S'abonner au fil des news (flux RSS)

Fil des news

20240425 / Energy crisis

As a test, starting tonight, following partitions will no longer start jobs between 6h and 22h (PartitionDown).

Jobs will only be started at night. Running jobs will continue running.

In test : E5 E5-GPU Lake-bigmem Epyc Cascade

2024/04/25 13:41 · ltaulell

20240422 / Energy crisis

Due to “high pressure from above & outside”, we are going to reduce PSMN's energy footprint by any means possible.

Changes are applied starting today. More info will come later.

2024/04/22 07:04 · ltaulell

20240410 / scratch/Cral on Cascade

We are changing RAM modules on one of /scratch/Cral servers. Files and directories may act weirdly for a few moments.

EDIT 11:45: done. Back to norminal.

2024/04/10 08:11 · ltaulell

20240408 / scratch/Cral on Cascade

One of scratch/Cral servers is having hardware issues, getting it to protected mode. Hence 1/4 of files disappearing momentarely. We know, we're on it. It might take a few days to resolve.

2024/04/08 11:52 · ltaulell

20240402 / PDU on Cascade

a PDU (power distribution unit) decided to die.

It has been replaced, impacted nodes restarted (s92node[81-84,87-90,109-112]).

Jobs have been lost.

2024/04/02 09:34 · ltaulell
news/blog.txt · Dernière modification : 2020/08/25 15:58 de 127.0.0.1