You are here

Grex

Grex System Status

Post date System Status: Update Notes
2016-01-22 - 19:28 PST Offline

Lustre filesystem servers failure

Around 9PM CST, Lustre filesystem servers rebooted.  /global/scratch on Grex is unavailable.

2015-11-27 - 11:11 PST Online

Grex is online

The storage controller that failed last week was replaced. High-availability features and performance of Lustre were restored. The replacement was done online, without affecting running  jobs.

2015-11-27 - 11:11 PST Conditions

Grex is online

The storage controller that failed last week was replaced. High-availability features and performance of Lustre were restored. The replacement was done online, without affecting running  jobs.

2015-11-20 - 13:56 PST Conditions

Grex is back online but file system might be slow

The faulty storage controller was taken out and the new replacement should be here next week. Until then, only one controller is serving the Lustre file system and some performance degradation is expected.  

Some jobs may have been lost. Please check your submitted jobs and resubmit if needed.

2015-11-19 - 08:19 PST Offline

Grex is unavailable due to storage issue

The storage controller hosting the Lustre file system is showing some signs of hardware issue as well. The vendor has been notified and they are working on the issue. 

2015-11-19 - 07:46 PST Testing

Grex is unstable due to hardware issue

The primary server hosting the home filesystem has a hardware issue and we will be switching to the secondary server. This may cause some disruption. 

2015-11-18 - 21:10 PST Testing

Grex is not available

Grex home filesystem had a problem again and that was the cause of outage. Grex is not completely recovered and you may experience some slowness. 

Pages