commit ac004fdf9cfa9c91532ed53cdb80b65fcf1f1636
Author: Andy McCrae <email address hidden>
Date: Wed Mar 1 16:17:07 2017 +0000
Adjust ordering for nova_db_setup
Running an api_db sync on an existing cluster without cells, for example
in an N->O upgrade requires the cells to have been setup.
Changing the ordering to setup the cells first before running the api_db
sync will resolve this.
However, in a greenfield deploy we require the api_db sync to happen
before we can create a cell. We should test to see if the api_db version
is 0 (it hasn't been setup yet) and then run the api_db sync if it
hasn't, otherwise skip to create the cell.
Reviewed: https:/ /review. openstack. org/440679 /git.openstack. org/cgit/ openstack/ openstack- ansible- os_nova/ commit/ ?id=ac004fdf9cf a9c91532ed53cdb 80b65fcf1f1636
Committed: https:/
Submitter: Jenkins
Branch: stable/ocata
commit ac004fdf9cfa9c9 1532ed53cdb80b6 5fcf1f1636
Author: Andy McCrae <email address hidden>
Date: Wed Mar 1 16:17:07 2017 +0000
Adjust ordering for nova_db_setup
Running an api_db sync on an existing cluster without cells, for example
in an N->O upgrade requires the cells to have been setup.
Changing the ordering to setup the cells first before running the api_db
sync will resolve this.
However, in a greenfield deploy we require the api_db sync to happen
before we can create a cell. We should test to see if the api_db version
is 0 (it hasn't been setup yet) and then run the api_db sync if it
hasn't, otherwise skip to create the cell.
Closes-Bug: #1667127 4ef44cad0a1626b 68229bca060
Change-Id: If3fff46dd58129