duplicate key on "sym_incoming_batch_pkey" when adding a table only on server locks the pull on a table

[Mario Di Ruscio]

Hi,

I have a system with a server and a client on postgres db.
I add a new table "my_table" on the server database but not in the client database.
When the client try to connect to the server
-an exception is raised for duplicate primary key in sym_incoming_batch table.
-in the client table sym_incoming_batch is generated a record with status='ER'
and sql_message='Could not find the target table my_table'
This is ok but when i add the table on the client and restart symmetric-ds but
the error still stays on and the pull of data on my_table doesn't works.
At the end i drop the record in sym_incoming_batch with status='ER' an restart symmetric-ds
and the synchronization works.

My questions are:
1.there is a way to resolve this lock operating only on server ?
2.why symmetric-ds retry the insert on the sym_incoming_batch
table and doesn't update it ?
3.What is the best way to upgrade the server database before the client database?

Thanks in advance.

[Chris Henson]

Can you post the stack trace? You should never get an dupe pk error for sym_incoming_batch.

If you want the client to ignore the fact that my_table is missing you can set dataloader.ignore.missing.tables=true