We are actively investigating intermittent slowness on the shared /fs/scratch filesystem.

OSC systems staff will execute a rolling reboot across all active compute clusters (Pitzer, Ascend, and Cardinal) beginning Thursday, November 05.

Known issues

Unresolved known issues

Known issue with an Unresolved Resolution state is an active problem under investigation; a temporary workaround may be available.

Resolved known issues

A known issue with a Resolved (workaround) Resolution state is an ongoing problem; a permanent workaround is available which may include using different software or hardware.

A known issue with Resolved Resolution state has been corrected.

Known Issues

Title Category Resolutionsort descending Description Posted Updated
Problems with GPFS filesystem, and OnDemand is not working filesystem Resolved

Updates on 12:00pm June 10:

The issue is fixed. All impacted services return to production.

We apologize for the inconvience. If you have any questions, please...

Read more
5 years 5 months ago 5 years 5 months ago
Problems with home directory servers filesystem Resolved

We had several auto-reboots with our home directory servers, starting from around 11pm August 30. Jobs might be impacted. The systems are working properly now.  

... Read more
5 years 2 months ago 5 years 2 months ago
Poor performance with hybrid MPI+OpenMPI jobs and more than 4 MPI Tasks on multiple nodes Software Resolved
(workaround)

RELION versions prior to 5 may exhibit suboptimal performance in hybrid MPI+OpenMP jobs when the number of MPI tasks exceeds four across multiple nodes.

Workaround...

Read more
1 year 3 months ago 1 year 3 months ago
Dec 27, 2016: Issues with /fs/project filesystem Resolved

Dec 27, 2016 3:46PM Update: Both project and scratch file systems (/fs/project and /fs/scratch ) are back to normal now.  Some users' jobs may be...

Read more
9 years 10 months ago 9 years 10 months ago
Intermittent home directory performance issues filesystem Resolved

Users may experience performance issues in home directory. It is recommended to use temporary directory ($TMPDIR, or scratch) or project storage to minimize the impact on...

Read more
3 years 4 months ago 3 years 3 months ago
OnDemand experiencing difficulties Web Services Resolved

OnDemand is experiencing some difficulties that may be related to the changes from the downtime. We are aware of these problems and are working on resolving them. We appreciate your patience.

12 years 9 months ago 12 years 9 months ago
Apptainer container builds fail on compute nodes due to /tmp namespace behavior Software Resolved
(workaround)

Building Apptainer containers on compute nodes may fail during apt update or other package operations...

Read more
10 months 5 days ago 9 months 1 day ago
Rolling reboot of all clusters, starting from 8 AM Tuesday, June 19, 2018 Batch, Owens, Ruby Resolved

Posted on June 12, 2018, at 4:40 PM:

We will have rolling reboots of three clusters (Owens, Ruby, and Oakley) including login and compute nodes, starting from 8 AM Tuesday...

Read more
8 years 5 months ago 8 years 4 months ago
OnDemand unresponsive login Resolved

Some of the login nodes on Owens and Pitzer are in bad states. User can't log into OnDemand. And scratch is unresponsive sometimes. We are working on this issue. We will update when we have more...

Read more
6 years 5 months ago 6 years 5 months ago
- --gpus-per-task is not working Batch Resolved

Updated: This is fixed. 

Original Post:

After the recent Slurm upgrade, the option --gpus-per-task is currently not functioning as...

Read more
1 year 10 months ago 1 year 10 months ago

Pages