Technology Blogs by Members
Explore a vibrant mix of technical expertise, industry insights, and tech buzz in member blogs covering SAP products, technology, and events. Get in the mix!
cancel
Showing results for 
Search instead for 
Did you mean: 
Former Member

This is a step by step tutorial that will show you how to install HANA SPS 8 on the Amazon Cloud. With this, I strive to reconstruct the entire installation process from scratch, never missing out on any command or step to take. Therefore, I also included some screenshots from the console in order for you to see how the process unfolds. Since some of these images were quite big in size, it might prove hard to read the information therein - in this case, double-clicking on any of these images will give you a full-sized image.


Launching an instance


First, we will install a SUSE Linux Enterprise Server 11 Service Pack 3 64-bit, sized r3.8xlarge.







Choose your corresponding Availability Zone. Activating the option Protect against accidental termination makes it impossible to accidentally delete this instance.


5 volumes (adding up to 1847 GB) must be created now. You can adjust your storage options according to your needs.


Name your instance, e.g. "SAP HANA SPS 8".


If you formerly already had an AWS instance, you can use your existing Security Group ID. Otherwise you can create a new one now:


After reviewing your data one more time, you are finally ready to launch your instance.




For accessing the operating system via SSH you can keep using your existing key pairs. Should you not have one yet, then you must create one now:



Your instance has now successfully launched,  and is available under a Public IP address via SSH.





Preparing an instance

Please log in as root user and create the following directories:


mkdir /hana

mkdir /hana/shared

mkdir /usr/sap

mkdir /hana/data

mkdir /hana/data/HDB

mkdir /hana/log

mkdir /hana/log/HDB




With the following commands, we will format our 5 volumes with the ext3 file system (proceed with "y"):


fdisk -l (shows you the list of devices on your machine)


mkfs -t ext3 /dev/xvdb

Proceed along with the following commands:

mkfs -t ext3 /dev/xvdc


mkfs -t ext3 /dev/xvdd


mkfs -t ext3 /dev/xvde


mkfs -t ext3 /dev/xvdf


Next, we will edit the file vi/etc/fstab and add the following values, so that our volumes will be automatically mounted every time we restart our system. Thus, they will always be available.


/dev/xvdb  /usr/sap       ext3    defaults        0 0

/dev/xvdc  /hana/data     ext3    defaults        0 0

/dev/xvdd  /hana/log      ext3    defaults        0 0

/dev/xvde  /hana/shared   ext3    defaults        0 0

/dev/xvdf  /hanadata      ext3    defaults        0 0

With the command mount -a, all volumes displayed within fstab will automatically be mounted. You can check this with the command df-h:


For the installation it is also required to install Java, which can be easily installed on your Linux machine with the following commands:


wget http://javadl.sun.com/webapps/download/AutoDL?BundleId=90215

mv FILE NAME java-linux-x64.rpm


rpm -ivh java-linux-x64.rpm


It is up to you whether you would like to download and decompress the data on a Windows machine or a Linux machine. Keep in mind, though, that the decompressed data must be readily available on your Suse Linux machine.

Within the Marketplace, we download the 3 RAR files – for instance on a Windows machine – and decompress these. For this, you will need a decompression tool such as WinRAR or 7-Zip.



The decompressed files can now be transferred to our SUSE server e.g. via SFTP:



Switch to the *DATA_UNITS/HDB_SERVER_LINUX_X86_64 directory and make the data executable with the following commands:


find -name hdbinst   -exec chmod 744 {} +
find -name hdbsetup  -exec chmod 744 {} +
find -name hdbuninst -exec chmod 744 {} +
find -name sdbrun    -exec chmod 744 {} +


In order to prevent the installation from aborting due to a hardware check, we will have to implement this script:


export IDSPISPOPD="1"

python

>>> import os

>>> 'IDSPISPOPD' in os.environ.keys()

True

>>> quit()



I have noticed once that my installation did not resume due to a file that was not executable. I would suggest to type this command:


chmod 744 /mnt/xvdu/HANA_SPS8_ALLIN/DATA_UNITS/HDB_AFL/LINUX_X86_64/hdbinst



You should also make the file hdblcm within the subdirectory /HDB_LCM_LINUX_X86_64 executable:

chmod 744 hdblcm


Begin installation:


./hdblcm


Proceed along, entering "y" whenever prompted.


After the process, a Logfile will be created. Please check it for any warnings or errors.


Now, we will rename our instance. For this, we will have to edit the hosts file.


vi/etc/hosts


Add the following line:        127.0.0.1     imdbhdb


sudo -su hdbadm



Now, we will rename our file. For this, we will have to switch back to our directory and type the following commands:


cd /usr/sap/HDB/SYS/global/hdb/install/bin


./hdbrename -hostmap ALTERNATE=imdbhdb





Now, your HANA system is ready to go.


Should you have any questions or comments, I would be glad to assist you.

8 Comments
Labels in this area