Two Day System Downtime

Date: 
Tuesday, February 21, 2017 - 7:00am to Wednesday, February 22, 2017 - 5:00pm
Location: 

Ohio Supercomputer Center

2/23/17 12:00PM Update:

Partial services will be available at approximately 1PM today including:

Oakley cluster including all login and compute nodes for general access
Ruby cluster including all login and compute nodes for restricted access
Owens cluster including login and half of compute nodes for general access
Home directory and scratch filesystems
OnDemand and other web portals

Effected services will NOT be available including:

Project filesystem (/fs/project)
Half of compute nodes on Owens. Users will experience slowness of scheduling jobs on Owens

 
5PM 02/22/2017 Update: We have to extend the downtime to complete filesystem checks. We do not know when all services will be back. We'll keep you posted once we know more.
 
A two day downtime is upcoming for all HPC systems from 7 a.m. Tuesday, Feb. 21 to 5 p.m. Wednesday, Feb. 22, 2017. The downtime will affect the Oakley, Ruby and Owens Clusters, web portals, and HPC file servers. Login services, including my.osc.edu, and access to storage will not be available during this time. 
 
In preparation for the downtime, the batch scheduler will begin holding jobs that cannot be completed before 7 a.m., Feb. 21. Jobs that are not started will be held until after the downtime and then started once the system is returned to production status. 

Highlights of the downtime activities include a software refresh for all systems:
  • Default compiler and MPI versions on Oakley and Ruby will be updated to match those on Owens.
  • Libraries will be rebuilt with the new default compilers.  Some applications will be rebuilt.
  • Many applications will be updated to match the default version on Owens.

We will keep you updated on these software changes and how they can effect your work here

 

To stay up to date on system notices, follow @HPCNotices on Twitter. As always, you can contact us at OSC Help.