Neutron db retry decorators do actually use native oslo db retries underneath, so nothing should be missed from oslo (though there were not many changes in oslo recently https://github.com/openstack/oslo.db/commits/master/oslo_db/api.py)
On top of oslo retries neutron has a plenty of checks and neutron specific handlings, which I don't think could be easily skipped without regressions.
So I'm wondering which exact problem are we trying to address here. Any examples would be good for better understanding.
Neutron db retry decorators do actually use native oslo db retries underneath, so nothing should be missed from oslo (though there were not many changes in oslo recently https:/ /github. com/openstack/ oslo.db/ commits/ master/ oslo_db/ api.py)
On top of oslo retries neutron has a plenty of checks and neutron specific handlings, which I don't think could be easily skipped without regressions.
So I'm wondering which exact problem are we trying to address here. Any examples would be good for better understanding.