How to recover missing documents in datasphere replication flow without having to reinitialize

[#1]

Hi 

I was wondering if it is possible to recover missing documents when replicating data from SAP ECC to SAP datasphere via SLT without having to do initial load again. I was informed that some accounting documents were missing for certain days.  Upon investigation, I found that the missing documents wasn't transferred due to error message "Transferring delta data failed due to the following error: An exception was raised / System failure." 

We are running the replication flow every minute so it will be hard to see the error if it happens and we want  to avoid doing initial load every time to fix it.

Thanks,

B

[MKreitlein · Accepted solution]

Hello @bhat_vaidya2 

Unfortunately there is no "standard" solution for this.

If your ECC is on HANA already, you could maybe create another CDS View, like a copy of the one you are extracting the data currently from .... and use this second in a "initial only" flow with data selection, if you know exactly which data is missing, or you load a whole period.

Hope the idea helps?!

BR, Martin

[bhat_vaidya2 · Accepted solution]

Thanks for your feedback,

I am extracting standard table via SLT. By the way, what is initial only ? Can I create a new replication with initial only which is more like full load when I already have a "initial and Delta" replication flow. My understanding you cannot replicate a table that already exist without stopping the replication flow.

[MKreitlein · Accepted solution]

Hello @bhat_vaidya2 , I'm sorry, I did not recognize that you are transfering the data from the table, I thought you are using CDS Views. For CDS Views you can also load either via Initial only OR via Initial & Delta... therefore you would need the 2nd CDS View ... here some information https://help.sap.com/docs/SAP_DATASPHERE/be5967d099974c69b77f4549425ca4c0/da62e1ee746448e8bc043e1be4... BR, Martin

[Martin_Kuma · Accepted solution]

Hi, you are correct you cannot use the same TARGET table, but you can use the same SOURCE table in a second Replication flow targeting another TARGET Table. You can afterwards 'merge' the two TARGET tables. However, how to you execute the flow? you schedule it every minute or you let the flow 'listen' to updates?