Status

From Grid5000
Revision as of 11:46, 22 November 2005 by Pmarty (talk | contribs) (→‎Grid administration schedule: syntax...)
Jump to navigation Jump to search

Grid administration schedule

If you experience problems using Grid'5000, please check the grid administration schedule, where past, present and future incidents (planned or not...) are notified for all sites.

Note that as far as possible most of the administration operations are planned on thursday mornings.

Other monitoring tools are also available to help you trace your experiment. If you still have questions, please report your problem or contact the technical staff from the most relevant site (or the whole technical committee for global cases).

Ganglia

Ganglia provides information about resources usage (memory, cpu, jobs...) for individual sites or the whole grid.

https://helpdesk.grid5000.fr/ganglia/

OAR and OARgrid

OAR is the grid scheduler, which may also be queried for current and planned jobs and nodes reservations, either from command lines (see OAR documentation) or as web services.

Viewing current reservations/jobs

It is possible to view a global snapshot or select an individual site:

Viewing usage history and schedule

It is possible to view a global Gantt Diagram ou select on an individual site:

GridPrems

GridPrems is an alternate grid scheduler running on certain sites.

https://helpdesk.grid5000.fr/gridprems/

Nagios

Nagios allows to monitor critical grid servers and services and automatically reports incidents and failures.

https://helpdesk.grid5000.fr/nagios/

Experimental network monitoring tools

Beware: experimental = unstable...

https://helpdesk.grid5000.fr/map/map.html (please initialize network data first using https://helpdesk.grid5000.fr/map/datafeed.php)