[rma_repair] Exception raised when osv_memory.autovacuum run

[andhit-r]

Hi Guys,

On v8.0 i encountered problem when osv_memory.autovacuum run. Here are the logs:

2017-10-24 02:21:36,138 3724 ERROR mrp1 openerp.addons.base.ir.ir_cron: Call of self.pool.get('osv_memory.autovacuum').power_on(cr, uid, *()) failed in Job 1
Traceback (most recent call last):
File "/home/andhit_r/odoo8/odoo/openerp/addons/base/ir/ir_cron.py", line 138, in _callback
getattr(model, method_name)(cr, uid, *args)
File "/home/andhit_r/odoo8/odoo/openerp/api.py", line 268, in wrapper
return old_api(self, *args, **kwargs)
File "/home/andhit_r/odoo8/odoo/openerp/addons/base/ir/osv_memory_autovacuum.py", line 31, in power_on
model._transient_vacuum(cr, uid, force=True)
File "/home/andhit_r/odoo8/odoo/openerp/api.py", line 268, in wrapper
return old_api(self, *args, **kwargs)
File "/home/andhit_r/odoo8/odoo/openerp/models.py", line 5108, in _transient_vacuum
self._transient_clean_rows_older_than(cr, self._transient_max_hours * 60 * 60)
File "/home/andhit_r/odoo8/odoo/openerp/api.py", line 268, in wrapper
return old_api(self, *args, **kwargs)
File "/home/andhit_r/odoo8/odoo/openerp/models.py", line 5072, in _transient_clean_rows_older_than
self.unlink(cr, SUPERUSER_ID, ids)
File "/home/andhit_r/odoo8/odoo/openerp/api.py", line 268, in wrapper
return old_api(self, *args, **kwargs)
File "/home/andhit_r/odoo8/odoo/openerp/models.py", line 3631, in unlink
'where id IN %s', (sub_ids,))
File "/home/andhit_r/odoo8/odoo/openerp/sql_db.py", line 158, in wrapper
return f(self, *args, **kwargs)
File "/home/andhit_r/odoo8/odoo/openerp/sql_db.py", line 234, in execute
res = self._obj.execute(query, params)
IntegrityError: null value in column "wiz_id" violates not-null constraint
CONTEXT: SQL statement "UPDATE ONLY "public"."rma_order_line_make_repair_item" SET "wiz_id" = NULL WHERE $1 OPERATOR(pg_catalog.=) "wiz_id""

It cannot erase all the wizard data (rma.order.line.make.repair) since all the wizard items (rma.order.line.make.repair.item) need wiz_id. Do you guys having this similar problem?

Propose Solution:

1. set required=False on wiz_id field, or
2. Add ondelete="cascade" parameter on wiz_id

[LoisRForgeFlow]

Hi @andhit-r

We haven't had this issue, but anyway your proposal seems good and can prevent it. Fix done in 2598b71

Thanks!!