ARSC system downtime for all systems (all)

Menu to filter items by type

Type Downtime News
Machine All Systems linuxws pacman bigdipper fish lsi
Downtime All Future Current Past

Contents for all systems

News Items

28 Aug 2015 Unscheduled Downtime

Last Updated: Fri, 28 Aug 2015 -
Machines: bigdipper
Start Time: 08/28/2015 -- 16:00
  End Time: 08/28/2015 -- 18:00
    Reason: The Bigdipper storage silo will be rebooted in an effort to clear slow
            system response times.  All pending batch_stage processes will be placed 
            on hold during this reboot.  Any file copies to and from $ARCHIVE and
            projects will need to be restarted following the reboot.

24 Aug 2015 Unscheduled Downtime

Last Updated: Mon, 24 Aug 2015 -
Machines: pacman
Start Time: 08/24/2015 -- 16:20
  End Time: 08/24/2015 -- 16:54
    Reason: Human error caused jobs on the 16 and 12 core nodes to be killed. 
            Users running jobs at the time were notified. 

24 Jul 2015 UnScheduled Downtime

Last Updated: Fri, 24 Jul 2015 -
Machines: linuxws pacman fish
Start Time: 07/24/2015 -- 16:00
  End Time: 07/24/2015 -- 22:00
    Reason: CENTER file system interruption

By 10:00 PM (Alaska Time) Friday, July 24, 2015 ARSC systems and
 services had recovered from an unscheduled interruption.

This evening our system analysts did a full reboot of everything that
 mounts ${CENTER} in order to restore the lustre filesystem to
 operation. It's not clear why this was necessary to recover, but it
 was. The ${CENTER} interruption started sometime late Friday
 afternoon.

The ${CENTER} file system is the high performance shared file system
 available on pacman, fish and the Linux workstations and is the
 preferred file system for computing on ARSC systems.

We apologize for any inconvenience this may have caused,  and please 
 contact us if you have any questions.

From our system administrators: Happy Golden Days!


Back to Top