Project

General

Profile

Actions

Feature #10677

closed

[Workbench] Copy and re-run container request

Added by Peter Amstutz over 7 years ago. Updated over 7 years ago.

Status:
Resolved
Priority:
Normal
Assigned To:
Radhika Chippada
Category:
-
Target version:
Story points:
0.5
Release:
Release relationship:
Auto

Description

Need a button on container request page that copies the container request to a new container request in "uncommitted" state so that user can modify inputs and re-run.

To keep this simple, the implementation should just create a new container request with the relevant fields (command, container_image, cwd, description, environment, mounts, name, output_path, runtime_constraints, scheduling_parameters), in "uncommitted" state, and then redirect to the new container request page.

To ensure the new request always results in a new run, use_existing should be false.


Subtasks 1 (0 open1 closed)

Task #10747: Review branch 10677-container-request-rerunResolvedRadhika Chippada12/16/2016Actions
Actions #1

Updated by Peter Amstutz over 7 years ago

  • Description updated (diff)
Actions #2

Updated by Peter Amstutz over 7 years ago

  • Subject changed from Copy and re-run container request to [Workbench] Copy and re-run container request
Actions #3

Updated by Radhika Chippada over 7 years ago

  • Target version set to 2017-01-04 sprint
Actions #4

Updated by Radhika Chippada over 7 years ago

  • Assigned To set to Radhika Chippada
Actions #5

Updated by Radhika Chippada over 7 years ago

  • Status changed from New to In Progress
Actions #6

Updated by Peter Amstutz over 7 years ago

  • Description updated (diff)
Actions #7

Updated by Lucas Di Pentima over 7 years ago

LGTM, just found a typo on a comment on apps/workbench/app/controllers/container_requests_controller.rb:82: 'wriable' (same one as in pipeline_instance_controller)

Was going to ask why we're creating a new instance and then doing an object duplication on ApplicationController's copy method, but then I saw that this is based on the same method existing on pipeline_instance, so I imagine that there's a good reason for doing it that way.

Actions #8

Updated by Radhika Chippada over 7 years ago

Fixed the typo.

Regarding "why we're creating a new instance and then doing an object duplication": This is not exactly a pure "copy" we are doing, but copy the state with the run results. Especially in this case, we are making a copy of the object and re-running it. I guess a "copy-to-rerun" or "copy-initial-state" would have been a better name for this "copy" method instead.

Thanks.

Actions #9

Updated by Radhika Chippada over 7 years ago

  • Status changed from In Progress to Resolved
  • % Done changed from 0 to 100

Applied in changeset arvados|commit:ccc1201e103e168b04801b135b662577d7ea475b.

Actions

Also available in: Atom PDF