[Mauiusers] Maui and failover strategy

Balle, Susanne susanne.balle at hp.com
Fri Jan 7 07:56:03 MST 2005


Hi,

I have been looking around for information on the Maui scheduler and its
failover strategy. The only thing I could find was:

> > > Single Point Of Failure
> > > ***********************
> > > Maui is designed to run on the node specified in maui.cfg,
"SERVERHOST".
> > > In case this node goes down, scheduling goes down, too. The only
way to
> > > bring up this service is to change parameter SERVERHOST and
restart
> > > Maui.
> >
> >   NCSA has funded development of full scheduler 'fail-over'
capability.
> > This feature allows a primary and back-up Maui to be run on separate
> > nodes.  Both daemons remain synchronized so that if the primary
fails,
> > the secondary immediately takes over with no loss of state.  All
scheduler
> > client commands are automatically rerouted and monitoring of the
primary
> > continues.  When the primary is restored, scheduling functionality
is
> > restored and scheduler clients again communicate with the primary.
This
> > feature is currently in alpha testing and will be generally
available in
> > the near future.

What is the status of this work?

Thanks for any info

Regards

Susanne

---------------------------------------------------------------
Susanne M. Balle, PhD
Hewlett-Packard
MS ZKO02-3/Q08
110 Spit Brook Road
Nashua, NH 03062

Phone: 603-884-7732
Fax:     603-884-0630

Susanne.Balle at hp.com



More information about the mauiusers mailing list