DSXchange

Hi,

I am quite new to Datastage parallel jobs and I need you help for resolving the below issue.

Iam using SCD stage for implementing Type2 SCD logic. I have build the job that works fine for the first run but when I run for second time with some changes in the data the job was aborted and showing the fatal error as

Error: Invalid Julian Day.

I have taken 1 business key 1 Type 2 column and 1 column indicator.

As per my logic

1) Any changes in the business key will insert a new row with flag as 'Y' that is working fine
2) Any change in the Type 2 column needs to update the existing flag as 'N' and insert a new row with flag as 'Y'.

This second change is not happening it aborts the job by showing the above error i.e Invalid Julian Day.

I can see no null value in my Type 2 column even then it is aborted.

Kindly help me with your suggestions, any help is appricaited.

Thanks,
Karunakar

Somewhere in the job you are converting from a string to the database date column and the format is not correct, probably in the date derivation for the update link.

ArndW wrote:Somewhere in the job you are converting from a string to the database date column and the format is not correct, probably in the date derivation for the update link. ...

Hi,

Thnaks for your reply.

I have the date columns in my table but I am just populating CurrentDate function in to that. I am not doing any conversions for that Date columns.

Even then I am getting the Inavlid Julian Day Fatal Error.

Please respond this.

Thanks,
Karunakar

Are the columns nullable? If so, remove your population with CurrentDate and see if the error remains.

ArndW wrote:Are the columns nullable? If so, remove your population with CurrentDate and see if the error remains. ...

Hi,

Hope you are using Change Capture Stage for this. In some cases DS PX log will not capture the exact error. So what i advise you is, before you run the job check the following.

1. Whether you can read the data (View data) from both ends.
2. Check the Data type for all columns for inputs.
3. Make sure you used proper partitioning.

If you are happy with the above, then run the job.