Bug 574930 - [RFE] cmdln/webui for job submission to permit user to set arbitrary ordered list of machine pools to try
Summary: [RFE] cmdln/webui for job submission to permit user to set arbitrary ordered ...
Keywords:
Status: CLOSED WONTFIX
Alias: None
Product: Beaker
Classification: Retired
Component: scheduler
Version: 0.5
Hardware: All
OS: Linux
low
medium
Target Milestone: ---
Assignee: beaker-dev-list
QA Contact:
URL:
Whiteboard: GroupModel
Depends On: 1127129
Blocks:
TreeView+ depends on / blocked
 
Reported: 2010-03-18 20:44 UTC by Kevin Baker
Modified: 2020-11-19 22:33 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: Enhancement
Doc Text:
Clone Of:
Environment:
Last Closed: 2020-11-19 22:30:19 UTC
Embargoed:


Attachments (Terms of Use)

Description Kevin Baker 2010-03-18 20:44:48 UTC
Description of problem:

Jeff Burke reported that the order in which the scheduler finds machines is too limited. In that it tries machines you own, in your group and then the general pool. To get around this Jeff kicks of jobs to fill the owner & group machines and then one more to get into the general pool. Then he cancels the owner & group jobs. Obviously not efficient.

Alternative:

Allow the job submitter to state the order they want their jobs to be scheduled in. In most cases the UI can suggest sensible defaults. For example, much like the UI interface for adding search terms, users could add groups to a list that sets the order they'd like the scheduler work in.

This /may/ be the default order

 My Machines
 My Group Machines
 General Pool

But permit the user to change it to anything

 General 
 Group A
 Group B
 My Group Machines
 My Machines

Comment 1 Nick Coghlan 2012-10-17 04:39:18 UTC
Bulk reassignment of issues as Bill has moved to another team.

Comment 2 Nick Coghlan 2013-04-15 08:07:56 UTC
This will be addressed as part of the System Pool design in Beaker 1.1 (see http://beaker-project.org/dev/proposals/system-pools.html#prefer-particular-system-pools-for-recipe-execution)

Comment 3 Nick Coghlan 2014-08-08 01:31:54 UTC
This is on hold until we evaluate the possibility of switching to a more capable scheduling engine.


Note You need to log in before you can comment on or make changes to this bug.