cancel
Showing results for 
Search instead for 
Did you mean: 

Synch error: Internal server error; contact your system administrator

Former Member
0 Kudos

Dear all,

Hopefully someone can help me with this problem.

We have a production system where more or less 800 users are using the MAM daily. Our MAM is a custom implementation, based on MAM 2.5. We are currently using MI SP19P8 at client side. Server is also on SP19.

We have a problem with 1 user, who gets the following error message after synchronization:


Internal server error; contact your system administrator.

When looking in MEREP_MON on the middleware server, we don't get any error or warning message, however I've noticed that 3 SyncBO objects are whole the time standing in "I-Waiting": ZMAMFR_001, ZMAMFR_070 and ZMAMFR_090.

I've also seen that there are a lot of objects in "O-Waiting", and only 1 object in O-Sent (which seems to be logical, as the system is still waiting for the first 3 objects, in "I-Waiting", to complete).

My conclusion is that we are dealing with some kind of time-out, when the middleware server is communicating with SAP. When I put the trace level on "all" at client side, this hypotheses seems to be confirmed. This is the error message I get:


Exception while callSync: Time limit exceeded.
Initial buffer size of BinaryReader is 66
Read current inbound counter from file: 41
Received exception info code 10 and detailed message "Exception while callSync: Time limit exceeded."
Exception while reading prolog from inbound stream
com.sap.ip.me.api.sync.SyncException: EXCEPTION_INFO_CODE=10
	at com.sap.ip.me.sync.in.BinaryInboundReader.readProlog(BinaryInboundReader.java:268)
	at com.sap.ip.me.sync.in.BinaryInboundReader.<init>(BinaryInboundReader.java:108)
	at com.sap.ip.me.sync.SyncManagerMerger.processInboundContainers(SyncManagerMerger.java:62)
	at com.sap.ip.me.sync.SyncManagerImpl.processSyncCycle(SyncManagerImpl.java:836)
	at com.sap.ip.me.sync.SyncManagerImpl.syncForUser(SyncManagerImpl.java:1278)
	at com.sap.ip.me.sync.SyncManagerImpl.processSynchronization(SyncManagerImpl.java:909)
	at com.sap.ip.me.sync.SyncManagerImpl.synchronizeWithBackend(SyncManagerImpl.java:464)
	at com.sap.ip.me.sync.SyncManagerImpl.synchronizeWithBackend(SyncManagerImpl.java:319)
	at com.sap.ip.me.api.sync.SyncManager.synchronizeWithBackend(SyncManager.java:79)
	at com.sap.ip.me.apps.jsp.Home$SyncRunnable.run(Home.java:641)
	at java.lang.Thread.run(Unknown Source)

We've also tried to reset and reinstall his MAM, without success.

Since all other users (more or less 800) don't experience any problem, I don't think it's a structural problem on the server.

So I've executed function module MAM25_001_GETLIST on our SAP production system, to see if I get any error / time-out when the orders for this user are loaded, but I could run this function module without any time-out / error. I've also done a MAM25_001_GETDETAIL for every order I got back with MAM25_001_GETLIST, also here without any time-out / error.

If there would be a data problem in SAP, I would expect these methods would result in a time-out / error.

I really have no idea anymore what can be wrong. Maybe somebody else has experienced this problem as well?

Thanks in advance for your feedback!

Best regards,

Diederik

Accepted Solutions (1)

Accepted Solutions (1)

AjithC
Employee
Employee
0 Kudos

Hi,

I think there is too much data for this user. The sync is taking more time. Set the ICF timeput parameters to a higher value.

Have a look at this note 921061 for more information on timeouts parameters..

Regards

Ajith

Former Member
0 Kudos

Hello,

Thanks for the feedback!

First of all I want to inform that we were able to solve the problem, however it is not clear what exactly caused this issue.

We solved it by removing all his orders, synchronizing, and then assigning one-by-one his orders again to the technician.

2 orders however were not assigned anymore.

I don't think it's a problem of too much data, since he didn't had a spectacular amount of service orders:28. We often have users with more.

The only 'different' thing with 1 of those 2 not-assigned orders, compared to the other ones, is that this order had 2 operations with control key ZCO3, one assigned to a normal service technician, and one assigned to a branch.

Normally the service order is send to the technicians, mentioned in the ZCO3 operations, but since a branch is not a real user, maybe this has caused the problem. I 'll try to check this in our test environment when I have more time.

If this was not the problem, I can not explain what went wrong...

Thanks for your feedback!

Best regards,

Diederik

AjithC
Employee
Employee
0 Kudos

It may not be the problem with number of orders alone, but with the number of dependent objects. With every order,we sent the Functional Location, Equipments, Notifiations & Partners and this have again cascading effect..

I remember we had a simlar case long back where in only one user was not able to sync (exactly the same error) because he was a superviser and he had too much of history data...

Just check the dependent data for the orders. If that is a test system, assign only that to a user, and see how much data comes in the merep_mon outbound data..

Regards

Ajith

Former Member
0 Kudos

OK, thanks for the info!

I'll keep it in the back of my mind, in case we have that problem again.

Have a good day!

Answers (0)