Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi,
I created a job that allows to insert nearly 24 000000 lines in sqlserver. After running I found the 24 000000 line, but bizarrely today I find only nearly 14 000 000 line.
In fact there is a loss of lines and I havn't changed anything !
Is there anyone who can help me please?
Thanks in advance.
most real reason - wrong forms (length, type, nulls and etc) from source
depending from settings, SQL output use batch size for insert, it is much faster (dramatically faster) but at same time - with any error all batch rows will be rejected
so, check -log files, check source data for compliance with Your output schema
I checked everything you just said but I suspect it is because of commit because I didn't make the component tMsSqlCommit.
Does it affect the job?
1) commit affected if You use tMSSQLConnection
2) if commit missed - You will have 0 records, but You have some of them
During my first run, i ddidin't use neither tMSSQLConnection or tMSSQLCommit.
if You do not use tMSSQLConnection, You can not use Commit, it managed by tMSSQLOutput (Advanced) and it auto commit for selected number of records, like 10 000 per batch
Yes that's what I did at the beginning, but since I lost lines I thought maybe that's because of commit 😕
Because I have no errors at the sql server log file and at tStatCatcher in Talend, so i don't understand the problem
Hello,
Is there any data truncation issue printed on console?
You can right click on tMssqlOutput, select Row, and then Reject to see if there is any record information rejected by your sql server.
Note:The reject row can't be used together with the 'die on error' option and 'use batch size' option in advanced setting tab of tMssqlOutput component, uncheck these two option and then you can use the reject row.
Best regards
Sabrina
I have already solved the problem of data truncation and i made a table for reject row.
Here is a screenshot :