Centre de Calcul IN2P3/CNRS
21 Avenue Pierre de Coubertin
69627 VILLEURBANNE cedex
Tél. : +33 (0) 4 78 93 08 80
Fax : +33 (0) 4 72 69 41 70
22 January 2010
We are going to process to some maintenance operation on HPSS and BQS on the 26/01. These two services will be unavailable on the 26/01 from 6am to 1pm (8am to 1pm for HPSS).
- This downtime is just on the Anastasie cluster. Pistoo will still be available during this operation.
- Jobs will be enqueued progressively (depending on the CPU needed by the job), 48 hours before the the intervention. Every job will be enqueued on the 26/01 from 6am to 1pm.
- This maintenance operation concerns the physical BQS database servers and system updates on every worker.
- Workers reboot will be done progressively, so they will be all fully available on the day after the operation.
- Xrootd, SRB, iRod, dCache services will still be available but data stored on tape won’t be available. Only cache data (stored on disk) will be readable.
- This maintenance operation will be a system update of some HPSS servers.
We will use this downtime to update the kernel Linux, as well as the AFS and GPFS clients. Oracle services will also be updated, but this won’t impact the production of it.
We will let you know about the evolution of these two operations.
Thanks for your understanding,