Update - We have also noticed that this issue has occasionally affected compute nodes, so far on savio4_gpu and savio4_htc. If you have a problem accessing scratch on a login node or a compute node, please let us know (including what node you had the problem on) so we can reboot the node in question.
We are in the process of scheduling a downtime to fix the problem.
Oct 07, 2026 - 14:08 PDT
We are in the process of scheduling a downtime to fix the problem.
Oct 07, 2026 - 14:08 PDT
Update - We are continuing to monitor for any further issues.
Sep 16, 2026 - 08:02 PDT
Sep 16, 2026 - 08:02 PDT
Monitoring - For the past week or so, on a sporadic basis, we have noticed that a login node is not able to access one of the disks that provide the /global/scratch (Lustre) filesystem. When this happens one may see an error message, "Cannot send after transport endpoint shutdown", when running a command that tries to access data on the inaccessible disk.
This is caused by a Lustre bug. A fix is available and will be rolled out when we are able to take a short downtime in coming weeks.
We are monitoring the nodes and rebooting as needed as a short-term work-around, but if you have a problem accessing scratch, please let us know (including what node you had the problem on) so we can reboot the node in question.
Sep 16, 2026 - 07:57 PDT
This is caused by a Lustre bug. A fix is available and will be rolled out when we are able to take a short downtime in coming weeks.
We are monitoring the nodes and rebooting as needed as a short-term work-around, but if you have a problem accessing scratch, please let us know (including what node you had the problem on) so we can reboot the node in question.
Sep 16, 2026 - 07:57 PDT
About This Site
This is a status page for the Savio HPC cluster operated by Berkeley Research Computing at UC Berkeley.
Home Directory Storage
Operational
Scratch Storage
Operational
Login Nodes
Degraded Performance
Compute Nodes
Operational
Data Transfer Nodes (DTNs)
Operational
Job Scheduler
Operational
Network
Degraded Performance
Open OnDemand (OOD)
Operational
MyBRC User Portal
Operational
Operational
Degraded Performance
Partial Outage
Major Outage
Maintenance
Major outage
Partial outage
No downtime recorded on this day.
No data exists for this day.
had a major outage.
had a partial outage.
Scheduled Maintenance
Planned Maintenance: Savio Cluster, Monday, October 26 Oct 26, 2026 08:00-17:00 PDT
Update -
Maintenance date is finalized for 10/26.
Oct 16, 2026 - 15:04 PDT
Oct 16, 2026 - 15:04 PDT
Scheduled -
Savio is tentatively scheduled to be offline for one day on Monday, October 26, 2026. During the outage we will fix a Lustre issue that has been causing intermittent inactive OSTs on login and compute nodes, and we will also do general node maintenance and repairs.
Slurm reservation: A reservation is already in place for October 26. If a job’s requested time limit would run into the maintenance window, the job will stay in the queue until the cluster is back online. This mostly affects jobs with long time limits, and it applies more widely as October 26 gets closer. To keep jobs running before the outage, set time limits that let them finish before the reservation begins.
Oct 16, 2026 - 15:01 PDT
Slurm reservation: A reservation is already in place for October 26. If a job’s requested time limit would run into the maintenance window, the job will stay in the queue until the cluster is back online. This mostly affects jobs with long time limits, and it applies more widely as October 26 gets closer. To keep jobs running before the outage, set time limits that let them finish before the reservation begins.
Oct 16, 2026 - 15:01 PDT