schedule_and_build_instances short-circuits on all instances if one build request is already deleted
Bug #1661024 reported by
Matt Riedemann
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Compute (nova) |
Fix Released
|
High
|
Matt Riedemann |
Bug Description
The 'return' statement here should be a 'continue':
That's in a block of code that's cleaning up an instance recently created if the build request was already deleted by the time conductor tried to delete the build request, i.e. the user deleted the instance before it was created (which actually deleted the build request in nova-api).
The return is wrong though since we're in a loop over build_requests, so if we hit that, and there are more build requests to process, those other instances won't get built.
It's easy to miss this context because the method is so large. We should break the build request cleanup code into a separate private helper method.
Changed in nova: | |
assignee: | nobody → Matt Riedemann (mriedem) |
tags: | removed: ocata-rc-potential |
To post a comment you must log in.
Fix proposed to branch: master /review. openstack. org/427839
Review: https:/