Primary informations
Username: coppinp
Cluster: baobab
Description
Hi, very few nodes seem to be up on Baobab. Most are in the draining state. E.g.
sinfo
shared-cpu up 12:00:00 120 drain cpu[084-090,193-195,197-202,205-213,216-217,220-225,237-242,244,247-276,279-280,282-284,289-299,304,319-340,342-352]
scontrol show node cpu280
Reason=health_BEEGFS__tcp_con_storage [root@2026-03-14T21:45:25]
At some point during the weekend /srv/beegfs/dpnc got filled. Could this be the reason?
(space has already been liberated in the meantime)
All the best,
Paul
