Group Responsible: Services & Tools

Affected Area: SSH Access

Maintenance Type: transparent

Starting today, Monday, October 12, 2026, the SCDF SSH key management page will accept only new Ed25519 keys, and each account can now hold up to 5 keys. On April 5, 2027, all non-Ed25519 keys will be removed. Action must be taken before this date. WHAT YOU NEED TO DO Navigate to https://useraccount.sdcc.bnl.gov/ssh-key and log in with your SCDF account. Upload an Ed25519 key to your account before April 5, 2027. On that date all non-Ed25519 keys (RSA, DSA, ECDSA) will be removed. If your account has only RSA keys on that date, you will lose SSH access until you add an Ed25519 key. HOW 1. Generate a key: ssh-keygen -t ed25519 -C "your-laptop" 2. Log in at https://useraccount.sdcc.bnl.gov and open the SSH key page. 3. Paste the contents of ~/.ssh/id_ed25519.pub and click Add. Keys currently marked "Deprecated" on that page are the ones that will be removed. For more information regarding SCDF SSH key policies and instructions please refer to the documentation here: https://docs.sdcc.bnl.gov/access/ssh/generate-ssh-key-pairs/ WHY Ed25519 is the current industry standard for SSH keys. It is faster, uses shorter keys, and avoids the configuration mistakes that weaken RSA and DSA keys. OpenSSH has already disabled DSA and SHA-1 RSA signatures by default, and this change keeps SCDF in step with that direction and with federal guidance to retire legacy algorithms. ALSO NEW: LINK YOUR ORCID You can now connect your ORCID to your SCDF account. Log in at https://useraccount.sdcc.bnl.gov, choose "Link ORCID", and sign in to ORCID. This is a self service for users utilizing services such as Jupyterhub. Questions: RT-RACF-Www@bnl.gov

Group Responsible: Services & Tools

Affected Area: SSH Access

Maintenance Type: transparent

Starting Monday, October 12, 2026, the SCDF SSH key management page will accept only Ed25519 keys, and each account can now hold up to 5 keys. WHAT YOU NEED TO DO After the change is live on October 12, 2026 and BEFORE APRIL 5, 2027, upload an Ed25519 key to your account. On that date all non-Ed25519 keys (RSA, DSA, ECDSA) will be removed. If your account has only RSA keys, you will lose SSH access until you add an Ed25519 key. WHY Ed25519 is the current industry standard for SSH keys. It is faster, uses shorter keys, and avoids the configuration mistakes that weaken RSA and DSA keys. OpenSSH has already disabled DSA and SHA-1 RSA signatures by default, and this change keeps SCDF in step with that direction and with federal guidance to retire legacy algorithms. A second announcement with further instructions will follow the SSH key management page update on October 12, 2026. Questions: RT-RACF-Www@bnl.gov

Group Responsible: Storage & Infrastructure

Affected Area: Phenix GPFS02 file system

Maintenance Type: service interruption

Phenix GPFS02 File System (also affects SPHENIX and Astro) is back up after power disruption last week. All volumes are still currently rebuilding on the NetApp appliance. although GPFS02 file system is up and mounted on the cluster servers there may be slow responses until the volumes rebuilds are completed.

Group Responsible: Storage & Infrastructure

Affected Area: Phenix GPFS02 file system

Maintenance Type: service interruption

Phenix GPFS02 File System (also affects SPHENIX and Astro) is still down after power disruption last week. All volumes are still currently rebuilding on the NetApp appliance. As noted, we attempted to mount the GPFS02 file system on the cluster servers and it was unsuccessful. The current fsck(file system check) began yesterday and is still running on the GPFS02 file system (~2.5PB) and is approx. 40% complete and this needs to finish before we can attempt to mount the file system again. We will send additional updates when available.

Group Responsible: Storage & Infrastructure

Affected Area: Phenix GPFS02

Maintenance Type: transparent

The Phenix GPFS02 File System (also affects SPHENIX and Astro) is now in an optimal state after power disruption as all volumes have completed rebuilding on the NetApp appliance.

Group Responsible: Storage & Infrastructure

Affected Area: Phenix GPFS02

Maintenance Type: service interruption

Phenix GPFS02 File System (also affects SPHENIX and Astro) is still down after power disruption last week. We received parts today which a NetApp engineer replaced. While we made some progress it was determined additional parts need to be replaced. These are being shipped overnight so they can be replaced tomorrow. We will provide an update once this has taken place.

Group Responsible: Storage & Infrastructure

Affected Area: PHENIX GPFS02 File System

Maintenance Type: service interruption

Phenix GPFS02 File System (also affects SPHENIX and Astro) is still down after power disruption yesterday. It was determined to be a hardware failure on the NetApp Storage Controller. The parts were ordered; delivery was expected this morning (9/25). Unfortunately, all parts needed to repair the NetApp Storage Controller have not arrived today to repair the GPFS02 file system. Will need to wait until Monday at the earliest for part arrival update from the vendor. The file system will be down until the repair is completed.

Group Responsible: Storage & Infrastructure

Affected Area: PHENIX GPFS02 File System

Maintenance Type: service interruption

Phenix GPFS02 File System (also affects SPHENIX and Astro) is still down after power disruption this morning. After working on the issue with the vendor, it was determined to be a hardware failure on the NetApp. The part has been ordered with delivery expected tomorrow morning (9/25). Will update when we have part on site and has been replaced. Still no ETA to when this will be resolved and returned to service.

Group Responsible: Storage & Infrastructure

Affected Area: Phenix GPFS02, this also affects sPhenix and Astro.

Maintenance Type: service interruption

Phenix GPFS02 File System (also affects sPHENIX and Astro) is still down after power disruption this morning. After working on the issue with the vendor, it was determined to be a hardware problem on the NetApp. The part has been ordered with delivery expected tomorrow morning (9/25). Will update when we have part on site and replaced. Still no ETA to when this will be resolved and returned to service.

Group Responsible: Data Center Operations

Affected Area: b725 datacenter

Maintenance Type: service interruption

SCDF is responding to unexpected power and cooling issues, correlated with F&O generator testing, within the b725 datacenter and some resulting service outages.

Group Responsible: Fabrics

Affected Area: Jupyter HTCondor pool

Maintenance Type: service interruption

There will be a brief interruption in Jupyter access to the HTCondor pool on 24 September starting at 10:00 Eastern as parts of the infrastructure will be undergoing maintenance and upgrades

Group Responsible: Data Center Operations

Affected Area: US ATLAS dCache storage service

Maintenance Type: transparent

Dear all, Please note that the BNL US ATLAS dCache storage element maintenance has been successfully completed. The service is now fully operational. Thank you for your understanding. Maintenance Description: Please note that the BNL US ATLAS dCache storage element underwent maintenance for a major software upgrade on September 14 from 9:00 AM to 5:00 PM EDT. The storage was unavailable to users during this period.

Group Responsible: IT Services

Affected Area: US ATLAS dCache storage service

Maintenance Type: scheduled downtime

Dear all, Please note that the BNL US ATLAS dCache storage element will undergo maintenance for a major software upgrade on September 14 from 9:00 AM to 5:00 PM EDT. The storage will be unavailable to users during this period. Thank you in advance for your understanding.

Group Responsible: IT Services

Affected Area: US ATLAS dCache storage service

Maintenance Type: scheduled downtime

Dear all, Please note that the BNL US ATLAS dCache storage element will undergo maintenance for a major software upgrade on September 14 from 9:00 AM to 5:00 PM EDT. The storage will be unavailable to users during this period. Thank you in advance for your understanding.

Group Responsible: IT Services

Affected Area: US ATLAS dCache storage service

Maintenance Type: scheduled downtime

Dear all, Please note that the BNL US ATLAS dCache storage element will undergo maintenance for a major software upgrade on September 14 from 9:00 AM to 5:00 PM EDT. The storage will be unavailable to users during this period. Thank you in advance for your understanding.