You are here
|Post date||System Status:||Update Notes|
|2014-01-10 - 15:53 PST||Online||
Orcinus Upgrade Update
Orcinus is operational. We are working on the final components of the new software stack. Please submit and run your jobs and contact WestGrid's support if you have any issues.
Our sincere apologies for all delays.
|2013-12-27 - 16:21 PST||Conditions||
Orcinus Upgrade Update - Friday, Dec 27, 2013
We have enabled access to the data and we are begining to schedule selected compute jobs for testing.
We are working on the new software stack and testing the installation. Our sincere apologies for all delays.
|2013-12-17 - 16:02 PST||Offline||
Scheduled Major OS and Filesystem Upgrades
Orcinus is down for a major OS and File System upgrade, and for cooling system maintenance.
Dec.24, 2013: STILL DELAYED: due to Lustre filesystem issues data access was only enabled late on the 21-st, and we're still working on the software stack and testing. The login nodes are available for data access, but the compute nodes are still unavailable so jobs cannot be submitted. We are now aiming for Dec.27.
(1) Dec.13-17: construct new filesystem and copy data from old system.
(2) Dec.18: Test new filesystem and remove the old one. Reconfigure filesystem management nodes for full redundancy.
(3) Dec.19: Upon successful testing filesystems will be made available to users through the login nodes.
(4) Dec.20: Cooling system tested under load and final updates installed.
(5) Dec.23: System in production and available for computation.
|2013-11-12 - 18:13 PST||Downtime Scheduled||
Orcinus Major OS and File System Upgrades
Orcinus will be down in order to complete a major OS and File System upgrade, starting on
Thursday, December 5, 2013 at 10:00 AM (PST)
We will attempt to resume normal operations ASAP, but the outage is expected to last at least 10 days.
We will post more details as they become available.
|2013-10-29 - 15:26 PDT||Online||
System fully operational
System fully operational.
|2013-10-24 - 15:06 PDT||Online||
Chiller issues resolved Orcinus operational
The chiller issues have been resolved and normal operations have resumed. Please examine your intermediate data and resubmit the lost computations. We are sorry for any inconveniences.
|2013-10-24 - 08:16 PDT||Offline||
We have suffered a chiller failure and were forced to power off all compute nodes. Running jobs were lost. We have no estimate yet of repair time. We will restore operations as soon as is possible.