chore: Handle deletion when the underlying GCP resource does not exist #3826
Conversation
| // and (false, nil) if the object was not found but should be presumed deleted. | ||
| // In an error, the state is not fully determined - a delete might be in progress. | ||
| // It returns (true, nil) if the object was deleted without error. | ||
| // In an error (false, err), the state is not fully determined - a delete might be in progress. | ||
| Delete(ctx context.Context, op *DeleteOperation) (deleted bool, err error) |
There was a problem hiding this comment.
I feel the returned deleted might be unnecessary, but I'm open to discussion:
- it's not used anywhere for now:
- the situation
return (false, nil) if the object was not found but should be presumed deleted, we could callFindto determine the existence of the GCP obj, and avoid unnecessary Delete calls if obj does not exist. Also We don't have specific handling for the case of (false, nil); instead, we treat it the same as (true, nil).
This can be called without calling Find. Unsure if there's a use case of it I haven't considered...
gemmahou
force-pushed
the
handle-delete-notexist
branch
from
080046e to
26bab12
Compare
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
| duration: 232.507687ms | ||
| status: 200 OK | ||
| code: 200 | ||
| duration: 245.142672ms |
There was a problem hiding this comment.
While idk why this changed, but I think the destroy version operation should return 200 instead of 404
Edit: This is handled in #3825, if version already got destroyed(404), we should not make the API call.
Change description
Fixes #3727
Tests you have done
make ready-prto ensure this PR is ready for review.