Monitring
Monitring
Monitring
Proactive monitoring of the SAP systems, will help to understand issues in advance & helps us to
take corrective actions and thus will lead to lesser downtime of the systems, improving
profitability of the business organisation.
Monitoring can be focused on ABAP stack(including database checks), Java Stack and Oslevel
checks.
This transaction code will be useful to view the processes that are running currently in an sap
insance. In this view you can check whether there are free workprocesses to execute the
processes. If all the workprocesses are in running state and no work process is idle it means that
wait times will increase for the processes that are waiting in the dispatcher queue leading to
performance degradation. If you find that there are no free workprocceses for maximum times
that you may consider, increasing the number of workprocesses.
This transaction code will be useful to view the processes that are running across all
instances/application servers of a SAP system. Similar to SM50 checks can be done in this
transaction as well.
This transaction code will be useful to view all the hostnames and application servers status. If any
application server is down, the same can be identified using status of the server column. We can
also figure out different Message types (Dialog, Batch, Update, Upd2, Spool, ICM etc) configured
for the respective servers.
This transaction code will be useful to view all the sap locks that are present in the system. As part
of monitoring, we need to look for any old sap locks that are more than 1 day. If any such locks,
we need to analyse the reason for that lock for such longer duration and take actions accordingly.
A lock can be set for such a long duration due to a long running background job or a lock is not
released due to an application error or a program terminated abruptly but lock not released etc.
This transaction code will be useful to view all the abap dumps that have occured in the system on
a given day. As part of daily monitoring, it is the responsibility of the basis administrator to
analyse the dumps and take necessary actions to avoid issues.
Some of the examples of abap dumps are timeout issue, database space issue, spool overflow
issue etc
This transaction is useful to view the log of the sap system for various operations. This log will be
very useful to identify various issue in advance and to take necessary measures. System log is the
place to check out for any timeout, network issues, database space issues, message server issues,
spool overflow, locktable overflow etc issues.
Additional details :
This transaction is useful to figure the status of update system. Incase an update is inactive we
can figure out the same from this transaction and necessary action can be taken and update can
be activated again.
Update got deactivated. what is the reason for update deactivation? How to activate the update ?
SM14 transaction can be called internally from SM13. These both transactions are useful for
update administration.
In SM13, you can select status (canceled, to be updated, v1 executed, v2 executed, all ) and time
interval during which you would like to view the status execute to check the overview of updates
as per the status and time interval selected.
Buffer statistics like hitratio, swaps, db access details, size of buffer and free size of buffer
etc
Important statistics related to Roll area, Page area, Extended memory and heap memory
As a basis administrator, it is our responsibility to ensure there is more hit ratio for the buffers and
less swaps to ensure efficient performance of the sap system. In case you see there are more
swaps and less hit ratios for most of the buffers, then tuning buffers to be carried out to ensure
optimal performance.
DB12 (Backup logs) : This transaction is useful to check the details of
last successful backup
overview of database backups ( Success / failure of backup with log details)
Archiving directory status (Free space of oraarch )
Overview of redolog files ( Number of redologs that are not yet backed up)
overview of redolog backups (Success / failure of backup with log details)
DB13 (DBA Planning calender) :
This transaction will be useful to schedule various database backups & clean up jobs like ( whole
database backup offline/online, Full backup online/offline, incremental backup offline/online,
redolog backup, update statistics, check db, cleanup logs, compress database, verify database,
initialize tape and validate structure jobs).
In this transaction, you can also check the status of every job that was scheduled and can
reschedule in case of failures.
Database backup
Redolog backup
BRSPACE log (extend tablespace issues etc)
BRCONNECT operations (Update optimiser statistics , database check etc)
As an sap basis administrator it is our responsibility to check and ensure backups and other
cleanup jobs are successful everyday. Incase of failures, should figure out root cause and take
actions like rescheduling and ensure these jobs are successful.
This transaction will be useful to have an overview of jobs with different statuses.
As part of daily monitoring, SAP basis administrator should use this transaction to findout canceled
jobs and active jobs(for eg: long running - more than 24hrs etc).
Incase of canceled jobs, root cause for the failure to be figured out from the logs of the respective
job and to be actioned by rescheduling etc.
Incase of long running jobs, we need to figure out the reason for long running and action them
accordingly.
In SM37, using extended job selection option, we can even select the jobs based on start
condition, steps (like abap program, external command or external program), period etc
ST04 (Database alert logs and performance) :
This transaction will be useful for (oracle) database administration. In this screen, goto Alerts and
drill down further. Click on "Database Check" to find out any errors or warnings related to
database like MISSING_STATISTICS, STATS_TOO_OLD, LAST_BACKUP_FAILED,
LAST_ARCHIVE_FAILED etc. After going through the error or warning in details take necessary
corrective actions based on the error like running update stats again, re-triggering backup etc
Under Alerts, you can view Alert monitor which will summarize status of the database under
different heads like
Space Management
Performance
Backup/restore
SAP Consistency
Health
Drill down on each of these to find out potential problems. These are color coded for ease of
administrator (Red for errors, yellow for Warnings and Green for OK status)
For Eg: If PSAPSR3 tablespace is >90%, you can see Space management in red color. Then it is
the responsibility of Basis administrator to take necessary actions on the same.
This transaction is useful to find out the status of spool request and output request. In SP01
transcation, you can list the spool requests or output requests between a given interval.
In the list generated, you can check out the status of spool requests and findout any errors by
drilling down further.
For eg: if so many spools are in waiting status, find out whether output device is available or not.
If many spool are in error status, figure out if there is any network issue and take necessary
actions.
If customers complain that they are not able print anything from SAP, check out whether there is
any spool overflow.
What is spool overflow ? How to troubleshoot spool overflow issue ?
SXI_Cache : This Tcode is specific to XI or PI system. This Tcode is used to findout whether cache
refresh is happening or not. Incase if cache refresh is happening successfully, it will indicate the
same in green color. Otherwise it will be in red indicating a problem with cache refresh.
If there is a problem with cache refresh then basis administrator has to troubleshoot the same.
SLDCHECK : This Tcode will be useful to figure out whether connection to the SLD system from
the system on which you are testing is fine or not. In case the connection is fine, all checks will
appear in green. Incase of any issues, it will appear in red or yellow and then basis administator
has to troubleshoot it and make sure SLDCHECK is working fine.
Ensuring SLDCHECK is working fine is important to keep all systems in the landscape in sync.
SXI_MONITOR : This TCode is specific to XI or PI system. This transaction will be useful to figure
out any errors or warnings in the processing of XI or PI messages. In case of any issues, this
needs to be informed to functional team and should be troubleshooted accordingly with the
functional team inputs.
DBO1 : This transaction code is useful to findout the database locks that are present in the SAP
system.
As part of daily monitoring, SAP Basis administrator has to figure out if there are any long pending
locks more than 1 day etc and analyse reasons for the same. Sometimes if programs/jobs got
terminated abruptly without removing the database locks set, this will lead to performance issues
as other programs which needs that lock cannot set etc and they have to wait indefinitely as these
locks won't get released automatically. In case of any long pending locks, Basis administrators
should contact DBA team if any an dfigure out the reason for these locks and action accordingly
Jack Stack Checks : Please refer below links to understand Daily Java monitoring of SAP
Landscape:
Please find below table which summarizes daily monitoring tasks that are to be performed by the
SAP Basis Administrator :
</></></></></></></>
Sno Task
ABAP Stack Checks
1Check process overview(SM50)
2Check overall system process overview(SM66)
3Check application servers status(SM51)
4Check for any pending locks (SM12)
5Check for Dumps in the system(ST22)
6Check System log for any errors(SM21)
7Check for any hanged updates or update status(SM13)
8Check for excessive swapping (ST02)
9Check for critical job status like backup,updatestats,checkdb etc(DB13)
10Check for longrunning/failed jobs status(SM37)
11Check database alertlogs and performance(ST04)
12Check spool job status (SP01)
13Check cache status (sxi_cache) for PI System
14Check SLD functionality(SLDCHECK)
15Check SXI_MONITOR for PI system
16Check for Database locks(DB01)
Java Stack Checks
1Check java portal accessibility using link
2Check server0 log for java system for critical errors
3Check accessibility of management console
4Check server node status
5Check default trace for critical java errors
6Check java reports for memoryconsumption/swapping
Os level checks
1Check filesystems usage (shouldb be <80%)
2Check for swap space using topas etc
3Check for work directory log files at oslevel for errors