Dear users of Nikhef (ELPROD),
A major power outage in the Amsterdam (NL) area affected the compute services of Nikhef ELPROD. The worker nodes (but not storage or grid central services) are currently powered off and all running jobs were lost. It also affects the Dzero-specific LCG-RB, and the classic storage element (tbn15.nikhef.nl).
The site currently publishes (correctly but somewhat uselessly) that it can run zero jobs, has zero free CPUs, and an almost infinite response time.
Also the LHC-OPN link through SARA seems to be affected. All storage traffic is now routed through our public internet links, limiting bandwidth to at most 2 Gbps.
We are in contact with the utility services to restore the site as soon as possible to full functionality. We are certain that the utility services will solve this issue as soon as possible.
Again: no other services at ELPROD should be affected, ONLY the worker node compute servic e (gazon and trekker.nikhef.nl).
Thank you for your consideration.
David Groep.