Re: [Queue-developers] Feedback requested on detailed plans and code for contrib project

[Patrick J. H. <sp...@si...>]

On Fri, 15 Sep 2000, Monica Lau wrote:

> Someone also brought up the idea of a back-up queue_manager.  There
> would be a master queue_manager and a slave queue_manager.  The slave
> queue_manager has all the information that the master queue_manager has,
> so if the slave ever detects that the master has failed, it will take
> over the master's role until the master starts back up again.  Is this
> similar to what you have in mind?

I would suggest setting up an SQL database, and submit jobs to the
database.  The _manager could verify a job was submitted.  If the _manager
failed, a restart could remove stalled jobs.  Perhaps a 'Basic Overseer
Service' like they have in AFS, could restart jobs _managers and _slaves
that die.  

Assuming every job submitted receives a unique ID that can be used as a
key between tables, using an SQL backend would also allow extensions of
the queing software without requiring modifications of all facets of
queue.  For example, a seperate table could contain infomation about jobs
submitted that could affect operation, without having to go through and
rewrite all data structures.  

The extra trouble of setting up a free database program might be worth the
flexability and design elegance that could be achieved.  There would be a
great benefit to reporting as well... just keep track of process
statistics and tie it back into the DB, along with start times, etc... 

Just a thought...

--------------------------------------------------------------------------
Patrick J. Hennessey                                  sp...@si...
Information Solutions                                       (512) 381-3722
SigmaTel, Inc.                                            www.sigmatel.com
Austin, TX					    	      we're hiring