DSXchange

I am trying to create one server job. Design looks like:

Merge--->Transformer--->sequential file

when I run this job it gives me following error:
Abnormal termination of stage Trial_Merge..Sequential_File_1.IDENT1 detected.

I am using a name for sequential file which is unique and have not been used before. I also tried same job with oracle stage as target, by removing transformer and used only sequential file but still getting same problem.

Are there any settings which I should do to avoid this problem?

Any help will be much appreciated.

Your error must be from a run without a transformer, otherwise it wouldn't have mentioned IDENT1 as the active process.

Are there any other warnings in the log? If you Reset the aborted job from the Director, does a new "From previous run..." log entry appear? If so, post the contents.

I am sorry. Yes that error was from run without a transformer.

chulett wrote:Your error must be from a run without a transformer, otherwise it wouldn't have mentioned IDENT1 as the active process.

I again ran the same process with transformer and got the same error at transformer:

Abnormal termination of stage Trial_Merge..Transformer_1 detected
I noticed that in director it shows as warning not as error.

When I reset job from director it shows status as 'has been reset'. There are no additional error/warning message in logs.

Are there any other warnings in the log? If you Reset the aborted job from the Director, does a new "From previous run..." log entry appear? If so, post the contents.

Have you named two sequential files in the Merge stage?

Thanks Ray for pointing me out. Actually in Directory path itself I gave file names and did not provide name in actual tab. Now I provided file name and directory path properly.
When I run again I am getting error:

"Abnormal termination of stage Trial_Merge..Transformer_1 detected"

One more thing - In input file tabs I am not able to view data. I mean that option is grayed out. Am I missing something else?

I have not provided anything in temporary directory name. I tried to look in help for what should be provided there but help also does not mention anything.

Have you loaded format and column metadata onto both input links?

Yes. I have. Still I am not able to view data

ray.wurlod wrote:Have you loaded format and column metadata onto both input links? ...

If I provide everything correctly then would I be able to view data?

No, there's no View Data capability in that stage which is why it is disabled. Ray was asking in an attempt to stop the aborts, not make it so you can view data.

Ok. Thanks for reply.

Yes, I have provided formats and metadata and in mapping tab I have provided all keys which should be mapped to each other and output columns as well.

Good. Now please answer the question about whether any "from previous run" information appears in the job log after you reset (not re-compile) an aborted job.

I ran the job. It aborted with warning:
"Abnormal termination of stage Trial_Merge..Transformer_1 detected"

After that I reset this job from director. Its status is "has been reset"
In logs there are no error/warning message. It has following info:
"Finished Resetting stage Trial_Merge..Transformer_1."

The "from previous run" message (if any) has Informational severity. Please check again.

I am sorry. I was looking into Type for Previous run. I found it in Info:

"From previous run
DataStage Job 233 Phantom 17828
Abnormal termination of DataStage.
Fault type is 11. Layer type is BASIC run machine.
Fault occurred in BASIC program DSP.Open at address 574."

I hope that is what you are talking about.

Nobody said it would be a warning or an error.

Next step? An exact search here for "Fault type is 11" since you're not the first to experience that particular error with the Merge stage.

Thanks Chulett!

I read all posts match to Fault type 11 criteria and tried several things which I thought, match my problem and nothing really helped. And after that I found that initial issue was temporary directory name. If I don't provide temporary directory name, process don't allow me to load even single row.
After I provided temp dir name record count is a contraint. I am trying to merge 2 seq files which have 25 million and 10 million records. I am trying to perform left join.

1. I ran process with 1 row.--->Success
2. 100 rows---> Success
3. 10000 rows---> Success
4. 1000000 rows--> Success
5. 5 million----> Success

But when I run for entire record set it starts failing.

Any suggestions regarding this?