Originally posted by: SystemAdmin

Hello,

I am trying to implement error handling in an EDIFACT map using WTX 8.1.0.3. I’ve tried two different approaches and I’m having an issue with getting the map to report the data element that is in error or what actually caused the map to fail.

My first approach is using the REJECT function on the element with the restart attribute using a second output card to report the errors or the bad data. This is working but not every time. For example, I have a test file where the sender id is missing and I get an error “Input type contains errors”, which I have set up in the map settings to fail. When I look at the error report, there is nothing there to state that what element caused the map to fail or where it is. I have other files where it works perfect but it is not consistent so I’m hesitant to use it.

My second approach is using the audit log. When I run the same file, the map status is error which is ok, the status states “E00”> Partner Inbound Interchange</object> however the sender id element is not reported on the audit log or the actual data. My audit settings are OnError for summary audit, burst audit is always everything else is never. I have dragged this field (and many others) to the data audit settings set it to Error, Error , Error and occurrence, Error, Error but it still doesn’t report the data element. It seems that the only status I am getting is an ‘E00” with input type contains errors, which isn’t very helpful. I’ve tried adding component rules so that I can get a more user friendly error message, but nothing works. I even tried to customize the myaudit.mtt type tree which is actually set up as X12 so I’m trying to customize it for EDIFACT. I’ve searched other post for error handling tried some of the suggestions but nothing seems to work.

I’m hoping someone with a little more experience with error handling can give me some suggestions that I didn’t think of to try on reporting the actual data/location that caused the map to fail. Basically I need to be able to report exact details of what caused the map to fail (for example, Grp2, TDT segment missing…..). Any ideas of what to try next would be appreciated. Thanks.
#IBM-Websphere-Transformation-Extender
#IBMSterlingTransformationExtender
#DataExchange

Originally posted by: repanzer

I've always had problems with using the data audit setting in the organizer not being consistent, just like you mention, where it works good for some objects, but not for others.

In that regards, I can only say to keep changing the settings until you get the desired result.

I'm assuming you want the error handling to show the specific element that caused the error. In my shop, we do it manually using the trace file.

Also, I don't believe the restart should or would work if the object does not loop (though I'm not claiming that I know for sure). At least that sounds right, because if it didn't, how could it restart at that object?

Maybe rather then grabbing a field and putting that into the data audit settings, you grab the segment the field is a part of instead.

Also, you can use the IN function in the data audit settings, something I still had issues getting the desired result when using.

For example, you grab the BGM segment and it looks like;

BGM Segment:Message PAYMUL Outbound Partner Set VD96A:PAYMUL

you can alter it to look like;

Segment IN Message PAYMUL Outbound Partner Set VD96A:PAYMUL

Maybe try changing it like that to get the result you want.

Let us know how it goes.
#IBM-Websphere-Transformation-Extender
#IBMSterlingTransformationExtender
#DataExchange

Originally posted by: SystemAdmin

Repanzer,
thank you these are great suggestions. Yes, I believe the restart is on a component that loops, so you are correct. I will try what your suggestions and let you know. Thanks again!
#IBM-Websphere-Transformation-Extender
#IBMSterlingTransformationExtender
#DataExchange

Originally posted by: SystemAdmin

hi newbag,
recently I did tried with audit log approach to track the exact point of error. What i did was in the organizer--> dataaudit settings

drag and dropped the highest level restart object (you can have restart attribute to inner groups also). and trac error, details and item data or occrence.

It gave me the exact error object. I captured the error records using the reject function.

I used the default audit type tree to read the audit log. this is second part any way.

if you can zip it your sample map and test data, i can try for you.

thanks
nagesh
#DataExchange
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender

Originally posted by: SystemAdmin

Nagesh,

thank you for offering your help. So far I have been able to get the error down to the group # but no more details than that. Because of the nature of the map I cannot send you the exact map, but what I can do is send you a generic edifact type tree and samples. I have been playing around with this map also as a validation map to see if I can get more details. Currently I have it mapped to in=out, so this is the only map rule. I will need to modify this map in order to use the reject function so when I have it ready I will post it. thank you and everyone for their suggestions.
#DataExchange
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender

Originally posted by: SystemAdmin

Nagesh,
I have attached the generic map, type tree and test files. the way the data audit log is currently set, is what I am getting the best results from. I also tried your suggestion of adding the highest level restart object. When I add a lot of segments the errors (if found) are duplicate making the log hard to understand. On two of the files I am able to narrow it down to the Group # but not the segment or data element (not sure if thats even possible). The rest of the files I am getting the E00 error. Whats strange is on two of the files I get the E00 error but the output looks valid. I would appreciate any advice on how to get a more accurate reason for rejection. Thank you.
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender
#DataExchange

Originally posted by: janhess

There are 2 problems with VESDEP_Unkonwnerror.txt.
It is treating the first * as a terminator.
If that is corrected the MsgVersion# is missing from the UNH envelope.
I think it doesn't report the first error in the log as it thinks it's read past the end of file.
#IBM-Websphere-Transformation-Extender
#DataExchange
#IBMSterlingTransformationExtender

Originally posted by: SystemAdmin

Hi Janhess,

You are correct on both issues. In trying to create test files, I accidentally typed over the version #. What I was trying to attempt, was to create a testfile with an invalid or missing terminator to see if the audit log would report a status code of E05. I guess this would explain why I am getting a E00 for this file. I guess I am a little confused as to why it would interpret the * as a terminator when its not in the UNA as a terminator. The good news is when I fixed the version #, dragged the UNA envelope to the data audit settings, I was able to get the E05 status on the audit log! I also recd other status codes E07, E09 as well. So now I am attempting to add other items to the data audit setting, run the test files thru again, and hopefully I wont get any duplicates. Thanks!
#IBMSterlingTransformationExtender
#DataExchange
#IBM-Websphere-Transformation-Extender

Originally posted by: SystemAdmin

hi newbeg,
I am sorry i late in reply. but the news is bad. for Standard (EDI/EDI Fact you cnat use)
Mercator can not travers up to the top if there are multi level or groups present (Standard EDI).

It can only travers upto the top in sigle object nested tree.

i.e
it works
file
level_1_rows(grp)1-s
f1
f2
level_2_rows(grp)1-s
f1
f2

it wont work
1) data doesn't loop
2) objects doesn't loop
3) no restart
4)
file
level_1_1_rows(grp)1-s
f1
f2
level_1_2_1_rows(grp)1-s
f1
f2
level_1_2_2_rows(grp)1-s
f1
f2
level_1_2_rows(grp) 1-s
f1
f2
#IBM-Websphere-Transformation-Extender
#IBMSterlingTransformationExtender
#DataExchange

Originally posted by: SystemAdmin

Hi Nagesh,

thanks for the information and thank you for trying.
#IBMSterlingTransformationExtender
#DataExchange
#IBM-Websphere-Transformation-Extender

Originally posted by: SystemAdmin

when I opened the zip file i am getting bad zip file error.

send the zip file to nageshbabu_a@yahoo.co.uk
#IBMSterlingTransformationExtender
#DataExchange
#IBM-Websphere-Transformation-Extender

Originally posted by: janhess

Nothing wrong with the zip file. Must be your download.
#IBMSterlingTransformationExtender
#DataExchange
#IBM-Websphere-Transformation-Extender