Status:
21.03.21, 11:50: We are done for now. Next scheduled downtime is April 16.
19.03.21, 20:25: Good progress today, both on the move from the old to the new server room and on storage integration in StorNext. Work continues tomorrow.
19.03.21, 06:10: Starting shutdown
Downtime:
During the downtime we will:
- Integrate new disk storage in StorNext
- Integrate a new tape robot
- Move large parts of the compute clusters to the new server room in Helga Eng's house.
We will start shutdown at 06.00 on Friday March 19.
The move to the new server room is a large job, and we don't know how many nodes we manage to physically move and integrate into the infrastructure in the new server room over the weekend.
Since we need to also move the Infiniband infrastructure, Infiniband-nodes that have not been moved will remain down until they have been moved.
Note that beehive24-phi, beehive25-phi, beehive32-gpu, beehive33-gpu, euclid17-gpu, euclid18-gpu, euclid19-gpu and euclid20-gpu will be retired. The hardware is not supported by later versions of Red Hat and the machines are from 2013 and 2014.
You will not be able to log on to the system from Friday morning and over the weekend.
Email will be available as normal
Early warning:
We plan a new downtime from Friday April 16. and the following weekend to upgrade StorNext.