The SAP HANA Reference for SAP Basis Administrators was started because during studies for HANATec Certification it was noticed that anybody interested in HANA Administration was missing a one-stop shop location and reference for the HANA Commands and Command Line Tools and Administrator's SQL queries respective to the Basis Administrator's HANA requirements.
The SAP HANA Reference for SAP Basis Administrators is a live SCN Community project, this means it is continuously updated and extended to include the latest valuable knowledge on SAP HANA from the Basis Administrator's perspective.
Everybody who has gems of knowledge anything from tips tricks to howto's on SAP HANA from the Basis Administration perspective is positively welcomed and encouraged to join this open source sharing SCN Community collaborative initiative to make this resource the most powerful SAP HANA Basis Administration
Let's together exploit the power of the SCN Community and unlock the value of the knowledge and experience in the community to the benefit of all.
To contribute either add a comment at the bottom or email or private message me.
Looking forward to more community contributions.
All the best,
Andy.
p.s.
if you click the View as PDF button on the right, then you can save this as a pdf onto your phone or pc for offline access and reading - of course, this work is live so come back from time to time and get the latest version.
Date | Author | Subject |
---|---|---|
8th June 2013 | martin.english | Developer Edition Filesystem Structure |
19th September 2013 | paul.aschmann | Forward port 80 to XS Engine (8080) |
19th September 2013 | paul.aschmann | XS Engine debug mode |
17th December 2013 | jakeechanove | Steps To Reset HANA System User Password |
21st January 2014 | jakeechanove | Configure ABAP to HANA SSL Connection |
23rd January 2014 | nicholas.chang | SAP HANA Lifecycle Manager - How to launch, start, install |
6th February 2014 | reinhard.jud2 | Additional information for the RowStore Reorg |
10th February 2014 | jakeechanove | Move HANA Data and Log Files to a different Mount Point |
11th February 2014 | applebyj | Licensing, Architecting and Sizing BW on HANA |
26th March 2014 | vvinodnair | Handy SQL Statements for Monitoring HANA |
7th April 2014 | singh.vinay | HANA Daily Monitoring Template |
28th April 2014 | richard.bremer | Reactivating the SYSTEM user when no User with USER ADMIN is Available |
10th June 2014 | lars.breddemann | Big thanks to Lars for taking time to correct and improve the look and feel |
18th June 2014 | fang%2Cyuan | The Configuration and Usage of SAP HANA Web-based IDE |
1st July 2014 | xing.jin2 | Preparing a SAP Hana system for SAP Lumira installation |
3rd July 2014 | devendrakumar.saxena | Troubleshooting Issues when implementing SAML SSO in HANA XS Engine |
./HDB stop
/usr/sap/hostctrl/exe/sapcontrol -nr &-function StopSystem HDB
./HDB start
/usr/sap/hostctrl/exe/sapcontrol -nr &-function StartSystem HDB
[ thanks to martin.english for providing this information ]
On the Developer Editions, sapcontrol lives in the instance EXE folder; i.e.
cdexe
./sapcontrol -nr &-function &
On the current (revision 56) developer edition,
cdexe takes you to
/usr/sap/HDB/SYS/exe/hdb
which is linked to
/usr/sap/HDB/exe/linuxx86_64/hdb
imdbhdb:~ # su - hdbadm
imdbhdb:/usr/sap/HDB/HDB00> cdexe
imdbhdb:/usr/sap/HDB/SYS/exe/hdb> pwd
/usr/sap/HDB/SYS/exe/hdb
imdbhdb:/usr/sap/HDB/SYS/exe/hdb> cd ..
imdbhdb:/usr/sap/HDB/SYS/exe> ls -la
total 8 drwxr-x- 2 hdbadm sapsys 4096 2013-01-25 14:17 .
drwxr-x- 3 hdbadm sapsys 4096 2013-01-25 14:17 ..
lrwxrwxrwx 1 hdbadm sapsys 25 2013-01-25 14:17 hdb -> ../../exe/linuxx86_64/hdb
imdbhdb:/usr/sap/HDB/SYS/exe>
/usr/sap/hostctrl/exe/sapcontrol -nr &-function GetSystemInstanceList
/usr/sap/hdbenv.sh
/usr/sap/exe/hdbnameserver
/usr/sap/hdbenv.sh
/usr/sap/exe/hdbindexserver -console
hdbuserstore SET & & & &
For example:
hdbuserstore SET millerj localhost:30115 JohnMiller 2wsx$RF
hdbuserstore LIST &
For example:
hdbuserstore LIST millerj
:KEY: millerjENV: localhost:30115USER: JohnMiller?
hdbsql -U &
For example:
hdbsql -U millerj
Ref: SAP Note 1625203 - saphostagent/sapdbctrl for newdb
The saphostagent functions for querying the database status and for
starting and stopping the database are now also available for the HANA.
If you want to use the connect without a password, the following
prerequisites must be met:
a. To query information from the database, you require a database user with
the monitoring role.
You can create a database user with the HANA studio. Connect as the SYSTEM
user and execute the following SQL commands:
CREATE USER SAPDBCTRL PASSWORD x&x
GRANT MONITORING TO SAPDBCTRL
Then logon to the HANA studio under the username &and execute
the following command:
ALTER USER SAPDBCTRL PASSWORD &
b. As the OS user &adm, you must provide the user key &SAPDBCTRL in
hdbuserstore on the database server. You can check the existence of the
user key with the following command:
hdbuserstore LIST &SAPDBCTRL
If the user key does not exist, you can create it as follows:
hdbuserstore SET &SAPDBCTRL &:&SAPDBCTRL &
The tool hdbuserstore is installed under the following paths:
/usr/sap/hdbclient
or
/usr/sap/&hdbclient
c. To check whether the sapdbctrl queries without passwords work, execute
the following command as OS user &:
/usr/sap/hostctrl/exe/saphostctrl -function GetDatabaseStatus -dbname
&-dbtype hdb
If sapdbctrl responds with the following error text:
ERROR: Database user authentication failed: SQLDriverConnect: Connect with
userkey failed! then you must check the connect data in hdbuserstore again.
SET SYSTEM LICENSE ‘&’
UNSET SYSTEM LICENSE ALL
SET SYSTEM LICENSE LicenseFile
ALTER SYSTEM RESET MONITORING VIEW SYS.M_VOLUME_TO_STATISTICS_RESET
initialises the statistics shown by this view
M_VOLUME_IO_STATISTICS_RESET
now shows the statistics since the reset time
SELECT round(sum(TOTAL_MEMORY_USED_SIZE/1024/1024)) AS "Total Used MB"
FROM SYS.M_SERVICE_MEMORY;
SELECT round(sum(CODE_SIZESTACK_SIZE)/1024/1024) AS "Codestack MB"
FROM SYS.M_SERVICE_MEMORY;
SELECT round(sum(MEMORY_SIZE_IN_TOTAL)/1024/1024) AS "Column Tables MB"
FROM M_CS_TABLES;
SELECT round(sum(USED_FIXED_PART_SIZE +
USED_VARIABLE_PART_SIZE)/1024/1024) AS "Row Tables MB"
FROM M_RS_TABLES;
SELECT SCHEMA_NAME AS "Schema",
round(sum(MEMORY_SIZE_IN_TOTAL) /1024/1024) AS "MB"
FROM
M_CS_TABLES
GROUP BY SCHEMA_NAME
ORDER BY "MB" DESC;
SELECT TABLE_NAME AS "Table",
round(MEMORY_SIZE_IN_TOTAL/1024/1024, 2) as "MB"
FROM M_CS_TABLES
WHERE SCHEMA_NAME = 'SYSTEM'
ORDER BY "MB" DESC;
select round((USED_PHYSICAL_MEMORY + FREE_PHYSICAL_MEMORY) /1024/1024/1024, 2) as "Physical Memory GB"
from PUBLIC.M_HOST_RESOURCE_UTILIZATION;
Execute the Linux command cat /proc/meminfo | grep MemTotal
Execute the SQL query:
select round(FREE_PHYSICAL_MEMORY/1024/1024/1024, 2) as "Free Physical GB"
from PUBLIC.M_HOST_RESOURCE_UTILIZATION;
Execute the Linux command:
awk 'BEGIN {sum = 0};
/^(MemFree|Buffers|Cached):/ {sum = sum + $2};
END {print sum}' /proc/meminfo
Shut down the instance.
Start the name server:
/usr/sap/HDB/hdbenv.sh
/usr/sap//HDB/exe/hdbnameserver
Start an index server in a new console:
/usr/sap/hdbenv.sh
/usr/sap/exe/hdbindexserver -console
ALTER USER SYSTEM password &
UPDATE: Section 7.3 in the admin guide for SP6 discusses how to do this: [http://help.sap.com/HANA/SAP_HANA_Administration_Guide_en.pdf ]
Reset HANA SYSTEM user password
• SSH to the HANA system and login as root
• SU to the &adm user
? su – anaadm
• Identify running HANA services
? ps aux | grep anaadm
• Stop the HANA server
? ./HDB stop
• Verify services are no longer running
? ps aux | grep anaadm
• Start the name server service – this may not get back to a prompt
? hdbnameserver &;
• Start index server. Run command even though may not be at a prompt.
? hdbindexserver –console
• Reset password for user SYSTEM
? ALTER USER SYSTEM PASSWORD Password1234
• Stop hdbindexserver service by hitting Ctrl-C
• Kill hdbnameserver service
• killall -9 hdbnameserver
• Verify all HDB services are stopped
? ps aux | grep anaadm
• Start HANA
• ./HDB start
• Verify HANA services are running
• ps aux | grep anaadm
• Update password in HANA studio
And you're done 🙂
cd /mnt/&DATA_UNITS/HANA_IM_LINUX__X86_64
Call the script in GUI mode to perform the rename:
./HANAconfig.sh --gui
Select Rename HANA System
Select Next
Specify the required entries.
$DIR_INSTANCE/../SYS/global/hdb
custom/config global.ini
indexserver.ini
nameserver.ini
$SAP_RETRIEVAL_PATH
sapprofile.ini
daemon.ini
/usr/sap/&trace
Instead of Client-Side JDBC or ODBC tracing an administrator would like to enable serverwide SQL traces.
1. Open In-Memory Computing Engine (IMCE) Studio
2. Go to the Administration Console -> Diagnosis Files
3. Click the Configure Trace ... icon in the top right corner
4. Select "ALL" for SQL tracing
5. Click Apply and OK
The SQL trace file will be listed along the other trace files in the following format:
sqltrace_&_&_000.py
Note: Make sure you are connected to the correct HANA instance 🙂
ref: 1627480 - How to enable serverwide HANA SQL traces?
The steps below are for Windows. Adjust them accordingly for UNIX and Linux.
1. Open a command prompt
2. Navigate to the location of the HANA Client (e.g. C:\SAP\hdbclient)
3. Run the following commands
1. hdbodbc_cons config trace filename &(e.g. C:\SAP\HANA_ODBC.txt)
2. hdbodbc_cons trace sql on
3. hdbodbc_cons trace api on
4. Confirm the settings took by running hdbodbc_cons show all. You should see something similar to the following:
Configuration:
Trace file name : C:\SAP\HANA_ODBC.txt
Trace flags : A:a
ODBC trace : enabled
Short trace : disabled
Debug trace : disabled
SQL trace : enabled
Time stamp prefix : disabled
Packet trace : disabled
Settings:
Update count : 0
Total size : 256
equivalent to 0 process-specific parts.
Version flag : 0
5. Disable tracing by running hdbodbc_cons trace off
6. Confirm the settings took by running hdbodbc_cons show all. You should see something similar to the following:
Configuration:
Trace file name : C:\SAP\HANA_ODBC.txt
Trace flags : (none set)
Settings:
Update count : 1
Total size : 256
equivalent to 0 process-specific parts.
Version flag : 0
Forced re-read of global configuration with last update.
ref: 1834470 - How to enable tracing on HANA
odbcreg -t hdbcodbc (for 64-bit driver)
or
odbcreg32 -t hdbcodbc32 (for 32-bit driver).
odbcreg -g or odbcreg32 -g to get a list of installed drivers.
The SAP HANA driver is called HDBODBC
As of HANA SPS5, the SAP HANA Automated Update Guide only mentions
SUM (SUM4HANA) for updating the SAP HANA appliance software
and related components.
You are not sure if SUM is mandatory and using hdbupd is obsolete by now.
+The use of SUM is mandatory and the highly recommended way for updating
the SAP HANA appliance. SUM takes care of:
Other tools such as the On Site configuration tool rely on this landscape
to be correct in order to work properly.
It is technically possible to use hdbupd but you should consider this option
only in special/extreme situations and after consulting a SAP representative.
So how do you recognize if your SAP HANA instance was installed with hdbinst or
Unified Installer?
It is really simple, if the directories /usr/sap/&SUM and /usr/sap/&lm_structure
exist then the Unified Installer was used and you should apply Support Package Stacks
and Revisions with SUM for HANA. If you do not have these directories, the recommendation
is to bring your instance up-to-date and install SUM for HANA and lm_structure using
SAP Note 1793303. After following this note you can use SUM for HANA to apply revisions
and SP Stacks.
export JAVA_HOME=/usr/sap/&SUM/jvm/jre
/usr/sap/&SUM/daemon.sh install
/etc/init.d/sum_daemon start
hdbupd or hdbsetup which are provided with the SAP HANA database installation.
hdbinst -a client
fullSystemInfoDump.py
$DIR_INSTANCE/exe/python_support
(Linux).
python fullSystemInfoDump.py
DIR_TEMP/system_dump where DIR_TEMP
hdbsrvutil -z | grep DIR_TEMP=
python fullSystemInfoDump.py &
python fullSystemInfoDump.py –h
[ts1adm@luvm252058a:/usr/sap/TS1/HDB01/exe/python_support | mailto:ts1adm@luvm252058a:/usr/sap/TS1/HDB01/exe/python_support]> python
fullSystemInfoDump.py
User name: SYSTEM
Password:
hdbsql -h
hdbsql -n PARMA -i 1 -u MONA –p RED
To log on using a &, enter the following command:
hdbsql -U &
Start HDBSQL:
hdbsql /
Log on to the database:
\c -n & -i & -u &,&
\s
host : wdfd00245293a:30015
database : ORG
user : SYSTEM
kernel version: 1.50.00.000000
SQLDBC version: libSQLDBCHDB 1.50.00 Build 0000000-0120
autocommit : ON
Welcome to the SAP In-Memory Computing interactive terminal.
Type: \h for help with commands
\q to quit
hdbsql=> select count () from EIM360.mara
14927
/usr/sap/exe/
BACKUP DATA USING FILE (‘&’)
BACKUP DATA USING FILE (‘/backupDir/data/monday/COMPLETE_DATA_BACKUP’)
hdbsql enables you to trigger backups via crontab.
The client software enables access to the hdbuserstore.
Use the following command:
hdbinst –a client (default location: /usr/sap/hdbclient)
Use the following command:
/usr/sap/hdbclient/hdbuserstore set &&:3&15
&&Example: /usr/sap/hdbclient/hdbuserstore set
BACKUP vebwtests1:30015 user password
/usr/sap/hdbclient/hdbsql –U&"BACKUP DATA USING FILE ('&')"
Example: /usr/sap/hdbclient/hdbsql -U BACKUP "BACKUP DATA USING FILE ('MONDAY')"
A data backup is then created in the default location. In the above example,
the prefix of all service-related backup files is “MONDAY”.
a. Find the Backup ID.
When a data backup is started, the system assigns a unique ID to the data backup.
To find the backup ID of the running data backup, use the monitoring view
M_BACKUP_CATALOG,
select BACKUP_ID from "SYS"."M_BACKUP_CATALOG" where entry_type_name =
'complete data backup' and state_name = 'running' order by sys_start_time desc;
To cancel the running data backup use the following SQL command:
backup cancel &
select state_name from "SYS"."M_BACKUP_CATALOG" where backup_id = &
RECOVER DATABASE UNTIL TIMESTAMP '2011-08-22 15:00:00' CHANGE ALL DATA USING PATH
('/backup/MONDAY/') CHANGE ALL LOG USING PATH ('/backup/logs1/','/backup/logs2/')
Determine the volume IDs and the lowest log item of the missing log backup from the name of the backup. If a previous restore has failed with the error message specified above, you can determine the name of the missing log backup from the error message.
The log backups are named after the schema log_backup_&_0_&log item>_&, for example,
log_backup_2_0_12345_67890.
In this example, the volume ID would be 2 and the lowest log item 12345.
Use hdblogdiag to determine the lowest log item that still exists in the log segments of the log area as follows:
Call
hdblogdiag seglist &
This call issues a list of log segments that looks as follows:
LogSegment[0/0:&-&(...
LogSegment[0/1:&-&(......
Take into account that the system outputs the log items in hexadecimal format and that you have to convert them into decimal numbers.
Example:
Assuming the parameter basepath_logvolumes has the value /HANA/log/HAN and the parameter use_mountpoints has the value yes, so that the log egments lie in the subdirectory mnt0001.
The log segments for the volume with the volume ID 2 then lie there in the subdirectory
hdb00002: > hdblogdiag seglist /HANA/log/HAN/mnt0001/hdb0002
LogSegment[0/0:0x129540-0x131c00(...
LogSegment[0/1:0x131c00-0x139540(...
Log segment 0 has the lowest log item 1217856 (hexadecimal 0x129540), log segment 1 has the lowest log item 1252352 (hexadecimal 0x131c00).
This means that the lowest log item that exists in the log segments is 1217856.
If the lowest log item of the missing log backup is bigger than the lowest log item that still exists in a log segment of the log area, a restore to the latest available point in time is possible.
In such a case, proceed as follows:
Generate a new backup catalog with
hdbbbackupdiag -generate (SAP Note1812057)
Start the restore for the point in time that you selected.
If a restore to the latest available point in time is not possible, you can restore the SAP HANA database to an earlier point in time.
You can restore the SAP HANA database only with the log backups that were written before the missing log backup.
You cannot use the log backups that were written after the missing log backup.
In this situation, proceed as follows:
Generate a new backup catalog with
hdbbbackupdiag -generate (SAP Note 1812057).
Start the restore for the point in time that you selected. For this, you must select "Initialize Log Area".
hdbbackupdiag
hdbbackupdiag -d $DIR_INSTANCE/../SYS/global/hdb/metadata -c
BackupCatalog.xml
Data backup file: thursday_databackup_0_1
RedoLogPosition: 0
Backup time: 2012-04-12T17:30:0302:00
"recovery strategy could not be determined" and
"Catalog backup not found".
generate
dataDir
--logDirs
-d
hdbbackupdiag --generate --dataDir $DIR_INSTANCE/backup/data
--logDirs $DIR_INSTANCE/backup/log -d $DIR_INSTANCE/backup/log
hdbbackupdiag --generate --dataDir $DIR_INSTANCE/backup/data
--logDirs $DIR_INSTANCE/backup/log,/mnt/disc2/log -d
$DIR_INSTANCE/backup/log
"hdbbackupdiag -d "
hdbcons 'mm flag / -rs as'
hdbcons 'mm flag / -rd as'
hdbcons 'mm callgraph -r -o /'
a. hdbcons 'mm bl -rt Pool' > /&&
b. hdbcons 'mm l -s -S -p' > /&&
hdbcons To get threads list
hdbcons "context list -s" > thread_callstack_test.txt
hdbcons "runtimedump dump -f /test/rte_dump.txt"
RAM = (Source data footprint - 60gb) * 2 / 4 * c1 + 90gb
or
RAM = ( colstore tables footprint * 2 / 4 + rowstore tables footprint /1.5) * c1 + 50gb
Disk persistence = 4 * RAM
Disk Log = 1 * RAM
c = source database specific compression factor (where applicable)
RAM dynamic + RAM static
Total RAM is
RAM = (Source data footprint - 60gb) * 2 / 4 * c1 + 90gb
or
RAM = ( colstore tables footprint * 2 / 4 + rowstore tables footprint /1.5) * c1 + 50gb
Disk persistence = 4 * RAM
Disk Log = 1 * RAM
Useful SAP Notes:
OSS 1736976 - Sizing Report for BW on HANA
OSS 1514966 - SAP HANA 1.0: Sizing SAP In-Memory Database
OSS 1799449 - Sizing report ERP on SAP HANA database
OSS 1855041 - Sizing Recommendation for Master Node in BW-on-HANA
You intend to migrate an SAP NetWeaver BW system from any database platform to the SAP HANA In-Memory Database and need information on hardware sizing of the HANA platform.
Starting with version 7.30 SP5 you can run SAP Business Information Warehouse (SAP BW) on SAP HANA as database platform.
This enables you to leverage the In-Memory capabilities of HANA and the SAP-HANA-optimized BW objects.
To simplify sizing of an SAP BW system that is supposed to be migrated, SAP now provide the database independent ABAP report
/SDF/HANA_BW_SIZING
which computes all information relevant to sizing a HANA database server.
A first version of the report is available with ST-PI 2008_1_7xx SP06.
Important: If you plan to migrate a BW system with significant size (e.g. 20 TB source system database size), please read SAP note 1855041.
For a detailed description of the report, its input parameters, and how to interpret its output, please refer to the document:
"SAP_BW_on_HANA_Sizing_Report.pdf" attached to SAP Note 1736976 - Sizing Report for BW on HANA.
The ABAP report now includes the following features:
ref: SAP Note 1736976 - Sizing Report for BW on HANA and the pdf attached to the Note
Initial Sizing observations for Suite on HANA (SAP enhancement package 6 for SAP ERP 6.0, version for SAP HANA, SAP enhancement package 2 for SAP CRM 7.0, version for SAP HANA, SAP enhancement package 2 for SAP SCM 7.0, version for SAP HANA, SAP enhancement package 3 for SAP Supplier Relationship Management 7.0, version for SAP HANA)
Initial sizing observations are based on lab measurements, load tests and measurements performed in customer test systems.
This information may change, if additional measurement results are available.
This information is only valid, if the system usage is not changed after the move; intense additional reporting and OLAP processing is not taken into account.
Application Server
Based on the measurements SAP doesn't expect any changes concerning CPU, memory and network requirements for the ABAP application server. This means that the existing hardware and network infrastructure can still be used. Please check the PAM (www.service.sap.com/pam) for further details.
HANA Main Memory
As far as sizing of the HANA appliance is concerned, main memory is the most important resource.
The current recommendation from SAP is that you should take half of the size of an uncompressed but "well-maintained" disk-based database and include a safety buffer of 20%. This means, if the database is currently approximately 1,8 TB in size (tables plus indexes), SAP recommends a HANA appliance of 1 TB.
You should not forget to consider your future data growth into your planning.
SAP doesn't expect changes in memory requirements for applications using liveCache.
HANA CPU and Disk Space
The appliance is configured in such a way that the CPU power and the I/O capacity is sufficient.
To fully support the parallel processing capabilities of HANA for optimal response times for analytical applications SAP prefers a factor of 3 to 4 more CPU power for HANA than for disk based databases without parallelization of single statements.
This also provides a buffer for running OLTP and OLAP load simultaneously.
If other software is running on the database server, please be careful to identify the CPU consumption of the database.
You can for example look at the CPU consumption of the database process(es) using SAP's monitoring tools.
Regarding the disk space of Suite on HANA, SAP recommends to use half of therequired
disk space that is needed for the Business Suite for disk-based databases.
Please note that this does not mean that each table just needs half of the disk space as compared to a disk based database.
For initial sizings you should use the Quick Sizer (www.service.sap.com/quicksizer) results for database size (DB Disk) and CPU requirements (DB SAPS) as input for HANA main memory, disk space and CPU sizing.
ref: Note 1793345 - Sizing for Suite on HANA
$ python hdbkrbconf.py -h
for help.
o HANA Database Rev 50 or later
- HANA Database has to be upgraded prior to run row store reorganization
o Catalog integrity check
- The following procedure call should run successfully and returns empty result set.
CALL CHECK_CATALOG
('CHECK_OBJECT_REFERENTIAL_INTEGRITY','','','');
o Row store reorganization is recommended, when allocated row store size is over 10GB and free page ratio is over 30%.
SELECT HOST, PORT,
CASE WHEN (((SUM(FREE_SIZE) /SUM(ALLOCATED_SIZE)) > 0.30)
AND SUM(ALLOCATED_SIZE) > TO_DECIMAL(10)/102410241024)
THEN 'TRUE' ELSE 'FALSE' END "Row store Reorganization Recommended",
TO_DECIMAL(SUM(FREE_SIZE)100 / SUM(ALLOCATED_SIZE), 10,2) "Free Space Ratio in %"
,TO_DECIMAL( SUM(ALLOCATED_SIZE)/1048576, 10, 2) "Allocated Size in MB"
,TO_DECIMAL( SUM(FREE_SIZE)/1048576, 10, 2) "Free Size in MB"
FROM M_SHARED_MEMORY
WHERE
( CATEGORY = 'TABLE' OR CATEGORY = 'CATALOG' )
AND PORT LIKE '%03'
GROUP BY HOST, PORT;
o Estimation on maximum possible memory space that can be reclaimed:
SELECT SUM(NUM_FREE_DATA_PAGES)16/1024 "Estimated Maximum Memory Saving in MB"
FROM SYS.M_DEV_MEMORY_SEGMENT
WHERE SEGMENT_TYPE = 0;
o Check disk space where log files are located.
SELECT (SUM(NUM_FREE_DATA_PAGES)16/1024)1.5 "Estimated Log Volume Size in MB"
FROM SYS.M_DEV_MEMORY_SEGMENT
WHERE SEGMENT_TYPE = 0;
1. Make a complete database backup
2. Upgrade database to Rev50 or later
3. Make sure to start up database at least once during upgrade
4. Shutdown HANA database
5. Set up configuration parameter in indexserver.ini via HANA Studio.
http://scn.sap.com/row_engine page_compaction_enable = true
6. Restart the database. Startup will take longer than normal startup when row store reorganization runs
7. Check the memory fragmentation status after reorganization:
SELECT HOST, PORT,
CASE WHEN (((SUM(FREE_SIZE) / SUM(ALLOCATED_SIZE)) > 0.30)
AND SUM(ALLOCATED_SIZE) > TO_DECIMAL(10)102410241024)
THEN 'TRUE'
ELSE 'FALSE'
END "Row store Reorganization Recommended",
TO_DECIMAL( SUM(FREE_SIZE)100 / SUM(ALLOCATED_SIZE), 10,2) "Free Space Ratio",
TO_DECIMAL( SUM(ALLOCATED_SIZE)/1048576, 10, 2) "Allocated Size in MB",
TO_DECIMAL( SUM(FREE_SIZE)/1048576, 10, 2) "Free Size in MB"
FROM M_SHARED_MEMORY
WHERE ( CATEGORY = 'TABLE' OR CATEGORY = 'CATALOG' )
AND PORT LIKE '%03'
GROUP BY HOST, PORT;
8. Check Indexserver trace
Please check the indexserver trace files after row store reorganization is done.
Row store reorganization is performed in 11 phases when successful.
It prints out "[RSReorg] start" message at the beginning,
"[RSReorg] success" message followed by "Reorganization statistics"at the end.
When there are errors during reorganization, row store reorganization is stopped in that phase and all the changes are rolled back and database starts up to return to its original state
Ref: SAP Note 1813245 - SAP HANA DB: Row store reorganization before reorganization is run.
SAP Support asks you to generate a runtime dump during a high CPU/Memory consumption
a) Log into the linux HANA host as sidadm user;
b) Run command 'hdbcons';
c) On the hdbcons console run command below:
> runtimedump dump
This will create a file under traces directory called 'indexserver....rtedump.trc'.
d) Attach generated trace file to the OSS Message
Or alternatively
hdbcons "runtimedump dump -f /test/rte_dump.txt"
o dbsl_equi_join
You can use this hint for a FOR_ALL_ENTRIES in order to inform the
DBSL that an EQUI join can be created from this FAE statement; this
EQUI join can then be resolved for an IN data set.
You must then specify this hint together with the hint of the DBI interface
(&prefer_join 1&) in the statement (see Note 48230).
ABAP
SELECT * INTO FROM FOR ALL ENTRIES IN itab
WHERE a = itab-a AND b = itab-b AND c = itab-c AND d = itab-d
%_HINTS HDB '&prefer_join 1&' hdb 'dbsl_equi_join'
From this, the system generates the following statement for the
database
SQL for the database
select * from where (a, b, c, d) IN ( (?, ?, ?, ?), ... (?, ?, ?, ?) )
o dbsl_add_stmt
Use: dbsl_add_stmt
This hint provides you with the option of adding an additional SQL
enhancement to the SQL statement in the DBSL. This hint is resolved
in the DBSL and the system appends the following text to the end of
the statement that was generated (after the keyword
dbsl_add_stmt).
ABAP
SELECT node FROM CONNECTION ('HANA') INTO TABLE T_NODE
%_HINTS HDB 'dbsl_add_stmt WITH PARAMETERS (''expression'' =''LEAVES("Node1")'' )'
SQL for the database
select node from table with parameters ( 'expression' = 'leaves("Node1")' )
This hint is available in the following releases:
720 DBSL as of patch level 103
720-EXT DBSL as of patch level 103
803 DBSL
• You identify an existing SAP HANA database system, created by an installation or a file system
copy, that is to be renamed.
• The target system ID (SID) must not exist.
• The target instance number must not exist.
• You are logged in as the root user.
• If a mountpoint contains the system ID, you must first rename it manually before performing the rename of the SAP HANA database.
a. Open a root shell.
b. Via the local mount point, change to the following directory on the shared file system:
///global/hdb/install/bin
c. Call the hdbrename program by entering this command:
./hdbrename
d. Specify the required data.
The system is now renamed.
SAP Note 1738390 - How to update SAP HANA Linux server hardware key
Stop the SAP HANA DB instance
Remove line 'id='
Execute 'hdbnsutil -convertTopology'
Start the SAP HANA DB instance
HDB proc
HDB info
eg /usr/sap/HDB/HDB00/HDB info
http://help.sap.com/HANA/HANA_dev_sqlscript_en.pdf
SAP on the 28th May 2014 published a wonderful SAP HANA Technical Operations Manual, which is here:
SAP HANA Technical Operations Manual
SAP HANA Update and Configuration Guide from SAP
[[http://help.sap.com/HANA/HANA_admin_en.pdf |
A great blog by Balaji Rajendran explaining some of the most important HANA tables/views required for administration
To be continued
The time stamps are supposed to be displayed as UTC time in Data Provisioning of HANA Studio.
It was a design decision to use UTC time to allow the replication schema to accommodate systems from differing time zones. Using UTC time simplifies this task as then it is not necessary to consider the time zone for each of the connected SLT and their respective source systems.
The time stamps you see in Data Provisioning match the internal time stamps which are used in the replication process on each connected SLT system.
The system works as designed.
1861325 - HANA: "Start Time" of replication on all tables are UTC time in Data Provisioning of HANA Studio
If you are not sure which SUSE distribution you are using, execute the following command to identify the release version and patchlevel.
cat /etc/SuSE-release
If you are running on SLES 11 SP2 the following values should be shown:
SUSE Linux Enterprise Server 11 (x86_64)
VERSION = 11
PATCHLEVEL = 2
To view which revision of SAP HANA you are using go to /usr/sap//lm_structure and view the
landscapeDescription.xml file
You want to remove the SAP HANA Extended Application Service (SAP HANA scriptserver) from the topology of your SAP HANA database (DB) because you do not use any applications that require SAP HANA scriptserver.
Or you would like to restore the backup created when the SAP HANA script server was not activated.
You must not simply stop the SAP HANA script server service. Since it has its own persistency, it is relevant for backups. A data backup fails if the SAP HANA scriptserver volumes are still entered in the topology but the service is not running.
ALTER SYSTEM ALTER CONFIGURATION ('daemon.ini', 'host', '<name>')
UNSET ('scriptserver','instances') WITH RECONFIGURE
To do this, carry out the following command for every with an SAP HANA scriptserver service:
ALTER SYSTEM ALTER CONFIGURATION ('topology.ini', 'system')
UNSET ('/host/', 'scriptserver') WITH RECONFIGURE
o Remove the volumes from the topology. To do this, carry out the following command for every of an SAP HANA scriptserver service:
ALTER SYSTEM ALTER CONFIGURATION ('topology.ini', 'system')
UNSET ('/volumes', '') WITH RECONFIGURE
Caution: It can be a risk to make changes to the topology.
Therefore, carefully check the host and the volume ID.
o Create a new data backup.
o Start the applications that use the database.
Ref: 1867783 - XFS Data Inconsistency Bug with SLES 11 SP2
HANA Alert 2 - Disk usage
HANA Alert 30 - Internal disk-full event
HANA Alert 50 - Number of diagnosis files
HANA Alert 51 - Size of diagnosis files
Volume Size vs. Disk Usage
In general, two types of storage space utilization have to be distinguished: Disk (file system)
usage and space consumed (internally) in the volumes attached to the HANA database.
Both aspects are summarized in the overview tab in HANA studio
Total Disk Usage represents the file system point of view and is identical to what
OS commands like 'df' are displaying:
df -h
/dev/sdd1 99G 32G 63G 34% /HANAmnt/data/HDB
/dev/sde1 50G 3.7G 44G 8% /HANAmnt/log/HDB
/dev/sdf1 30G 19G 9.3G 67% /HANAmnt/HDB
Volume size refers to the size of the HANA volumes on file system level
This is effectively the output of the following SQL statement:
select v1.host, v1.port, v1.service_name, v1.volume_id,
v1.subpath, v2.data_size, v2.log_size, d.disk_id,
d.device_id, d.path, d.usage_type, d.total_size,
d.used_size
from (
( m_volumes as v1 join M_VOLUME_SIZES as v2 on v1.volume_id = v2.volume_id )
right outer join
m_disks as d on d.disk_id = v2.disk_id
)
If there is a huge difference between “Volume Size” and “Disk Usage”, this would
indicate that a lot of space on files system level is consumed by files not related
to data volumes, log volumes or trace files. In this example, this is true for the
trace volume. All trace files together just consume around 100 MB, but consumed
disk space is actually close to 19 GB:
select sum(file_size) / 1024 / 1024 as size_mb
from m_tracefiles
SIZE_MB
117,222
In cases like this, the 'du' command, executed against the file system in question,
can reveal the root cause:
259M/HANAmnt/HDB/hdbstudio_update/repository/plugins
620M/HANAmnt/HDB/hdbclient
797M/HANAmnt/HDB/global/hdb/custom/config/lexicon/lang
7.4G/HANAmnt/HDB/exe/linuxx86_64/HDB_1.00.51.374135_1005022
7.9G/HANAmnt/HDB/exe/linuxx86_64/HDB_1.00.55.376513_1078043
The file system containing the trace files also contains the executables of HANA software
(present and previous version), hence the difference in volume and disk usage.
Volume Size and Volume Utilization
Data Volumes
If the percentage of unused space is high, shrinking the volume can be considered.
The relevant system management command is outlined in SAP HANA Reference:
ALTER SYSTEM RECLAIM DATAVOLUME.
In order to find out which column store tables are occupying most of the space inside the volumes
you can use the following SQL statement:
select top 30 table_name, sum(physical_size)
from M_TABLE_VIRTUAL_FILES
group by table_name
order by 2 desc
Log Volumes
The number of log segments that consume space can either be determined using HANA Studio
or using SQL statements against view M_LOG_SEGMENTS
SAP note 1679938 explains how to free space in case that the log volume is full.
Trace Files
If the trace files are contributing to the disk usage of the file system where they are located
(see section ‘Volume Size vs. Disk usage’), then 2 cases have to be distinguished:
Space consumed by a large number of files that are not exceptionally large or
Individual files that are significantly larger than expected.
The number and size of existing trace files is influenced by the following factors:
(1) Activated trace components. This can be verified by using the following SQL statement:
select distinct("SECTION") as "TRACE"
from SYS.M_INIFILE_CONTENTS
where (("SECTION"='trace')
or ("SECTION"='sqltrace' and "KEY"='trace' and "VALUE" = 'on')
or ("SECTION" like 'traceprofile_%')
or ("SECTION" = 'expensive_statement' and "KEY"='enable' and "VALUE" = 'true'))
and "LAYER_NAME" = 'SYSTEM'
(2) Incorrect configuration of trace file rotation:
The relevant parameters are maxfiles and maxfilesize
(see HANA administration guide chapter 'Configuring Trace File Rotation')
ref: 1870858 - HANA Alerts related to files system utilization
If the log_mode is set to legacy the logvolume keeps all log segments since the last full backup in the logvolume.
If no backups are performed the logvolume can run out of space.
Solution
1. Stop the database:
HDB stop
2. Change into the folder mnt00001 of the logvolume (Default: /usr/sap//global/hdb/log):
cd /usr/sap//global/hdb/log/mnt00001
3. You have to move one of the logvolumes temporarily to another volume where enough space is available.
You should free at least 2 GB of space to ensure that the database has enough space to start.
To find out the space consumption of each volume execute:
du -sh
4. Move a volume which consumes at least 2 GB of space (e.g. hdb00003)
to a volume with enough free space,
e.g. to the data volume (Default: /usr/sap//global/hdb/data):
mv hdb00003 /usr/sap//global/hdb/data
5. Create a symbolic link to the new folder in the old location:
ln -s /usr/sap//global/hdb/data/hdb00003 /usr/sap//global/hdb/log/mnt00001/hdb00003
6. Start the database (HDB start) and perform a backup
7. Use the following SQL-Statement to clean up the logvolume:
ALTER SYSTEM RECLAIM LOG;
8. Stop the database again and remove the symbolic link:
rm -f /usr/sap//global/hdb/log/mnt00001/hdb00003
9. Move the log volume back to its original location:
mv /usr/sap//global/hdb/data/hdb00003 /usr/sap//global/hdb/log/mnt00001
10. Start the database (HDB start)
ref: 1679938 - Log Volume is full (log mode = legacy)
HANA content is organized in software components called "delivery units" ("DUs").
DUs are distributed in the physical form of DU-TAR-GZ-files.
As for texts (e.g. view column descriptions), the DU TAR-GZ-files contain only the texts in the original language (typically, English), but no translated texts (e.g. for German, Chinese etc.).
Translated texts for DUs are distributed in the form of Languages-DU-TAR-GZ-files.
At the moment (HANA SP5/revision 45), those cannot be imported using the SAP HANA Studio.
To deploy translated texts, the command line tool "Regi" can be used.
"Regi" is bundled with the HANA client installation (it is also bundled with the HANA server installation).
Note that Regi is only included in the 64bit Windows and Linux versions of the HANA client.
Here is a step-by-step guide to deploy translated texts using Regi:
1) Start a command line window and navigate to the folder of your HANA client installation.
On Windows, the default location is:
C:\Program Files\sap\hdbclient
On Linux, the default location is:
/usr/sap/hdbclient
2) Set the user, password, host and port of the connection to the HANA system using environment variables.
On Windows, the commands are:
set REGI_HOST=:315
set REGI_USER=
set REGI_PASSWD=
where "" is a two-digit number identifying the instance number of the HANA system.
On Linux (bash), replace the command "set" with "export":
export REGI_HOST=:315
export REGI_USER=
export REGI_PASSWD=
3) Import the Languages-DU-TAR-GZ-file with the following command:
regi import languages
Note that further help for the REGI command can be accessed as follows:
regi help import languages
ref: 1805967 - Deployment of Translated Texts for Delivery Units
This is the procedure of the homogeneous system copy of SAP HANA using backup and recovery.
The prerequisites are that you have installed an SAP system with SAP HANA as the primary database
and you have installed SAP HANA Studio in which the database that is to be copied is set up as a system.
The version of the SAP HANA database server must be at least Version 1.00 Revision 28.
For carrying out the system copy, you must use SWPM 1.0 Support Package 1 or higher.
The scenario for this reference example is: You want to copy the BW system PRD connected to the database
PR1 to the BW system DEV connected to the database DV1. The name of the ABAP schema in PR1 is SAPPRD.
Therefore, PRD is the source system (the system that is to be copied), PR1 is the source database (the database system that is to be copied), DEV is the target system (the copied system), and DV1 is the target database (the target database system).
SAPPRD is the (database) schema (the schema name) or the database user of the system.
Procedure
1. Use SAP HANA Studio to create a database backup.
For a full description of the procedure and the SAP HANA backup options, refer to
SAP HANA Database - Backup and Recovery Guide on SAP Help Portal.
a) In SAP HANA Studio, right-click the database system that is to be copied and choose "Backup...".
b) If possible, select "Complete Data Backup" as the Backup Type and "FILE" as the Device Type.
Under Location, enter a directory in which the backup files are to be stored and enter a prefix for the backup files under Backup File Prefix.
Ensure that the backup directory contains sufficient free space for the backup and that no backup that has the same prefix already exists there.
Then choose "Next".
c) Check your entries and choose "Finish" to start the backup or choose "Back" to correct your entries.
d) Wait until the backup has been created and then close the dialog box.
2. Transfer the backup to the target system.
a) Copy all backup files to the standard backup directory of the target database system.
The backup files are saved in the standard backup directory of the source database system and begin with the prefix entered in step 1.
The standard backup directory is defined in the database parameter
global.ini -> persistence -> basepath_databackup.
3. Use SWPM to import the backup.
a) Start SWPM.
b) Choose -> -> System Copy -> Target System -> ->
Based on -> Standard System.
c) Follow the dialogs.
d) In the dialog SAP System Database,
choose Homogeneous System Copy (SAP HANA-specific Backup/Recovery).
e) In the Database Schema dialogs, enter the schema names and the passwords that match the data in the backup.
For example, if you install a system DEV and use a backup of the system PRD for the installation, you must specify SAPPRD as the schema in the dialogs but not SAPDEV.
The same applies to the DBA Cockpit schema.
f) In the dialog Database Recovery, enter the name (prefix) of the backup.
In addition, you can specify whether you want the system to check whether the backup exists.
If this check is deactivated and the backup does not exist, the installation will terminate with an error at a later time.
You also have to specify the password of the adm user of the database and the SAPControl URL.
The system prefills the SAPControl URL; it usually does not have to be changed.
4. Installation
a) After you complete the following dialogs, you can start the installation.
b) During the installation, the system may issue an error message such as:
Error during execution of HdbCmdClazz 'ABAP_PRE_LOAD_EXECUTION'
com.sap.hdb.core.main.cmd.HdbCmdMain
Error during execution of SQL command:
ALTER SYSTEM ALTER CONFIGURATION ('indexserver.ini', 'SYSTEM' ) SET
('distribution', 'client_distribution_mode') = 'statement' WITH RECONFIGURE SAP
In this case, a new license is required for the target database.
This license must be implemented manually using the SAP HANA Studio.
To do this, right-click the target database system in the SAP HANA Studio and choose "Properties".
In the subsequent dialog, choose "License" and then choose "Install License Key".
Select a valid license for the target database and confirm your selection with "OK".
As soon as the license has been installed successfully, you can proceed with the installation in
the SWPM by choosing "Retry".
5. Additional adjustments after the installation
a) Check the secondary database connections.
After you copy the database, the target system has the same database connections (for example, for the DBA Cockpit) as the source system; this may cause problems.
Therefore, you must check the database connections in transaction DBCO and adjust them if required.
b) Check the RFC connections.
After you copy the database, the target system has the same RFC connections as the source system; this may cause problems.
Therefore, you must check the RFC connections in transaction SM59 and adjust them if required.
c) Check the spool configuration.
After you copy the database, the target system has the same spool configuration as the source system.
Therefore, you must check the spool configuration in transaction SPAD and adjust it if required.
6. The system copy is now complete.
ref: SAP Note 1763333 - Homogeneous system copy on SAP HANA
Data required for the license key request:
Installation number, system ID, system type, SAP product, product version, hardware key, operating system, and main memory amount
The incoming requests are processed automatically within a few hours (maximum processing time: three working days).
Determining the hardware key and the SID
This information is displayed in the SAP HANA Studio on the attributes page of the system.
Perform the following steps to read this:
In the SAP HANA Studio, select your system in the Navigator and choose the option to display the attributes in the context menu of the system.
In the navigation area, choose the option to display the license.
Note that you require the system authorization LICENSE ADMIN to access this page.
License key request
Proceed as described below to create a license key request on
SAP Service Marketplace at service.sap.com/licensekey.
On the "Request license keys" tab page, choose either "Search an installation" or "select it from a list" and search for an ERP or an SAP NetWeaver installation.
In the installation list, you can enter the installation number directly in the blank field in the column for the installation.
Alternatively, you can also use the installation search. Here, the system displays all of the installation numbers for which you have to request the license key authorization.
License key request for a new SAP HANA database
Choose "New system" and enter all required data.
To proceed to the hardware data, choose "Continue".
Enter your hardware key, the amount of main memory (in GB), and the validity date.
You can also split your acquired SAP HANA license (total amount) over several instances (also over several installation numbers).
Choose "Save" to save your entries.
Choose "Continue" to display all the system data again.
Enter your e-mail address and choose "Send".
If the required amount of main memory exceeds the licensed amount, the license key is reduced by the difference accordingly so that, at the most, the licensed amount can be called.
If the entire amount has already been called, only temporary keys are issued.
Therefore, always check how much main memory you have currently licensed by contract.
For test and development systems, you can request license keys as required.
These license keys are permanent and contain the licensed memory amount number for each request at the most.
Downloading the license key
In this system overview, click the required system ID under an installation number.
Choose "Continue" and select the relevant entry.
Choose "Display license script" and "Download to PC".
You can then choose "Open - File - Save as" to save the license key file directly to your host.
Installing the license key
You can install a permanent license key, which you received by e-mail or download, in your SAP HANA database only by using SAP HANA Studio.
The prerequisite for this is a database user with the system authorization LICENSE ADMIN.
Carry out the following steps to do this:
In SAP HANA Studio, navigate to the properties page of the database for which you have requested the license key.
On the properties page, choose the button for installing the license key.
Navigate to your license file and select it.
After a successful confirmation, the properties page of the system is updated with the new license information if this is valid for the database.
Another option of installing a license key is to use the SQL command line of SAP HANA Studio.
In this case, you also require a database user with the system authorization LICENSE ADMIN, who can install a license key with the following SQL command:
SET SYSTEM LICENSE ''
Note that you must use the complete content of the license file, including the line breaks, for this command.
If the specified license data is correct and corresponds to the database (hardware key, system ID), the license is updated in the database, and you can verify it in the system view M_LICENSE.
Emergencies
If a production standstill occurs, you can create a message with priority 1 under component XX-SER-LIKEY to request a temporary license key that is valid for one week.
The following data is required for the license key to be issued:
System ID, hardware key, and installation number.
The key is issued immediately. Request a permanent license key within this week.
ref: Note 1644792 - License key req./installation SAP HANA databases (HANA SPS3)
With SAP NetWeaver 7.4 SP04 the components of the following instances based on Application Server Java are released for use with SAP HANA Database 1.00:
Application Server Java
NW Product Description
Adobe Document Services
PDF Export
Demo Applications
BPM
EP Core - Application Portal
Enterprise Portal
BI Java
Advanced Adapter Engine
PI Adapter Engine (Java EE)
Enterprise Services Repository
Composite Application Framework
Composition Platform
Guided Procedures
AS Java Extensions
Development Infrastructure
Please Note: there are some warnings and conditons
Make sure to check SAP Note:: 1849151 - SAP NetWeaver 7.4 Java On HANA Release Information
And the Platform Availability Matrix
Latest Update: 14th June 2013 - this can be subject to change
Definition: "SAP HANA appliance": "SAP HANA appliance" means a single SAP HANA node server or a scale-out cluster of multiple server nodes within one logical SAP HANA system.
This described scenario has also been called Multiple Components One System (MCOS) in regards to other SAP applications. The term "SAP HANA system" is used interchangeably with the term "SAP HANA appliance".
In this context, some new terms are introduced: "Multiple DBs One HANA" (MDOH), and "Multi-SID".
SAP's Support Status:
Important: When considering the support status, bear in mind that the items in the "special considerations" section which follows must be taken into account.
While SAP HANA is delivered by the hardware partners with one DB installed, additional DBs can be installed by an SAP consultant who has access to SAP internal documentation for this purpose.
SAP does support multiple DBs on a distributed (multi-node) non-production SAP HANA system as an SAP-managed project solution.
Additionally., SAP plans to offer generally available support for this in the near future; this note will be updated when this status changes.
Special Considerations:
Please ensure that your system is sized appropriately for any additional DBs that you wish to deploy. SAP recommends working closely with your hardware partner to ensure adequate capacity planning.
Please be aware that running multiple DBs on one SAP HANA system may impact performance of various types of operations, as contention for memory resources may occur.
This performance impact may appear despite adequate sizing.
If performance issues arise when running multiple DBs on one SAP HANA system, stop all but one of the DBs and see if the performance issue persists.
ref: SAP Note 1681092 - Support for multiple SAP HANA databases one HANA
There is a very interesting discussion on this subject over here.
Also note this means that SAP generally does not support a packaged application from SAP residing together in the same SAP HANA system as the scenario "Custom Data Marts".
The following applications can run together with other applications on one Production SAP HANA system
"Custom Data Marts"
SAP CO-PA Accelerator
ERP Operational Reporting with SAP HANA
SAP Finance and Controlling Accelerator: Material Ledger
SAP Finance and Controlling Accelerator: Production Cost Planning (aka"CO-PC Accelerator")
"SAP NetWeaver BW powered by SAP HANA (BW on HANA). For more information,refer to SAP note 1666670.
SAP Suite Accelerator (aka "Application Acelerator")
Things to take into consideration when running multiple components on one HANA system
Upgrades and patches:
- Applying SAP HANA support packages and revisions will affect all applications residing within the SAP HANA database or utilizing other SAP
HANA components.
- Applications deployed on a single SAP HANA may have to be patched and/or upgraded at the same time due to direct dependencies on SAP HANA patch levels.
- Packaged Applications releases, support packages or patches may have direct dependencies on SAP HANA revisions, patches and support packages.
- Packaged applications may have independent release cycles to SAP HANA and may have independent timeline for support of SAP HANA release levels and support packages.
Resource allocation and capacity planning:
If you are running multiple applications on one SAP HANA system, there is a risk that one application's processing operations could consume a
significant amount of available CPU and memory resources, thereby reducing the amount of such resources available for the processing operations of the other applications at a given point in time.
Such contention for system resources may negatively impact performance of any one application's processing operations.
Therefore, you must consider one application's impact on another.
An additive sizing approach is required for all applications running on a single SAP HANA system.
This means that when performing capacity planning, you must determine the resource allocation needs for each application, then add them together to estimate the required sizing for your SAP HANA system.
It is important to avoid underestimating sizing, as this will help mitigate the risk of performance issues due to contention for system resources.
System Management, Workload Management and Monitoring:
SAP HANA monitoring, debugging and support tools are not individual application or schema-specific, they are supported at the SAP HANA database level.
SAP HANA scale out distribution options (including node fail over) is offered at the database level, not at an individual application or component level.
Stopping and starting the database is a database wide operation.+
Backup &; Recovery:
- Backup and Recovery is currently supported only at the SAP HANA database and component level (e.g. SLT, Data Services) and not at a specific
application level in an MCOD scenario.
This means that a point-in-time recovery for the SAP HANA system will impact all applications residing on that SAP HANA system.
Please note that various point-in-time recovery scenarios (recovering SAP HANA or recovering a connected SAP Business Suite system) will likely entail re-initializing and re-loading replicated data into SAP HANA.
Depending on data volume, this could be a time-consuming process.
Lifecycle management:
- All applications should be isolated from each other in separate database schemas.
- System copies of applications are not available at an individual applications level.
High Availability &; Disaster Recovery:
- HA and Disaster Recovery is currently supported only at the SAP HANA database and component level (e.g. SLT, Data Services) and not at a specific application level.
Security:
- Customers are responsible for implementing security models to protect and/or isolate application artifacts (e.g. users, data, metadata, database
objects)
Production support systems ("Break fix systems") and Quality assurance system:
- It is recommended that "break fix" and QA systems have the same configuration as the production system so that customers can replicate
issues and test corrections in systems that are representative of the production environment.
ref: 1661202 - Support for multiple applications on SAP HANA
and the complimentary OSS Note
1826100 - Multiple applications SAP Business Suite powered by SAP HANA
1. Create a data backup of the source database.
More information: SAP HANA Backup and Recovery Guide, available on SAP Help Portal
2. Check that there is sufficient disk and main memory capacity on the target database host.
The available main memory should be able to hold at least twice the size of your row store data.
To find out how much this is, execute the following SQL statement:
select sum(allocated_page_size)
from M_CONVERTER_STATISTICS
where type = 'RowStoreConverter';
The result is the total size in bytes.
Multiply this value by 2 to obtain the absolute minimum size of main memory required.
More information: SAP Note 1514966
3. Configure additional indexservers on the target database.
The total number of index servers on the target database must be the same as on the source database.
(The number of index servers is normally the same as the number of hosts.)
Thus additional index servers must be configured in your target database.
To configure additional index servers, you need to change the daemon.ini file.
To change the daemon.ini file, execute the following statement (for example, with instanceids 40 and 42):
alter system alter configuration('daemon.ini','system')
set('indexserver.c','instanceids')='40,42' with reconfigure;
This statement adds entries such the following to the daemon.ini file:
[indexserver.c]
instanceids = &,&,...
&and &must be even numbered ports starting at 40 with an interval of 2.
For example, for a triple host source system, you would need two additional instanceids and you can use 40, 42, ...
4. Check that the additional index servers are configured in your target
instance.
In SAP HANA studio, go to the Landscape tab.
The additional index server processes should be dislayed.
The description indicates which are the added index servers ("HDB Tenant Index Server on Port 300&", where & corresponds to the instanceids you elected earlier).
5. Shut down the target database.
6. Move or delete available data and log backup files from the target database.
7. Move or delete the backup catalog files, if they exist:
$(DIR_INSTANCE)/../SYS/global/hdb/metadata/
BackupCatalog.xml
$(DIR_INSTANCE)/../SYS/global/hdb/metadata/
BackupCatalog_Recover.xml
8. Copy the data backup files and, optionally, the log backup files from the source database to the corresponding directories in the target database.
Note:
The directories for data and log backup are defined by the parameters basepath_databackup and basepath_logbackup in the global.ini file.
9. Start the recovery of the target database system.
In SAP HANA studio, select the target database system and choose Recovery... from the context menu.
10. Select a recovery type.
For example, Recover the database to its most recent state.
11. Choose Next.
12. Check that the specified default locations of the data and log backup files for the target database matche the backup files that you have copied.
13. Select the option Initialize log area and confirm the warning.
14. Select the option Install new license and specify the license file.
15. Choose Next.
16. Choose Finish to start the recovery.
ref: SAP Note 1749467 - Copying SAP HANA From a Multiple- to a Single-Host System
[Latest update: 17/01/2014]
HANA multi-node configurations are used for different purposes:
a) High-availability with standby nodes
b) Scaling memory (scale-out)
See also chapters 10 and 14.3 of the HANA Administration Guide (http://help.sap.com/HANA/HANA_admin_en.pdf ).
The following restrictions for usage of
SAP EHP 6 for SAP ERP 6.0, version for SAP HANA
SAP EHP 2 for SAP CRM 7.0, version for SAP HANA
SAP EHP 2 for SAP SCM 7.0, version for SAP HANA
apply:
a) High availability scenarios in a multi-node cluster with a standby-node are released for the Suite on HANA, but are restricted to the simplest case of two servers (one worker, one standby) only, since scale-out in general is not yet supported for Suite on HANA.
b) Scale-out scenarios with multiple worker nodes to scale memory are not yet released for Suite on HANA.
It is recommended to scale-up memory by using a hardware configuration that maximizes the available database memory.
ref: SAP Note 1825774 - SAP Business Suite Powered by SAP HANA - Multi-Node Support
ref: SAP Note 1950470 - Hardware Prerequisites for Business Suite on SAP HANA Scale Out
Attention has been drawn to this document by +HANA Distinguished Engineer and SAP Mentor Tomas Krojzl .
This document is guidance for large Customers implementing HANA.
The document is a must read for all HANA Basis Architects involved in HANA implementations.
High-end customer landscapes based on SAP HANA+ - http://public.dhe.ibm.com/common/ssi/ecm/en/xsw03129usen/XSW03129USEN.PDF
You have received new hardware from your Hardware-Partner for your SAP HANA appliance or recently bought a new SAP HANA on top of new hardware.
When installing the newest SAP HANA Revision the hardware check fails for your hardware configuration.
Beginning with SAP HANA Revision 18 a new Hardware check is mandatory when executing the SAP HANA Installer.
The reason for this check is the elimination of problems due to wrong or unsupported Hardware configurations in combination with the SAP HANA appliance.
Since hardware validation is a continuing process as new hardware becomes available on the market, the check for allowed configurations will change over time to include newly certified hardware.
When installing any SAP HANA Revision your new hardware may not be recognized by the check, because it was not yet included at the corresponding release date.
The solution to this problem is, to go to SAP Note
'1658845 - Recently certified SAP HANA hardware not recognized'.
Attached to that note you will find the most recent version of the Hardware Check including all validated hardware.
Before triggering the installation again please replace the file
'HANAHwCheck.py'
in the
'server\linuxx86_64\SAP_HANA_DATABASE\server'
subdirectory of your SAP SMP download with the version attached to this note.
ref: OSS 1658845 - Recently certified SAP HANA hardware not recognized
For the receiving SLD to display the HANA database as the CIM instance "HANA Database System" (SAP_HDBSystem), the SLD must meet the prerequisites specified in Note 1649323.
If the SLD does not meet the pre-requisites then it is necessary to apply the Java patches to the SLD according to SAP Note 1649323.
To successfully register the SAP HANA Database in the SLD use either:
- sldreg, following the HANA Installation Guide with SAP HANA unified installer, page 46, chapter 2.3.7.5 Configuring a SAP HANA System to Connect to System Landscape Directory (SLD)
You can use the HANAconfig program to configure a SLD connection.
For more related information, see SAP notes 1673424 and also 1649323.
Make sure you meet these prerequisites:
You identify the existing system, created by an installation or a file system copy, that is to be configured with SLD.
You mount the DVD with the HANA configuration tool, which you can find at the following location on the SAP HANA DVD:
&\DATA_UNITS\HANA_IM_LINUX__X86_64
You are logged in as the root user
Configuring Connection to SLD in GUI
In GUI mode, you are prompted to enter the required parameters.
If you specify the host name, make sure that it is fully qualified, such as:
myhost.sap.com (not just myhost).
1. Connect to the system with an X server client to enable GUI system access.
2. Open a root shell and go to the directory where you mounted the SAP HANA DVD, by entering a command like the following:
cd /mnt/ &DATA_UNITS/HANA_IM_LINUX__X86_64
3. Call the script in GUI mode:
./HANAconfig.sh –gui
The wizard starts.
4. Choose Configure SLD Connection, then Next . Follow the instructions of the wizard.
Configuring Connection to SLD in Interactive Mode
Before you continue, make sure that you meet these prerequisites:
You are logged on to the host where the server software is installed.
You are logged on as the root user.
In interactive mode, the program queries all essential parameters that are not defined using command line options and if invalid parameters are entered the program interrupts the process.
1. Open a root shell.
Running SAP HANA Unified Installer
2. Go to the following directory:
cd /mnt/&DATA_UNITS/HANA_IM_LINUX__X86_64
3. Enter the following command:
./hanconfig.sh sldconfig
4. Specify the data requested by the program.
Configuring Connection to SLD in Batch Mode
Before you continue, make sure that you meet these prerequisites:
You are logged on to the host where the server software is installed.
You are logged on as the root user.
In batch mode, you specify all required values using command line options.
If you omit required values or specify invalid values the program interrupts the process.
1. Open a root shell.
2. Go to the following directory:
cd /mnt/&DATA_UNITS/HANA_IM_LINUX__X86_64
3. Run the HANAconfig program in batch mode using the sldconfig task:
./hanconfig.sh --batch sldconfig &
For the list of all available options, you can call the tool help, by entering ./HANAconfig --help
4. Specify the data requested by the program.
- RZ70
ref: 1871902 - Cannot find registered HANA DB system in the Technical Systems view of the SLD
SAP HANA On-site Configuration does not support multiple systems installed on appliance.
This limitation is related only to the SPS04 release.
In SPS05 the On-site Configuratio supports multiple systems on single SAP HANA appliance.
If you need to configure SLD connectivity for multiple systems in SPS04 the best approach is to update the systems and the On-site Configuration tool
to SPS05. In this way you will have system and tools with the correct and integrated behaviour.
If this is not possible you can configure the SLD connection manually by using the sldreg tool.
Example:
su &adm -l -c '"/usr/sap/&SYS/exe/hdb/sldreg" -configure
"/usr/sap/&SYS/global/slddest.cfg"'
Replace &with the SID of your system.
For addtional SLD configurations you can take a look in the SAP HANA
Database # Administration Guide on http://help.sap.com/HANA_appliance
ref: SAP Note 1787755 - SLD configuration for multiple SAP HANA databases in SPS04
You experience memory related issues, e.g:
Checkout the manual steps and useful SQL statements in the OSS Note:
The setup and configuration for Diagnostics Agent on SAP HANA Systems is described in the SAP Note:
The optimal settings for Suse Linux are described in SAP Note:
1824819 - SAP HANA DB: Recommended OS settings for SLES 11 / SLES for SAP Applications 11 SP2
Since SAP HANA Platform SPS08, RedHat Linux is now aa supported Operating System.
More information is available in
2009879 - SAP HANA Guidelines for RedHat Enterprise Linux (RHEL) Operating System
2004651 - SAP HANA Platform SPS 08 Release Note
see SAP Note:
1496410 - Red Hat Enterprise Linux 6.x: Installation and Upgrade
see SAP Note:
2013638 - SAP HANA DB: Recommended OS settings for RHEL 6.5
There is an SAP Note:
2009879 - SAP HANA Guidelines for Red Hat Enterprise Linux (RHEL) Operating System
Pay particular close attention to this SAP Note because it contains a very nice pdf attachment entitled:
Red Hat Enterprise Linux RHEL 6.5 configuration guide for SAP HANA
Thanks to paul.aschmann for this contribution
This is helpful if you would like your web app to be available on the default/standard HTTP port
iptables -t nat -A PREROUTING -p tcp --dport 80 -j REDIRECT --to-ports 8000
Thanks to Paul Aschmann for this contribution
Enable the developer mode which provides details on execution/runtime errors when helps during development (rather than getting generic 500 errors).
From HANA Studio, open Administration
-> Configuration
-> xsengine.ini
-> httpserver
-> developer_mode = true
see SAP Note 1971603 - How to check SLT Trigger Status
SAP Note 1984858 - Network ports used by SAP HANA database for internal/external communication
SAP Note 1969397 - Troubleshooting Guide for SAP Business One 9.0, version for SAP HANA
2031547 - Overview of SAP-certified 3rd party backup tools and associated support process
SAP Notes:
1957450 - SAP HANA Backup Support Process for CommVault Simpana
1970558 - SAP HANA Backup Support Process for HP Data Protector
1957450 - SAP HANA Backup Support Process for CommVault Simpana
1913568 - SAP HANA Backup Support Process for Symantec NetBackup
1913500 - SAP HANA Backup Support Process for IBM Tivoli Storage Manager
1986958 - Common issues / Configuration recommendations for Symantec NetBackup for SAP HANA
Internet of Things IoT represents a strategic direction for SAP with HANA.
Published on Friday 8th May, is the SAP Note for the SAP HANA Internet of Things IoT Edition:
2008304 - SAP HANA platform, Internet of Things (IoT) edition 1.0 Central Note
2010194 - SAP HANA platform, Internet of Things (IoT) edition 1.0 SP01 Release Information Note
SAP have published a fantastic SAP Note full of SAP HANA Frequently Asked Questions, it is here:
SAP have now published the FAQ for SAP Hana Memory
1999997 - FAQ: SAP Hana Menory
Contributed with the permission of fang%2Cyuan
For SAP HANA developer, SAP HANA Studio is a common client tool to develop or manage on SAP HANA.
At the same time, SAP HANA XS engine also provides a web-based development environment. Only using a web
browser, you can perform some simple development. When you either cannot install and use SAP HANA studio
locally or you need to develop from a remote location, you can use this IDE.
This web-based IDE is called SAP HANA Web-based Development Workbench, which contains four modules:
The whole blog is here, The Configuration and Usage of SAP HANA Web-based IDE
There is an excellent blog written by xing.jin2 where all of the steps to preparing a SAP Hana
system for SAP Lumira installation are explained and with screenshots.
Prepare a HANA System for Lumira Server Install
Amazon Web Services is a popular Cloud based hosting option for SAP Hana.
The main url for information on SAP Hana on AWS Amazon Web Services is:
Amazon Web Services have produced some excellent introduction guides:
SAP Hana on AWS Amazon Web Services Quick Start Guide
SAP Hana on AWS Amazon Web Services Implementation and Operations Guide
SAP Hana on VMWare vSphere, the full documentation is here:
https://www.vmware.com/business-critical-apps/sap-virtualization/sap-hana.html
FAQ - SAP Hana on VMWare vSphere for Production Environments
SAP Hana on VMWare vSphere Best Practices Guide
There is a fantastic SAP Note from SAP explaining all of the most important information regarding
HANA Support for VMWare Virtualized environments.
Visit the SAP Note regularly, because it is the most up to date source of information in this are:
1788665 - SAP HANA Support for VMware Virtualized Environments
The SAP HANA rules framework provides a set of tools enabling business users and application developers
to build decision logic based on the organization's data. This rule engine applies rules and actions as defined
by end users without affecting how the application runs.
Using the SAP HANA rules framework, you can plan your decision support application development by:
2014934 - SAP HANA Rules Framework 1.0 Release Note
SAP MarketPlace contains all of the implementation guides and security information.
related SAP Notes available in one place on the Internet
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
13 | |
5 | |
4 | |
4 | |
3 | |
3 | |
3 | |
3 | |
3 | |
2 |