Set of import set rows getting skipped
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎01-09-2018 01:43 AM
HI all,
We are running through a strange issue post Jakarta upgrade with a range of import set rows in one of our transforms(between 4000 and 6700) getting skipped.
We are transforming CMDB computers through Eracent database and coalesces on sys_id.
There are a total of 7507 reacords of which 2800 records are getting skipped with error message : Unable to resolve target record, coalesce values not present: [Script].
I tried with creating a different data source and added the skipped import set rows as an attachment using the same import set table and transform maps and guess what it got procssed successfully.
Has anyone experienced the same issue after Jakarta upgrade?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎01-09-2018 04:55 AM
Hi Ravi,
Just giving an example screen shot. please check on your database transform map if the choice action is create or not. Or if you have any transform map script running ON Before which will ignore the record if the sys_id does not match. Please find the below screenshot as reference.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎01-10-2018 01:23 AM
HI Preetam,
There is no onbefore transform map script. I have following code in explicit transform map mapping to prevent insert of new computer records.
if (action == "insert"){
error = true ;
error_message = "New record in SN is restricted." + source.u_name ;
}
Choice action for all the field maps are "create" except for one reference field which i have set to reject in order to reject the row if there is new incoming value for the record.
Setting it to "Reject" would not create a new record in reference table and will set the row to error.
I have set error = true and not used ignore or skipped anywhere but i don't know why the rows are getting skipped.
Strange part is when i reprocess the skipped records by exporting the records in .xls format and attaching it to a data source of type attachment using the same import set table and transform maps, rows gets processed.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎01-11-2018 03:35 AM
Hi Ravi,
The choice action have been set to "reject" is it also the coalesce field. let me know if this is the case. If so then please check after you run the transform map first time. If the record is created on the target table. Let me know the result.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎01-11-2018 04:10 AM
HI Preetam,
No, this is not a coalesce field. We are coalescing on the sys_id and mapping is done through source script.
I was able to workaround by adding a log statement in the source script. Added a log in the while statement under the fucntion "buildByTransformMap" .If you want to have a look at the source script, here is your reference :
Colasec field in SCCM Integration, Colasec field in SCCM Integration
I don't know how addition of a log statement has affected the transformations which is completely strange. Even I added an obsolete variable definition var x = 5 ; under the while and this worked as well. Seems like an issue of concurrent script execution.
Commenting out the log statement caused the issue to regenerate.
We have an open HI ticket for the issue. I will update here if I get a response from them.
Thanks for looking into it.
Regards,
Ravi