DIALOG PROCESS

Dialog Process:

• It is the only process that is used to communicate with end users Interactively using Classic GUI or Through RFC or through Http Protocol.
• The Average Response Time for each dialog step is 600 ms.it is displayed through ST03n
• Each Process can execute up to a maximum runtime of 600 seconds.if any process goes beyond this time an additional time of 60 seconds is given and the process is timed out.(ST22)(it is not mandatory to allocate additional 60 seconds
• Each Process Serves 5-10 users.(Calculated based on ST07)
• Each Process Requires 75mb to 300 mb Memory
• Each Process can Handle the requests that are initiated for background, spool & update and update them in temporary tables.
• Dialog Process is not restricted to a User and Users are not restricted to a process which is referred as process multiplexing
• Parameters to configure dialog processes is rdisp/wp_no_dia
• There should be at least minimum two dialog processes and maximum up to 100 processes.
• The Maximum runtime of a dialog process is restricted by using parameter rdisp/max_wprun_time
• Dialog Processes are monitored in SM50/SM66/dpmon/sapmmc /sapmc
• The Status of dialog processes are waiting, running, stopped, on hold and ended
• The sum of dialog processes should always be more than the sum of non-dialog processes in an ideal instance
• The Process can be terminated by using SM50/DPMON whenever it is in RFC/CPIC/PRIV mode causing bottleneck to the system

SM51(Monitoring):
Monitor the Number of Application Servers
precisely count the apps and identify the app servers which are missing.
check whether if there is a notification for downtime(OS Upgrade,DB Client Update,hardware Update,parameter Changes,sap kernel update etc.) Check-in Sm02 for Message or check Email.

• Check with L2, L3, L4 Teams and escalate the issue immediately.
• If there are no issues and if authorizations are available at OS Level then start the Application Server at OS Level using command ‘startsap r3’
• SM51 Provides the release levels of the system(OS Version, Database Version, and SAP kernel Version
• It displays the Number of Application Servers with provided services such as DVEBMGS..D01, D03…. Including the Type of Processes Such as Dia, Btc, Upd, Upd2, Spool, ICM and Enq.Message Server and Gateway are not displayed in SM51. From Nw7.3 on-wards Enqueue Server also not displayed.
• It provides the Navigation to other instances without login using RFC mode.Execute Sm51–>Double Click the instance–>it displays SM50->navigate to the Transactions of that Instances.
• Double Click on the instance to navigate to the work process overview of the instance.
• It is possible to navigate to 57(62 in EHP7) transactions approx from SM51(Table TCDCOUPLES displays the list using transaction SE12/SE16). All the Transactions may not be allowed to a monitoring consultant.Check with SAP Security Team if any of the transactions are not allowed.
• It is possible to display the inactive instances from RZ03.Earlier versions until Net Weaver 7.0 it is possible to start and stop the instances from Transaction Rz03. It is deactivated due to security reasons.
• Cleanipc <instance number> will be used to release the memory occupied by the instance.(Login as <SID>ADM) AND Execute Command >cleanipc 00.
  This will release the memory and facilitate the instance to start with initial memory.
• Check the available.log in work directory for the exact downtime.
  check /usr/sap/SID/DVEBMGS00/work through transaction AL11.
• The Instance Failure logs are displayed in the Log dev_disp in the working directory.
• Check the Instance status using dpmon followed by profile
  dpmon pf=profile name.
• Check Whether the Server is Reachable using command ‘ping’.
  if the server is not available then refer to system admin team immediately.

Work Processes Overview.
These are Monitored by using various tools.

• Dpmon (Dispatcher Monitor) It is an OS Utility provided by SAP to monitor the
• Dispatcher and Work Processes. It is widely used on UNIX Platforms.
• They are Monitored at OS Level Using Command
  ‘ps -ef |grep dw*'(UNIX)
• They are Monitored at OS Level Using ‘Task Manager'(Windows).
• They are Monitored by Using SAPMMC(Microsoft Management Console), SAPMC(Management Console).

SM50 Process Monitoring:
It is used to display the list of work processes configured on the Instance.
It is possible to display Dialog, Update, background, Spool, and Enqueue Processes Precisely it displays which process is performing what task and which user is working on that process,how long(time), which table, program etc.
It displays the work Process trace from the startup of the Instance.it is also possible to cancel(kill) the Process with Core(trace)/without core(no trace).

SM50 Displays the Following:
Serial Number of the Process:(it is a whole Number start from zero)

1. Type: Work Process Type(DIA,BTC,UPD,UPD2,SPO,ENQ)
   Gateway and Message Server are not displayed in SM50 and they are maintained through separate Transaction because they are not internal to process the user Task.(SMMS for Message Server and SMGW for Gateway Monitoring)
2. PID: Process ID at OS Level(Each Process is allocated with a Process ID)
3. Status of the Process: Running, on hold, Stopped, Waiting, Ended,
4. Status Reason: if the Process is held(running,on hold) for a longer time then check this column why the process is on hold. it could be due to PRIV, CPIC, RFC(Remote Function Call)etc Modes.(CPIC is an SAP Protocol used to communicate between SAP Systems. it stands for Common Programming Interface for Communication).
   (Stopped Status can also be observed along with On hold in SM66)
5. Restart: if the Process is terminated/aborted/timed out then restart the process YES/NO….No for Debugging the Process Failures.Yes for Restart after the failure.(AUTO Restart by dispatcher)
6. Error: The Number of times process has been restarted(the Number is marked in Red color)
7. Semaphore: The Number of the Semaphore where the work process is blocked at Disk Level.Green Semaphore(Processing that Semaphore and processing the Request)Red Semaphore(Waiting for the Semaphore) It means each process while processing the request it may communicate with enq,upd, buffers, print,pxa, number range buffer etc.
8. CPU time: The Cumulative CPU time.
9. Time: The Time Consumed by Process referred as runtime.
10. User Name: Name of the User
11. Program: The Name of the Program/Report
12. Client: The Logged on Client Number
13. Action: The Type of Action(read,logical read,sequetialread,insert,delete, roll-in/out etc)
14. Table: Name of the database Table as part of the Monitoring identify the expensive processes/Users along with their runtime and update the check list.
    Based on Approvals/Situation/Requirement the work process can be killed/terminated using the SM50/Dpmon tool. Dpmon is an emergency tool to monitor the processes at OS Level.it is the only tool used when the GUI Logins are congested/blocked.
15. Maximum Number of Work Processes:
    Parameter rdisp/wp_max_no is used to set the limit for all the configurable work processes on the system.Ensure that there is sufficient memory for the processes else the system may be choked.from Net Weaver 7.3 this value is set to a maximum of 600.
16. Reserved (restricted) work processes:
    apart from work process types (dialog, background, update, spool, there is also the type reserved. The number of initial work processes (as with the other types also) is specified in the profile parameter rdisp/wp_no_restricted. No further reserved WPs can be created while the system is running.These are created on demand when there is a system blockade.Example during deadlocks an additional process is created to resolve the issue.
    Reserved Work Processes are only of type Dialog. 36+3(rdisp/wp_no_restricted)