Dispatcher Work process hangs

[Former Member]

Hello Technology Experts,

Last 1 month I am facing a problem in our Production Server:

Please find the details about the same:

Production instance application server's dialog work process hangs indefinitely.

Users can not do anything and for everyone it hangs.

Our product detail:

SAP ECC.60 EHP7 Support patch 9,

Kernel 7.42 support Patch 101.

Analysis and Actions:

1. I have checked all the work process log files and could not find any reason for the same.

2. Previously, we have seen that all the work processes were waiting for the CRM Server.

Since we are not using CRM server, we have removed this CRM connection in SM59 and SMLG.

3. Still we are facing the same problem of Dialog work processes are getting hung Frequently (2 to 3 times per week).

4. Once we restart the complete SAP instances, then evrything works again for couple of days and again the same situation.

5.  I have sent all the Log files, Core dump, SM51/SM66, dpmon screenshots  to SAP, but they said , they could not find any reason for Dialog work process hang situation.

6. But they have recommended to go with Enq Split as per SAP note -2119669

Question:

1. Can anyone please let me know whether you have faced similar situation before?

2. And if in case dialog work process waits for some server or process, why all the dialog work processes are getting hung?

3. Is there any configuration in work processes where we can say that when one of the workprocess hangs, then the other process should not wait!!

4. or set the workprocess to wait for particular time, then restart the work process automatically?

fyi.

we will also do the Enq Split this weekend , as suggested by SAP.

Please let me know in case work process trace files needs to be posted for further analysis.

Thank you.

--

Regards,

Eswaran

[isaias_freitas · Accepted solution]

Hello,

Do you have a screenshot from the transaction SM50, showing the work processes when the hang situation is occurring?

If login to SAP is not possible, have you collected the "work process table" information using the operating system level tool "dpmon"?

If you provide either of the above, we might be able to suggest something .

Regards,

Isaías

[Former Member · Accepted solution]

Hello Isaias,

Please find the attached dpmon screenshot during work process hung situation.

Please suggest your findings to solve this situation.

Thank you in advance.

Regards,

Eswaran

[isaias_freitas · Accepted solution]

Hello,

The "wp2.png" file does not show the database related fields, so it is not possible to confirm whether both files show the same issue.

The "wp1.png" file shows the processes reading from the DYNPLOAD and REPOLOAD tables.

One possible cause would be something related to slow database performance.

Perfming the ASCS split would not help in this case. However, I strongly recommend that you proceed with it, in order to avoid enqueue related issues.

Regards,

Isaías

[Former Member · Accepted solution]

What DBMS system are you using? Check DB state/perfromance metrics for the same time frame as already suggested by Isaias.

[Former Member · Accepted solution]

I found that when I restart the Gateway using SMGW transaction, then the problem with Work process hanging is gone.

But I think this is just an workaround.

may be there are some gateway profile parameters which will solve the problem of work process hang situation forever.

The below SAP Note helps me just to get the idea of the whole situation.

934109 - Work Process Status on-hold

Regards,

Eswaran

[isaias_freitas · Accepted solution]

I don't see anything related to the gateway on those screenshots...