(RDMLX) No CIK value - randomly happening to end users

[jimoreilly]

I am wondering if I am doing something wrong when I put new components into production.

Sometimes only one or two components may change and the schema remains the same. I have been having end user problems where after the new pieces are in place, some of the users applications fail with the following error.

From user . . . . . . . . . : WEBUSR11
From module . . . . . . . . : DCXP8602
From procedure . . . . . . : DCXP8602AH
Statement . . . . . . . . . : 4
Message . . . . : Fatal Error : Component : LRNG_WAM Main LongRange WAM
Statement : 78 Message : No CIK value or property was supplied by the client
application. Routine : *COMP.

Not everyone gets this error, but those that do, then have trouble getting the app to work again. Sometimes they have to restart the phone, sometimes they have to reload LR onto the phone. Sometimes it will just start to work again.

Do I need to force the users to reload the schema everytime I change a component of the app, even though the schema did not change?

[MarkDuignan]

This error message may just be a side effect of the real problem.
ie: The real problem is listed earlier in the job log.

Can you please get a full IBM i job log - and post it to the forum or to LANSA support?

There are also two other standard log / error files to look for when a LANSA for the Web interaction fails.

I will ask LANSA support to post the full set of details you should look for when any LANSA for the Web interaction fails.

[LongusRanger]

This table of standard log and trace files to look for (and their standard location) might help when attempting to resolve L4Web related issues - includes VLF-WEB, VLF.NET and LongRange-LANSA ………
File Platform Typical Location
QPJOBLOG spool files IBM i Various IBM i output queues – sometimes on QEZJOBLOG
lx_wam.log IBM i log directory
x_err.log All IBM i – log directory
Windows – temp directory
LANSAWEB.log Windows Temp directory
X_Tracennn.txt All temp directory

IBM i log directory is usually \LANSA_<pgmlib>\x_lansa\log on IFS
IBM i temp directory is usually \LANSA_<pgmlib>\x_lansa\tmp on IFS
Windows temp directory is usually C:\Windows\Temp. Use %TEMP% to find it.

If you can't find the problem, these logs and trace files should help.

[jimoreilly]

The entry in the job log is always preceded by just the change in userid. The web task is reused many times by other transaction prior to the failure. The failures are always just the change of userid and then the CIK message.

I can have the app fail on one idevice, switch to another one, do exactly the same transactions and it will work on that one.
Often I have processed many transactions including in the changed component that all work prior to the one that fails.



I cannot view the log files you are referencing in notepad, they open but not readable.

DCM1710 Completion 00 01/29/14 18:00:17.717824 DCXS86XX LN@COMLIB *STMT W3@P1200 LN@PGMLIB *STMT
From user . . . . . . . . . : WEBDEV11
From module . . . . . . . . : DCXP8602
From procedure . . . . . . : DCXP8602AH
Statement . . . . . . . . . : 4
To module . . . . . . . . . : W3XP1200
To procedure . . . . . . . : ChangeUserProfile__FPCcT1
Statement . . . . . . . . . : 15
Message . . . . : Current user profile changed to WEBDEV11.
Cause . . . . . : User profile for this job changed to WEBDEV11. Job
attributes changed: Current library, user portion of the library list,
output queue, printer device, job accounting code, CCSID, country, language
CPF9898 Diagnostic 40 01/29/14 18:00:17.795058 DCXS86XX LN@COMLIB *STMT *EXT
From user . . . . . . . . . : WEBDEV11
From module . . . . . . . . : DCXP8602
From procedure . . . . . . : DCXP8602AH
Statement . . . . . . . . . : 4
Message . . . . : Fatal Error : Component : LRNG_WAM Main LongRange WAM
Statement : 78 Message : No CIK value or property was supplied by the client
application. Routine : *COMP.
Cause . . . . . : This message is used by application programs as a general
escape message.

[jimoreilly]

I also tried debuging the component. For all requests except the one that fails, I can step through the code. On the failure, I never get into the debug part of the app. The component fails prior to: Mthroutine Name(HandleRequest) Options(*REDEFINE)

[MarkDuignan]

This look like it should be handled by the support team.
Could you get all the logs (please include the QPJOBLOG spool file) material together and pass this over to your regional support team?

[MarkDuignan]

If you can reproduce this problem what would also be useful to help LANSA support to resolve this is a server side system trace. It should show whether a CIK (licensing) value has been sent by the client.

Server side trace files go into the lansa_xxxxxxxxxx\webserver\images\LongRange\tracing folder on your IFS where xxxxxx is your LANSA system library name (Note: I am assuming your are using an IBM i server because your trace files are in EBCDIC format).

The procedure would go something like this:
• On the iOS/Android device use the ‘spanner’ icon on the menu to display the Settings panel.
• Select the MISCELLANEOUS SETTINGS -> Troubleshooting and Support.
• Turn on the SERVER-SIDE LOGGING -> System-Level option.
• Run the failing transaction.
• Turn off the SERVER-SIDE LOGGING -> System-Level option.

Get the trace file from the tracing folder on the IFS.

If there is no trace it is probably an IFS authority issue with writing a file to the tracing folder.

[jimoreilly]

attached is a sample program that fails with CIK error.

add program MJEX0001 to any schema as online form, pass START as parameter.

Program shows five buttons on row 1 - B1 B2 etc.

selecting B1 adds a table of 100 rows in column 1, B2 adds a table of 200 rows in column 2 etc.

Program fails for me on B4.

[Mark_Dale]

So far we have not been able to reproduce the problem, using this example.

I can step through all buttons 1 to 5 and then press Click or touch a table row without any crash or joblog being produced.

We will try a wider range of devices.

[jimoreilly]

Fails on my system every time, as soon as I touch B4.

Perhaps host levels are different:

My Lansa level looks like this:

V12 SP1 build 4042

EPC858 EPC859 EPC861 EPC862 EPC866 EPC868


IBM is V7R1 using Apache

System level trace on B1 B2 B3 generates a 5MB trace file. The B4 touch does not add to the trace file.

Using Client side trace, I can view the trace log on the iphone, but cannot email it, looks to be too big, never completes building the email.