Updated openstack/openstack
Project: openstack/keystone-specs 84a5147010e9a62c741437fe3c19472b2a632d14 Simplify manage-migration spec by introducing database triggers By introducing database triggers to this specification, we can vastly simplify the overall implementation and operator experience and skip straight to zero downtime. There's no need for the operator to manage any metadata in the database about the state of the migration, beyond what sqlalchemy-migrate will do for us, and there's no need to write versionedobjects for each possible object state during a live migration. Hopefully that will all translate to fewer bugs. On the downside, it means that we have to manually write triggers, which may be need to be specifically tailed for each of our supported databases (MySQL, PostgreSQL, and SQLite). Change-Id: I38c02a3a3388ed57032f01e8f522ce53230231f8
This commit is contained in:
parent
9125288862
commit
f8105f2020
|
@ -1 +1 @@
|
|||
Subproject commit 5e18978fb10f6efe510b79d3d4731580db5af102
|
||||
Subproject commit 84a5147010e9a62c741437fe3c19472b2a632d14
|
Loading…
Reference in New Issue