Question

mWorkOrder 3.9  lags writing back to IEE after Oracle 19c upgrade

  • 4 November 2022
  • 4 replies
  • 115 views
CUCSOLUTIONS
Rank
Userlevel 3
Badge +7

mWorkOrder 3.9  lags push data back to IEE after Oracle 19c upgrade from Oracle 12c.

We are experiencing more lags in data writing back to IEE from mWorkOrder 3.9.

We are using APPS9 UD17 and prior to the Oracle upgrade didn’t have this consistent lag.

 

Our biggest obstacle is when a survey answer includes a photo and the photo often gets stuck and we end up with errors or no pictures.

 

During the Oracle upgrade permissions were changed remvoing sys and system access from appowner. I was under the impression the mobile app uses IFSSYS for all mobile transactions, however maybe this isn’t the case.

 

We were planning on deploying mworkorder 3.14, however am afraid I may introduce yet other issues moving to the newer mworkorder version prior to resolving the issues at hand.

 

Where can I begin looking to determine where this lag may be occurring?

Is there a set of permission standards deployed with APPS9 noting which IFS, SYS, SYSTEM, APPOWNER accounts should have which access

Any particular log files I can review? 

I appreciate any insight to others experiencing issues such as this and how they were resolved.

 

Thank you in advance.

 

 

 

 

 

4 replies

kathlk
Rank
Userlevel 6
Badge +14

Hi,

Yes Mobile Work order use IFSSYS to make any database connection and run as the end user. Does this issue persist for all users? I think that’s one of test you can perform. And only the lag noticed in outgoing transactions/ have you seen the same from server where Sync tasks execution time or Device Initialization time. Di you try to restart TAS server after the upgrade? 

As this is for the outgoing transactions, first log file you need to check is TAS log. Also please check the sync trace level for your user. if trace is enabled then there is a small lag but it’s normally not noticeable. 

Regards

Kapila

Regards KATHUK
CUCSOLUTIONS
Rank
Userlevel 3
Badge +7

Kathlk;

Thank you for your response.

The issues doesn’t persist for all users, about 10%.

After the Oracle Upgrade, yes we restarted the TAS server. 

We actually restart all of the VMs monthly after each Windows Update is applied.

Our TAS has its own small db on a separate SQL server.

FYI - I have set our Push queue to 10 processes since we have Pooled and Assigned orders. This seemed to help with order processing for MWO functions.

I have checked the TAS trace log and sync log for each user and the only errors I am seeing are connections lost and the following.

UTL_HTTP_FAULT: pushData: Unable to get database connection. (weblogic.common.resourcepool.ResourceDeadException: 0:weblogic.common.ResourceException: Could not create pool connection for datasource 'fndbas'. The DBMS driver exception was: IO Error: Socket read timed out)

Sending push entity data failed: Unable to get database connection. (weblogic.common.resourcepool.ResourceDeadException: 0:weblogic.common.ResourceException: Could not create pool connection for datasource 'fndbas'. The DBMS driver exception was: IO Error: Socket read timed out) - weblogic.common.resourcepool.ResourceDeadException: 0:weblogic.common.ResourceException: Could not create pool connection for datasource 'fndbas'. The DBMS driver exception was: IO Error: Socket read timed out

 

and this once occurs about 2-3 times a week.

Sending push entity data failed: ORA-00060: deadlock detected while waiting for resource - Failed executing statement (ORA-00060: deadlock detected while waiting for resource

ORA-06512: at line 1)

 

I appreciate any insight.

Thank you

Marjie

kathlk
Rank
Userlevel 6
Badge +14

Hi Marjie,

 

The error you get ‘UTL_HTTP_FAULT: pushData: Unable to get database connection’ might caused by the change you did to increase no of processors in the Batch Queue (Push Data). As soon as you increase the processors count in the Batch Queue, it start to cerate parallel processers (Of cause that’s what you need) and start to call MWS in the same scale which means MWS now want to get more Oracle connections from the database. So if you have not increased the MAX DB connections (Default in App10 is 50 I think), the MWS / WebLogic throws that error ‘Unable to get Database connections’ . So if you are increasing the default process to larger value, then make sure to do the same in MWS where increase the Max Database connection in the Pool. Also make sure your Oracle Database can cope with the demand. 

 

Hope this help for you to configure the MWS and database properly.

 

Regards

Kapila

 

Regards KATHUK
dulwlk
Rank
Badge +3

Hi 
Did you find any solution for the ORA-00060: deadlock detected issue ?

Reply


Terms & ConditionsCookie settings