Supports rollback when allocate callback function fails #3776
Conversation
… executed, for example, in the GPU allocation scenario Signed-off-by: wangyang0616 <wangyang8126@gmail.com>
volcano-sh-bot
added
the
do-not-merge/work-in-progress
volcano-sh-bot
added
the
size/M
wangyang0616
force-pushed
the
branch_for_apollo_1.13.7
branch
from
906efb6
to
e91123f
Compare
wangyang0616
changed the title
volcano-sh-bot
removed
the
do-not-merge/work-in-progress
Signed-off-by: wangyang <wangyang8126@gmail.com>
wangyang0616
force-pushed
the
branch_for_apollo_1.13.7
branch
from
e91123f
to
566f751
Compare
| } | ||
| } | ||
| return fmt.Errorf("Failed to exec allocate callback functions") | ||
| } | ||
| } |
There was a problem hiding this comment.
How about aggregate errs and call unallocate func when there is err?
There was a problem hiding this comment.
When each pod allocation is successful, but the overall vcjob does not meet expectations, unallocate is used to roll back the previous resource scheduling information.
For direct resource errors and exceptions during the scheduling process, I understand that it would be more appropriate to roll back directly based on the error.
Signed-off-by: wangyang0616 <wangyang8126@gmail.com>
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: william-wang The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
volcano-sh-bot
added
the
approved
fix issue #3777 for task 1