems arise while you are starting the Hadoop services, see the Troubleshooting section.

  2. The HBase service might show a green check mark even when the HBase Master is not live. To check that the HBase Master is live, click the HBase service in the left pane of the Dashboard tab. If it shows that the HBase Master is not live, click Restart the HBase Master. Once the HBase Master is live, proceed to step 10.
  3. In the Ambari Dashboard, verify that these required Hadoop services continue to be in the Started state:If any of those services have stopped, restart them.
  4. Check the actual status of HBase by starting a terminal session (by selecting Applications > System Tools > Terminal from the toolbar) and running the HBase shell command called status 'detailed':
    [trafodion@trafodion ~]$ hbase shell

hbase(main):001:0>status 'detailed' 
version 0.98.0.2.1.7.0-784-hadoop2
0 regionsInTransition
master coprocessors: []
1 live servers
    trafodion.local:60020 1423613814142
        ...
0 dead servers

hbase(main):003:0>

    If there are no live HBase RegionServers, return to step 9 to restart those servers. If the HBase RegionServers are live, proceed to step 12.

Bringing Up the Trafodion Instance

NOTE: Before bringing up the Trafodion instance, make sure that the required Hadoop services and the HBase RegionServers are still running. See steps 10 and 11 in Starting the Hadoop Services.

  1. Before starting Trafodion, verify the network interface where the connectivity traffic will flow:
    1. Start a terminal session by selecting Applications > System Tools > Terminal from the toolbar.
    2. Run the ifconfig command and note which network interface (that is, eth<n>) has an IP address.
    3. Compare that network interface name with the dcs.dns.interface setting in $MY_SQROOT/dcs*/conf/dcs-site.xml.
    4. If those names are different, change the dcs.dns.interface setting in dcs-site.xml to match the network interface name. You are now ready to start Trafodion.
  2. In the terminal window, run these commands to bring up the Trafodion instance:
    cds
sqstart

    If sqstart runs successfully, you should see this output after running the sqcheck command:

    [trafodion@trafodion scripts]$ sqcheck
Checking if processes are up.
Checking attempt: 1; user specified max: 2. Execution time in seconds: 11.

The SQ environment is up!


Process		Configured	Actual	    Down
-------		----------	------	    ----
DTM		2		2	    
RMS		4		4	    
MXOSRVR		2		2

    If sqstart is not working (that is, if it does not complete after about five minutes):

    1. Type Ctrl-C in the session.
    2. Run ckillall:
      Background SQ Startup job (pid: 15136)
# of SQ processes: 9 ^C

[trafodion@trafodion scripts]$ ckillall
This command might cause irreparable damage to your database. Execute sqstop
instead. Do you really want to continue? y/n: y
Going to execute ckillall
[trafodion@trafodion scripts]$
    3. Restart the HBase RegionServers using the Ambari console. See step 9 in Starting the Hadoop Services.
    4. Retry sqstart.

    Once sqstart runs successfully, proceed to step 14 to launch sqlci.

  3. In the terminal session, perform a quick sanity check using the Trafodion Conversational Interface (sqlci). First, get the version of the software and metadata using the GET VERSION OF SOFTWARE and GET VERSION OF METADATA commands. For example:
    [trafodion@trafodion scripts]$ cd ..
[trafodion@trafodion sql]$ cd ..
[trafodion@trafodion ~]$ sqlci
Trafodion Conversational Interface 1.0.0
(c) Copyright 2014 Hewlett-Packard Development Company, LP.
>>get version of software;

  System Version 1.0.0. Expected Version 1.0.0.
  Software is current.

--- SQL operation complete.
>>get version of metadata;

  Current Version 3.0. Expected Version 3.0.
  Metadata is current.

--- SQL operation complete.

    The output of the GET VERSION commands should indicate that the software and metadata are current. 
    If you see error messages when running the GET VERSION commands, report the problem to tbd and try initializing Trafodion as follows to resolve the problem:

    >>initialize trafodion, drop;
 
--- SQL operation complete.
>>initialize trafodion;
 
--- SQL operation complete.

    Rerun the GET VERSION OF SOFTWARE and GET VERSION OF METADATA commands to make sure that the software and metadata are current.

  4. Next, create a table with a couple of records. For example:
    >>create table test1 (f1 int, f2 int);

--- SQL operation complete.
>>insert into test1 values(1,1);

--- 1 row(s) inserted.
>>insert into test1 values(2,2);

--- 1 row(s) inserted.
>>select * from test1;

F1            F2
-----------   -----------
          1             1
          2             2

--- 2 row(s) selected.
>>get tables;

Tables in Schema TRAFODION.SEABASE
==================================

TEST1

--- SQL operation complete.
>>exit;

    Those SQL commands should run successfully.

Connecting to the Trafodion Sandbox

  1. Get the IP address of the Trafodion sandbox so that you can later connect to it from a client application. In the Ambari console, click the Hosts tab, scroll down to see the Summary, and note the IP address of trafodion.local.
  2. Download and install the Trafodion JDBC and/or ODBC drivers so that you can connect to the Trafodion sandbox from a client application using the sandbox's IP address. For instructions, see the Trafodion Client Installation Guide (pdf, 1.0 MB). That guide explains how to install the drivers, how to connect to a Trafodion instance, and how to run sample programs to test the connection.

Troubleshooting

Problem: App Timeline Server fails to start

Solution: Update the yarn.timeline-service.store-class value as follows:

  1. In the Ambari console, select the Services tab.
  2. Click YARN in the left pane.
  3. Click the Configs tab.
  4. Under Application Timeline Server, enter the following value in the yarn.timeline-service.store-class field:
    org.apache.hadoop.yarn.server.timeline.LeveldbTimelineStore
  5. Click the Save button.

Problem: Hadoop services do not start.

Solution: You might need to set the proxy server in /etc/yum.conf:

  1. Start a terminal session by selecting Applications > System Tools > Terminal from the toolbar.
  2. In the terminal session, log in as root with the password of traf123:
    [trafodion@trafodion ~]$ su root
Password:
[root@trafodion trafodion]#
  3. Edit the /etc/yum.conf file by setting the proxy to http://trafodion:8080:
    [root@trafodion trafodion]# vi /etc/yum.conf

#<proxy-server>:<port-number>
proxy=http://trafodion:8080

[root@trafodion trafodion]#
  4. Close the terminal window.
  5. Return to step 8c to restart all the Hadoop services.